March 15, 2014 / Vol. 39, No. 6 / OPTICS LETTERS
1649
Terahertz wave parametric amplifier Saroj R. Tripathi,1,* Yuusuke Taira,1 Shin’ichiro Hayashi,2 Kouji Nawata,2 Kousuke Murate,1 Hiroaki Minamide,2 and Kodo Kawase1,2 1 2
Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan
RIKEN, 519-1399 Aramaki-Aza Aoba, Aoba-ku, Sendai 980-0845 Japan *Corresponding author: [email protected]‑u.ac.jp Received December 26, 2013; accepted February 4, 2014; posted February 14, 2014 (Doc. ID 203686); published March 13, 2014
The importance of terahertz (THz) wave techniques has been demonstrated in various fields, and the range of applications is now expanding rapidly. However, the practical implementation of THz science to solve the real-world problems is restricted due to the lack not only of convenient high power THz wave emitters and sensitive detectors but also of efficient quasi-optical active devices such as amplifiers. In this work, we demonstrate the direct amplification of THz waves in room temperature using magnesium oxide-doped lithium niobate (MgO:LiNbO3 ) crystals as the nonlinear gain medium. The input THz wave is injected as a seed beam along with the pump beam into the nonlinear crystal and it is amplified by the optical parametric process. We report gain in excess of 30 dB with an input THz pulse energy of less than 1 pJ. We believe that this demonstration will contribute to the convenience and further applicability of THz frequency techniques. © 2014 Optical Society of America OCIS codes: (040.2235) Far infrared or terahertz; (190.4410) Nonlinear optics, parametric processes; (190.4400) Nonlinear optics, materials. http://dx.doi.org/10.1364/OL.39.001649
The rapid advancement of optoelectronics and laser technology has enabled the convenient generation of terahertz (THz) waves, bridging the frontiers between the microwave and infrared regions of the electromagnetic spectrum. Applications for THz radiation have already been demonstrated in fields such as biomedicine [1], material science [2], nondestructive testing [3,4], and many more that have considerable potential. However, the impact of a majority of these applications is thus far restricted to the controlled environment of research laboratories, and many rely on sophisticated and expensive equipment. In order to further drive the evolution of THz techniques so that they can be applied to real-world problems, high performance, compact, convenient, and low cost THz sources and detectors must be made available. A new generation of higher efficiency quasi-optical THz devices such as modulators and amplifiers will play a pivotal role in the transformation of THz science from particular applications to the commonplace. In recent years, the development of high-power THz wave sources has received a great deal of attention, and various methods to achieve high-power THz waves have been reported. Examples include the THz wave parametric generation [5], optical rectification of femtosecond pulses in organic and inorganic crystal [6–8], quantum cascade laser (QCL) [9,10], laser plasma interaction [11], and so on. Besides this, amplification of THz radiation has also been demonstrated using QCL in order to fulfill the necessity of high power for practical applications [12,13]. However, these methods generally have their limitations. For example some sources require expensive laser pump sources, sophisticated optics, and so on. Similarly, THz amplification in a QCL requires cooling to cryogenic temperatures, thereby making the system cumbersome and complicated to operate. Therefore, a direct amplification method for THz radiation at room temperature is an attractive way to overcome the sophistication and expense of current technologies. In this Letter, we report experiments that demonstrate a method to amplify single longitudinal mode, pulsed THz 0146-9592/14/061649-04$15.00/0
