At the Dawn of a New Era in Terahertz Technology - School of ...

5 downloads 118 Views 1MB Size Report
Available: http://webbook.nist.gov/chemistry/thz-ir/. [44] THz image gallery. [Online]. Available: http://www.thznetwork.org/wordpress/index. php/thz-images/.
INVITED PAPER

At the Dawn of a New Era in Terahertz Technology Development of a direct T-ray laser source is producing results, including advanced lasers and photoconductors as well as a spectrometry system. By Iwao Hosako, Norihiko Sekine, Member IEEE , Mikhail Patrashin, Shingo Saito, Kaori Fukunaga, Member IEEE , Yasuko Kasai, Philippe Baron, Takamasa Seta, Jana Mendrok, Satoshi Ochiai, and Hiroaki Yasuda

ABSTRACT

|

The National Institute of Information and

Communications Technology (NICT, Japan) started the Tera-

magnetic wave; terahertz-range quantum well photodetector; terahertz time domain spectroscopy

hertz Project in April 2006. Its fundamental purpose in the next five years is to enable a nationwide technical infrastructure to be created for diverse applications of terahertz technology. The technical infrastructure includes the development of semiconductor devices such as terahertz quantum cascade lasers, terahertz-range quantum well photodetectors, and high-precision tunable continuous wave sources. It also includes pulsed terahertz measurement systems, modeling and measurement of atmospheric propagation, and the establishment of a framework to construct a materials database in the terahertz range including standardization of the measurement protocol. These are common technical infrastructure even in any terahertz systems. In this article, we report the current status of developments in these fields such as terahertz quantum cascade lasers (THz-QCLs) (with peak power of 30 mW, 3.1 THz), terahertz-range quantum well photodetectors (THz-QWPs) (tuned at 3 THz) an ultrawideband terahertz time domain spectroscopy (THz-TDS) system (with measurement range of from 0.1 to 15 THz), an example of a database for materials of fine art, and results obtained from measuring atmospheric propagation.

KEYWORDS

|

Array detector; atmospheric propagation;

material database; quantum cascade laser; terahertz electro-

Manuscript received March 17, 2007; revised April 13, 2007. I. Hosako, N. Sekine, M. Patrashin, K. Fukunaga, Y. Kasai, P. Baron, T. Seta, J. Mendrok, S. Ochiai, and H. Yasuda are with the National Institute of Information and Communications Technology (NICT), Tokyo 184-8795, Japan (e-mail: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]). S. Saito is with the National Institute of Information and Communications Technology (NICT), Kobe 651-2492, Japan (e-mail: [email protected]). Digital Object Identifier: 10.1109/JPROC.2007.898844

0018-9219/$25.00  2007 IEEE

I. INTRODUCTION Terahertz electromagnetic radiation (from 100 GHz to 10 THz) that lies in the boundary region between light and radio waves has attracted a great deal of attention in recent years due to its ability to achieve innovative sensing systems. Terahertz waves could handle ultrabroadband signals, have very large absorption due to water or water vapor, and are transparent through many materials (e.g., plastic, paper, cloth, and oil) that are opaque in visible and IR light. Many materials have a so-called fingerprint spectrum in the spectrum range. Therefore, terahertz waves are expected to be applied to ultrafast wireless communications, scanning systems for hazardous materials, and assay devices for medical examinations. They are also expected to be applied to the multiresidue analysis of agrichemicals, medical diagnostics, environmental assessment, process monitoring systems for industrial products, and biometric security. There is an atmosphere prevailing that terahertz technology represents the dawn of a new era. The Ministry of Internal Affairs and Communications (MIC), Japan, conducted an exhaustive fact-finding study on the current status and prospects of terahertz technology in fiscal 2004 (the chairperson was Professor Masayoshi Tonouchi of Osaka University). Fifty Japanese experts and 15 researchers from overseas who were eminent in various scientific fields participated in MIC’s investigative research as committee members. They discussed the current status and prospects of terahertz-wave engineering, terahertz photonics, terahertz electronics, and other aspects of these technologies (including roadmaps) for a total of 100 hours. Part of the final report of MIC’s investigative research was published in May 2006 [1] and part of the roadmap for specific applications has been Vol. 95, No. 8, August 2007 | Proceedings of the IEEE

1611

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

presented elsewhere [2]. The National Institute of Information and Communications Technology (NICT), Japan, started research and studies on the Terahertz Project in April 2006. Its fundamental purpose in the next five years is to enable some kind of a technical infrastructure to be created for diverse applications of terahertz technology. Our final goal is to create a technical basis for ultrahigh bit-rate wireless communication systems. The technical infrastructure includes the development of semiconductor devices such as terahertz quantum cascade lasers (THz-QCLs), terahertz-range quantum well photodetectors (THz-QWPs), and high-precision tunable continuous wave sources. It also includes pulsed terahertz measurement systems, modeling and measurement of atmospheric propagations, and the establishment of a framework to construct a materials database in the terahertz range including standardization of the measurement protocol. The technical infrastructure is common even in security and biosensing systems. For example, the combination of a small light source with relatively high power output and a large-format sensitive array detector would significantly decrease the cost of a terahertz imaging system for use in security. A model of atmospheric attenuation would give us optimum bands for stand-off sensing systems. A materials database in the terahertz range would enable components to be discriminated by security and biosensing systems. We report the current status of developments in these fields in Section II, such as those for devices including THz-QCLs and THz-QWPs. We then report the development of an ultrawideband terahertz time domain spectroscopy (THz-TDS) system in Section III, and present an example of a materials database in Section IV. The results obtained from measuring atmospheric propagation are presented in Section V.

conductor devices, especially QCLs, as candidates that would satisfy these requirements. QCLs are semiconductor devices originally proposed by Kazarinov and Suris [3] and first developed by Faist et al. [4] in the midinfrared region and Ko¨hler et al. [5] in the terahertz region. In QCLs, light emission is obtained through intersubband transitions in the quantum wells. This is a critical difference between semiconductor lasers currently used in optical communications in which light is emitted by interband transitions. Fig. 1 shows the conduction-band profile of the active region of a QCL under the operating bias field [6]. QCLs consist of a periodic sequence of alternately grown thin layers of different semiconductor materials. Light emission occurs through the intersubband transition between the upper (3) and lower laser states (2), and the electrons are then removed by the extractor and reinjected into the next active part by resonant tunneling. QCLs have some characteristic properties due to this specific structure. The first is emission frequency. As can be seen from Fig. 1, the emission wavelength is not determined by the physical parameters of the material, but by the well/barrier width. Therefore, we can set any emission frequency in the range of the band-offset energy of the quantum wells. Second, since the active modules are repeatedly connected, electrons are recycled in the active region. This results in high efficiency and power. The third property results from the dispersion relation of the subbands in the quantum wells. In interband transitions, the curvature of the conduction band has an opposite sign from that of the valence band. As a result, the transition energy depends on the in-plane momentum ðk//Þ

II. TERAHERT Z S EMICONDUCTOR DEVICES A. Quantum Cascade Lasers Terahertz pulses that are generated with femtosecond (fs) lasers and terahertz emitters, such as photoconductive antennas and electrooptic crystals, have been used in research