waves emitted from an injection-seeded THz wave parametric generator (is-TPG). The emitted THz wave is amplified in a nonlinear crystal by means of an optical parametric process. 30 dB gain was achieved for an input THz pulse with energy of a few picojoule. THz wave parametric generation and amplification is based on both second and third order nonlinear processes [14–17]. Among the polar nonlinear crystals, such as LiTaO3 and GaP, we selected LiNbO3 for a number of reasons. First it has a large nonlinear coefficient (d33 25.2 pm∕V at λ 1.064 μm [18], and estimated d33 165 pm∕V at THz wave region [19].) It also has a high figure of merit [7] and high transparency over a wide wavelength range (0.4–5.5 μm). Moreover, the laser induced damage threshold of this crystal is high, making it possible to use higher peak power pump sources without damage to the crystal. LiNbO3 has four infrared- and Raman-active transverse optical (TO) phonon modes, called A1 -symmetry modes. The lowest of these modes (ω0 ∼ 248 cm−1 ) is particularly useful for efficient tunable THz generation and amplification because it has the largest parametric gain and the smallest absorption coefficient [20]. When an intense laser beam propagates through a nonlinear crystal, the photon and phonon transverse fields are coupled and behave as new mixed photon–phonon states, called polaritons. The efficient parametric scattering of light via a polariton (called stimulated polariton scattering) results in the generation of an idler beam whereby THz radiation can be generated. The output of such a system produces a wide range of THz frequencies. However, as shown in Fig. 1(a), when a single frequency seed beam is injected at a specific fixed angle of incidence to the crystal, such that phase matching conditions are met, coherent THz radiation is generated. This process, known as injection seeding, offers a wide frequency range of THz radiation that is tuned simply by varying the wavelength of the seed beam. Based on this fundamental principle, we use the THz wave thus generated (which we call the weak THz field) as a seed beam, © 2014 Optical Society of America
1650
OPTICS LETTERS / Vol. 39, No. 6 / March 15, 2014
Fig. 1. (a) Principle of THz wave generation and (b) THz wave amplification. Here, kseed , kpump , kTHz , and kidler represent the wave vectors of the seed, pump, THz, and idler beams respectively.
as shown in Fig. 1(b). By seeding the THz wave with a specific wavelength, the corresponding idler beam can be amplified, and this in turn leads to amplification of the THz radiation. In this process the noncollinear phase matching condition must be satisfied: i.e., kpump kTHz kidler ; where kpump , kTHz , kidler are the wave vectors for pump, THz wave, and idler beam, respectively. This condition is accomplished by adjusting the angle of incidence of the THz seed beam to the LiNbO3 crystal. The experimental setup is shown in Fig. 2. The system consists of two distinct parts. First, a THz wave emission
Fig. 2. Experimental setup for the emission and amplification of THz radiation. The pump laser (λ 1064 nm, energy 700 μJ∕pulse, pulse width 420 ps, repetition rate 100 Hz, M 2 1.09) is amplified by two optical amplifiers in a double pass configuration. Each amplifier consists of 0.7-at. % Nd3 doped YAG as a gain medium with diameter and length 3 mm and 70 mm respectively. The inset shows the phase matching condition for generation and amplification of THz radiation with a frequency of 2.01 THz. In this figure, polarizing beam splitter and continuous wave are abbreviated as PBS and CW, respectively.
section that comprises a pump laser, laser amplifier, seed laser, and nonlinear crystal. The second part is the THz amplifier section, composed of a pair of nonlinear crystals and a pyroelectric detector to measure the emitted THz power. As a pump source, we used a diode end pumped single-mode microchip Nd3 :YAG laser that is passively Q-switched by a Cr4 :YAG saturable absorber. This laser has very low noise and its peak power fluctuations are lower (< 2%) than other active Q-switched lasers [21]. The pumping beam is amplified by two optical amplifiers and extracted using a polarizing beam splitter (PBS). This setup amplifies the laser power to 14 mJ∕pulse. A continuous wave tunable external cavity diode laser (ECDL, Velocity 6300, New Focus Inc.) with an average power of 5.5 mW is used as an injection seeder for the idler beam. Both pump beam and seed beam irradiate the 5 mol. % MgO:LiNbO3 crystal at a specific angle as shown in Fig. 1(a). We used a 50 mm long crystal with an antireflection coating for a wavelength of 1064 nm. The polarization orientation of the pump, seed, idler, and THz waves are all parallel to the z axis of the crystal. Since the maximum gain coefficient of LiNbO3 occurs at approximately 2 THz [22], we optimized our system for the frequency of 2.01 THz by tuning the seed beam wavelength to 1072.36 nm. In this case, the optimum angle between pump and seed beam is calculated to be 1.96 deg. The emitted THz radiation has pulse energy of 12.1 nJ, a temporal width of 100 ps, and a line width of 5 GHz. For amplification of this THz radiation, we used two nonlinear MgO:LiNbO3 crystals in series. These crystals are pumped by the Nd:YAG laser beam after it is transmitted through the emitter crystal. This recycling of the pump laser makes the system physically far more compact. The energy of the pump beam in this second stage is about 7.2 mJ∕pulse. The THz wave is collimated using a cylindrical lens and efficiently coupled to the nonlinear crystal using a silicon (n 3.4) prism. The THz wave is incident on the crystal so as to satisfy the nonlinear phase matching condition. In this case, the relationship kpump > kidler ≫ kTHz holds. The angle between the pump and idler beam is therefore small whereas the angle between the pump beam and the THz wave is large. This geometry of beams makes it relatively easy to arrange for the weak field THz radiation to be brought to the non-linear crystal as a seed beam. In order to minimize absorption loss of the THz wave and to ensure efficient coupling between the pump beam and the THz wave inside the lithium niobate crystal, we arranged for the pump laser to be as close as possible to the Silicon prism. Moreover, the pump laser and THz wave are temporally overlaid by adjusting the optical path length. In this arrangement, the weak THz field acts as a seed beam and due to the interaction between the laser and THz seed beam in the crystal, amplified THz radiation is emitted. The amplified THz wave within the nonlinear crystal will suffer total internal reflection due to the high refractive index of the MgO:LiNbO3 unless we use an appropriate output coupling medium. We use a Si-prism coupler on the y surface of the crystal to extract the THz wave from the crystal [23]. This pulsed THz radiation is finally detected by a calibrated pyroelectric detector.
March 15, 2014 / Vol. 39, No. 6 / OPTICS LETTERS Output THz energy (nJ/pulse)
2
1.5
1
0.5 (a) 0
1
0.1 0.01 0.001 Input THz energy(nJ/pulse)
0.0001
40 (b)
Gain (dB)
30
20
10
0
1
0.1 0.01 0.001 Input THz energy (nJ/pulse)
0.0001
Fig. 3. (a) Input/output characteristics of the THz amplification system. Here, the THz energy is varied using the combination of set of calibrated THz wave attenuators (TFA-4; CDP Corp) with transmittances of 1%, 3%, 10% and 30%. (b) Gain plotted as a function of input THz energy. The gain is calculated as G 10 log 10E out ∕Ein , where E out and E in are the THz input and output energy respectively.
Figure 3(a) shows how the output pulse energy of the THz amplification system varies as a function of the input pulse energy. Both input and output THz energy are measured using the pyroelectric detector with the input energy ranging from 1 nJ∕pulse to 0.1pJ∕pulse. The input THz energy is varied using the THz wave attenuator. In order to better visualize the results, we plot the gain as a function of input pulse energy in Fig. 3(b). We find a system gain in excess of 30 dB with low values of input energy per pulse (0.1pJ∕pulse). The gain varies inversely with input THz pulse energy. Gain falls almost to unity when the input THz pulse energy is higher than 1 nJ. We attribute this to saturation of the parametric gain of the nonlinear crystal. It is satisfying to note that our system generates THz pulse energies at levels that are easily detected by commercially available room temperature THz detectors. In our system, the gain saturates when the input pulse energy is more than 1 nJ∕pulse. However, commercially available THz detectors are capable of detecting 1 nJ pulses with good signal-to-noise ratio, so amplification beyond this level is not crucial. However, when the THz energy is less than 1 nJ∕pulse, our amplifier demonstrates an unprecedented ability to amplify such weak signal. We have demonstrated direct amplification of THz radiation with gain of more than 30 dB. Pronounced gain is observed when the input THz pulse energy is as small as a few picojoule. However, in the higher input pulse energy regime, gain decreases due to saturation of the
1651
parametric gain of the nonlinear crystal. Significant amplification of THz radiation as demonstrated here represents an advance in the general applicability of THz techniques. This amplification method of THz radiation can be implemented to compensate for losses in strongly absorbing samples in spectroscopic and imaging systems. Furthermore, amplification of THz radiation above the detection threshold of room temperature detectors greatly improves the applicability of, and enhances the versatility of THz sources. At the same time, costs and sophistication are reduced by eliminating the need for high power THz sources. Our initial demonstration may be extended to amplify the weak THz signals emitted by other types of source. The technique could be applied, for example, to difference frequency THz generation in nonlinear crystals and unitraveling carrier photo-diodes. We would like to thank Professor T. Taira of the Institute of Molecular Science, Okazaki, Japan, for his guidance on the subject of high power microchip lasers. We also like to thank K. Wood of QMC Instruments Ltd., for proofreading our manuscript. We also gratefully acknowledge financial support from the JST project, Japan. References 1. E. Pickwell and V. P. Wallace, J. Phys. D 39, R301 (2006). 2. H. Hirori and K. Tanaka, IEEE J. Sel. Top. Quantum Electron. 19, 8401110 (2013). 3. K. Kawase, Y. Ogawa, Y. Watanabe, and H. Inoue, Opt. Express 11, 2549 (2003). 4. S. R. Tripathi, H. Ogura, H. Kawagoe, H. Inoue, T. Hasegawa, K. Takeya, and K. Kawase, Corros. Sci. 62, 5 (2012). 5. S. Hayashi, K. Nawata, H. Sakai, T. Taira, H. Minamide, and K. Kawase, Opt. Express 20, 2881 (2012). 6. S. R. Tripathi, K. Murate, H. Uchida, K. Takeya, and K. Kawase, Appl. Phys. Express 6, 072703 (2013). 7. J. Hebling, K.-L. Yeh, M. C. Hoffmann, B. Bartal, and K. A. Nelson, J. Opt. Soc. Am. B 25, B6 (2008). 8. H. Hirori, F. Blanchard, and K. Tanaka, Appl. Phys. Lett. 98, 091106 (2011). 9. B. S. Williams, Nat. Photonics 1, 517 (2007). 10. K. Vijayraghavan, Y. Jiang, M. Jang, A. Jiang, K. Choutagunta, A. Vizbaras, F. Demmerle, G. Boehm, M. C. Amann, and M. A. Belkin, Nat. Commun. 4, 1 (2013). 11. H. Hamster, A. Sullivan, S. Gordon, W. White, and R. W. Falcone, Phys. Rev. Lett. 71, 2725 (1993). 12. C. Mauro, R. P. Green, A. Tredicucci, F. Beltram, H. E. Beere, and D. A. Ritchie, J. Appl. Phys. 102, 063101 (2007). 13. N. Jukam, S. S. Dhillon, D. Oustinov, J. Madeo, C. Manquest, S. Barbieri, C. Sirtori, S. P. Khanna, E. H. Linfield, A. G. Davies, and J. Tignon, Nat. Photonics 3, 715 (2009). 14. K. Kawase, J. Shikata, K. Imai, and H. Ito, Appl. Phys. Lett. 78, 2819 (2001). 15. K. Kawase, H. Minamide, K. Imai, J. Shikata, and H. Ito, Appl. Phys. Lett. 80, 195 (2002). 16. A. Lee, Y. He, and H. Pask, IEEE J. Sel. Top. Quantum Electron. 49, 357 (2013). 17. Y. Takida, T. Ohira, Y. Tadokoro, H. Kumagai, and S. Nashima, IEEE J. Sel. Top. Quantum Electron. 19, 8500307 (2013).
1652
OPTICS LETTERS / Vol. 39, No. 6 / March 15, 2014
18. I. Shoji, T. Kondo, A. Kitamoto, M. Shirane, and R. Ito, J. Opt. Soc. Am. B 14, 2268 (1997). 19. Y. Avetisyan, Y. Sasaki, and H. Ito, Appl. Phys. B 73, 511 (2001). 20. J. Shikata, K. Kawase, K. Karino, T. Taniuchi, and H. Ito, IEEE Trans. Microwave Theory Tech. 48, 653 (2000).
21. H. Sakai, H. Kan, and T. Taira, Opt. Express 16, 19891 (2008). 22. S. Hayashi, T. Shibuya, H. Sakai, T. Taira, C. Otani, Y. Ogawa, and K. Kawase, Appl. Opt. 48, 2899 (2009). 23. K. Kawase, J. Shikata, H. Minamide, K. Imai, and H. Ito, Appl. Opt. 40, 1423 (2001).
1649
Terahertz wave parametric amplifier Saroj R. Tripathi,1,* Yuusuke Taira,1 Shin’ichiro Hayashi,2 Kouji Nawata,2 Kousuke Murate,1 Hiroaki Minamide,2 and Kodo Kawase1,2 1 2
Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan
RIKEN, 519-1399 Aramaki-Aza Aoba, Aoba-ku, Sendai 980-0845 Japan *Corresponding author: [email protected]‑u.ac.jp Received December 26, 2013; accepted February 4, 2014; posted February 14, 2014 (Doc. ID 203686); published March 13, 2014
The importance of terahertz (THz) wave techniques has been demonstrated in various fields, and the range of applications is now expanding rapidly. However, the practical implementation of THz science to solve the real-world problems is restricted due to the lack not only of convenient high power THz wave emitters and sensitive detectors but also of efficient quasi-optical active devices such as amplifiers. In this work, we demonstrate the direct amplification of THz waves in room temperature using magnesium oxide-doped lithium niobate (MgO:LiNbO3 ) crystals as the nonlinear gain medium. The input THz wave is injected as a seed beam along with the pump beam into the nonlinear crystal and it is amplified by the optical parametric process. We report gain in excess of 30 dB with an input THz pulse energy of less than 1 pJ. We believe that this demonstration will contribute to the convenience and further applicability of THz frequency techniques. © 2014 Optical Society of America OCIS codes: (040.2235) Far infrared or terahertz; (190.4410) Nonlinear optics, parametric processes; (190.4400) Nonlinear optics, materials. http://dx.doi.org/10.1364/OL.39.001649
The rapid advancement of optoelectronics and laser technology has enabled the convenient generation of terahertz (THz) waves, bridging the frontiers between the microwave and infrared regions of the electromagnetic spectrum. Applications for THz radiation have already been demonstrated in fields such as biomedicine [1], material science [2], nondestructive testing [3,4], and many more that have considerable potential. However, the impact of a majority of these applications is thus far restricted to the controlled environment of research laboratories, and many rely on sophisticated and expensive equipment. In order to further drive the evolution of THz techniques so that they can be applied to real-world problems, high performance, compact, convenient, and low cost THz sources and detectors must be made available. A new generation of higher efficiency quasi-optical THz devices such as modulators and amplifiers will play a pivotal role in the transformation of THz science from particular applications to the commonplace. In recent years, the development of high-power THz wave sources has received a great deal of attention, and various methods to achieve high-power THz waves have been reported. Examples include the THz wave parametric generation [5], optical rectification of femtosecond pulses in organic and inorganic crystal [6–8], quantum cascade laser (QCL) [9,10], laser plasma interaction [11], and so on. Besides this, amplification of THz radiation has also been demonstrated using QCL in order to fulfill the necessity of high power for practical applications [12,13]. However, these methods generally have their limitations. For example some sources require expensive laser pump sources, sophisticated optics, and so on. Similarly, THz amplification in a QCL requires cooling to cryogenic temperatures, thereby making the system cumbersome and complicated to operate. Therefore, a direct amplification method for THz radiation at room temperature is an attractive way to overcome the sophistication and expense of current technologies. In this Letter, we report experiments that demonstrate a method to amplify single longitudinal mode, pulsed THz 0146-9592/14/061649-04$15.00/0
waves emitted from an injection-seeded THz wave parametric generator (is-TPG). The emitted THz wave is amplified in a nonlinear crystal by means of an optical parametric process. 30 dB gain was achieved for an input THz pulse with energy of a few picojoule. THz wave parametric generation and amplification is based on both second and third order nonlinear processes [14–17]. Among the polar nonlinear crystals, such as LiTaO3 and GaP, we selected LiNbO3 for a number of reasons. First it has a large nonlinear coefficient (d33 25.2 pm∕V at λ 1.064 μm [18], and estimated d33 165 pm∕V at THz wave region [19].) It also has a high figure of merit [7] and high transparency over a wide wavelength range (0.4–5.5 μm). Moreover, the laser induced damage threshold of this crystal is high, making it possible to use higher peak power pump sources without damage to the crystal. LiNbO3 has four infrared- and Raman-active transverse optical (TO) phonon modes, called A1 -symmetry modes. The lowest of these modes (ω0 ∼ 248 cm−1 ) is particularly useful for efficient tunable THz generation and amplification because it has the largest parametric gain and the smallest absorption coefficient [20]. When an intense laser beam propagates through a nonlinear crystal, the photon and phonon transverse fields are coupled and behave as new mixed photon–phonon states, called polaritons. The efficient parametric scattering of light via a polariton (called stimulated polariton scattering) results in the generation of an idler beam whereby THz radiation can be generated. The output of such a system produces a wide range of THz frequencies. However, as shown in Fig. 1(a), when a single frequency seed beam is injected at a specific fixed angle of incidence to the crystal, such that phase matching conditions are met, coherent THz radiation is generated. This process, known as injection seeding, offers a wide frequency range of THz radiation that is tuned simply by varying the wavelength of the seed beam. Based on this fundamental principle, we use the THz wave thus generated (which we call the weak THz field) as a seed beam, © 2014 Optical Society of America
1650
OPTICS LETTERS / Vol. 39, No. 6 / March 15, 2014
Fig. 1. (a) Principle of THz wave generation and (b) THz wave amplification. Here, kseed , kpump , kTHz , and kidler represent the wave vectors of the seed, pump, THz, and idler beams respectively.
as shown in Fig. 1(b). By seeding the THz wave with a specific wavelength, the corresponding idler beam can be amplified, and this in turn leads to amplification of the THz radiation. In this process the noncollinear phase matching condition must be satisfied: i.e., kpump kTHz kidler ; where kpump , kTHz , kidler are the wave vectors for pump, THz wave, and idler beam, respectively. This condition is accomplished by adjusting the angle of incidence of the THz seed beam to the LiNbO3 crystal. The experimental setup is shown in Fig. 2. The system consists of two distinct parts. First, a THz wave emission
Fig. 2. Experimental setup for the emission and amplification of THz radiation. The pump laser (λ 1064 nm, energy 700 μJ∕pulse, pulse width 420 ps, repetition rate 100 Hz, M 2 1.09) is amplified by two optical amplifiers in a double pass configuration. Each amplifier consists of 0.7-at. % Nd3 doped YAG as a gain medium with diameter and length 3 mm and 70 mm respectively. The inset shows the phase matching condition for generation and amplification of THz radiation with a frequency of 2.01 THz. In this figure, polarizing beam splitter and continuous wave are abbreviated as PBS and CW, respectively.
section that comprises a pump laser, laser amplifier, seed laser, and nonlinear crystal. The second part is the THz amplifier section, composed of a pair of nonlinear crystals and a pyroelectric detector to measure the emitted THz power. As a pump source, we used a diode end pumped single-mode microchip Nd3 :YAG laser that is passively Q-switched by a Cr4 :YAG saturable absorber. This laser has very low noise and its peak power fluctuations are lower (< 2%) than other active Q-switched lasers [21]. The pumping beam is amplified by two optical amplifiers and extracted using a polarizing beam splitter (PBS). This setup amplifies the laser power to 14 mJ∕pulse. A continuous wave tunable external cavity diode laser (ECDL, Velocity 6300, New Focus Inc.) with an average power of 5.5 mW is used as an injection seeder for the idler beam. Both pump beam and seed beam irradiate the 5 mol. % MgO:LiNbO3 crystal at a specific angle as shown in Fig. 1(a). We used a 50 mm long crystal with an antireflection coating for a wavelength of 1064 nm. The polarization orientation of the pump, seed, idler, and THz waves are all parallel to the z axis of the crystal. Since the maximum gain coefficient of LiNbO3 occurs at approximately 2 THz [22], we optimized our system for the frequency of 2.01 THz by tuning the seed beam wavelength to 1072.36 nm. In this case, the optimum angle between pump and seed beam is calculated to be 1.96 deg. The emitted THz radiation has pulse energy of 12.1 nJ, a temporal width of 100 ps, and a line width of 5 GHz. For amplification of this THz radiation, we used two nonlinear MgO:LiNbO3 crystals in series. These crystals are pumped by the Nd:YAG laser beam after it is transmitted through the emitter crystal. This recycling of the pump laser makes the system physically far more compact. The energy of the pump beam in this second stage is about 7.2 mJ∕pulse. The THz wave is collimated using a cylindrical lens and efficiently coupled to the nonlinear crystal using a silicon (n 3.4) prism. The THz wave is incident on the crystal so as to satisfy the nonlinear phase matching condition. In this case, the relationship kpump > kidler ≫ kTHz holds. The angle between the pump and idler beam is therefore small whereas the angle between the pump beam and the THz wave is large. This geometry of beams makes it relatively easy to arrange for the weak field THz radiation to be brought to the non-linear crystal as a seed beam. In order to minimize absorption loss of the THz wave and to ensure efficient coupling between the pump beam and the THz wave inside the lithium niobate crystal, we arranged for the pump laser to be as close as possible to the Silicon prism. Moreover, the pump laser and THz wave are temporally overlaid by adjusting the optical path length. In this arrangement, the weak THz field acts as a seed beam and due to the interaction between the laser and THz seed beam in the crystal, amplified THz radiation is emitted. The amplified THz wave within the nonlinear crystal will suffer total internal reflection due to the high refractive index of the MgO:LiNbO3 unless we use an appropriate output coupling medium. We use a Si-prism coupler on the y surface of the crystal to extract the THz wave from the crystal [23]. This pulsed THz radiation is finally detected by a calibrated pyroelectric detector.
March 15, 2014 / Vol. 39, No. 6 / OPTICS LETTERS Output THz energy (nJ/pulse)
2
1.5
1
0.5 (a) 0
1
0.1 0.01 0.001 Input THz energy(nJ/pulse)
0.0001
40 (b)
Gain (dB)
30
20
10
0
1
0.1 0.01 0.001 Input THz energy (nJ/pulse)
0.0001
Fig. 3. (a) Input/output characteristics of the THz amplification system. Here, the THz energy is varied using the combination of set of calibrated THz wave attenuators (TFA-4; CDP Corp) with transmittances of 1%, 3%, 10% and 30%. (b) Gain plotted as a function of input THz energy. The gain is calculated as G 10 log 10E out ∕Ein , where E out and E in are the THz input and output energy respectively.
Figure 3(a) shows how the output pulse energy of the THz amplification system varies as a function of the input pulse energy. Both input and output THz energy are measured using the pyroelectric detector with the input energy ranging from 1 nJ∕pulse to 0.1pJ∕pulse. The input THz energy is varied using the THz wave attenuator. In order to better visualize the results, we plot the gain as a function of input pulse energy in Fig. 3(b). We find a system gain in excess of 30 dB with low values of input energy per pulse (0.1pJ∕pulse). The gain varies inversely with input THz pulse energy. Gain falls almost to unity when the input THz pulse energy is higher than 1 nJ. We attribute this to saturation of the parametric gain of the nonlinear crystal. It is satisfying to note that our system generates THz pulse energies at levels that are easily detected by commercially available room temperature THz detectors. In our system, the gain saturates when the input pulse energy is more than 1 nJ∕pulse. However, commercially available THz detectors are capable of detecting 1 nJ pulses with good signal-to-noise ratio, so amplification beyond this level is not crucial. However, when the THz energy is less than 1 nJ∕pulse, our amplifier demonstrates an unprecedented ability to amplify such weak signal. We have demonstrated direct amplification of THz radiation with gain of more than 30 dB. Pronounced gain is observed when the input THz pulse energy is as small as a few picojoule. However, in the higher input pulse energy regime, gain decreases due to saturation of the
1651
parametric gain of the nonlinear crystal. Significant amplification of THz radiation as demonstrated here represents an advance in the general applicability of THz techniques. This amplification method of THz radiation can be implemented to compensate for losses in strongly absorbing samples in spectroscopic and imaging systems. Furthermore, amplification of THz radiation above the detection threshold of room temperature detectors greatly improves the applicability of, and enhances the versatility of THz sources. At the same time, costs and sophistication are reduced by eliminating the need for high power THz sources. Our initial demonstration may be extended to amplify the weak THz signals emitted by other types of source. The technique could be applied, for example, to difference frequency THz generation in nonlinear crystals and unitraveling carrier photo-diodes. We would like to thank Professor T. Taira of the Institute of Molecular Science, Okazaki, Japan, for his guidance on the subject of high power microchip lasers. We also like to thank K. Wood of QMC Instruments Ltd., for proofreading our manuscript. We also gratefully acknowledge financial support from the JST project, Japan. References 1. E. Pickwell and V. P. Wallace, J. Phys. D 39, R301 (2006). 2. H. Hirori and K. Tanaka, IEEE J. Sel. Top. Quantum Electron. 19, 8401110 (2013). 3. K. Kawase, Y. Ogawa, Y. Watanabe, and H. Inoue, Opt. Express 11, 2549 (2003). 4. S. R. Tripathi, H. Ogura, H. Kawagoe, H. Inoue, T. Hasegawa, K. Takeya, and K. Kawase, Corros. Sci. 62, 5 (2012). 5. S. Hayashi, K. Nawata, H. Sakai, T. Taira, H. Minamide, and K. Kawase, Opt. Express 20, 2881 (2012). 6. S. R. Tripathi, K. Murate, H. Uchida, K. Takeya, and K. Kawase, Appl. Phys. Express 6, 072703 (2013). 7. J. Hebling, K.-L. Yeh, M. C. Hoffmann, B. Bartal, and K. A. Nelson, J. Opt. Soc. Am. B 25, B6 (2008). 8. H. Hirori, F. Blanchard, and K. Tanaka, Appl. Phys. Lett. 98, 091106 (2011). 9. B. S. Williams, Nat. Photonics 1, 517 (2007). 10. K. Vijayraghavan, Y. Jiang, M. Jang, A. Jiang, K. Choutagunta, A. Vizbaras, F. Demmerle, G. Boehm, M. C. Amann, and M. A. Belkin, Nat. Commun. 4, 1 (2013). 11. H. Hamster, A. Sullivan, S. Gordon, W. White, and R. W. Falcone, Phys. Rev. Lett. 71, 2725 (1993). 12. C. Mauro, R. P. Green, A. Tredicucci, F. Beltram, H. E. Beere, and D. A. Ritchie, J. Appl. Phys. 102, 063101 (2007). 13. N. Jukam, S. S. Dhillon, D. Oustinov, J. Madeo, C. Manquest, S. Barbieri, C. Sirtori, S. P. Khanna, E. H. Linfield, A. G. Davies, and J. Tignon, Nat. Photonics 3, 715 (2009). 14. K. Kawase, J. Shikata, K. Imai, and H. Ito, Appl. Phys. Lett. 78, 2819 (2001). 15. K. Kawase, H. Minamide, K. Imai, J. Shikata, and H. Ito, Appl. Phys. Lett. 80, 195 (2002). 16. A. Lee, Y. He, and H. Pask, IEEE J. Sel. Top. Quantum Electron. 49, 357 (2013). 17. Y. Takida, T. Ohira, Y. Tadokoro, H. Kumagai, and S. Nashima, IEEE J. Sel. Top. Quantum Electron. 19, 8500307 (2013).
1652
OPTICS LETTERS / Vol. 39, No. 6 / March 15, 2014
18. I. Shoji, T. Kondo, A. Kitamoto, M. Shirane, and R. Ito, J. Opt. Soc. Am. B 14, 2268 (1997). 19. Y. Avetisyan, Y. Sasaki, and H. Ito, Appl. Phys. B 73, 511 (2001). 20. J. Shikata, K. Kawase, K. Karino, T. Taniuchi, and H. Ito, IEEE Trans. Microwave Theory Tech. 48, 653 (2000).
21. H. Sakai, H. Kan, and T. Taira, Opt. Express 16, 19891 (2008). 22. S. Hayashi, T. Shibuya, H. Sakai, T. Taira, C. Otani, Y. Ogawa, and K. Kawase, Appl. Opt. 48, 2899 (2009). 23. K. Kawase, J. Shikata, H. Minamide, K. Imai, and H. Ito, Appl. Opt. 40, 1423 (2001).
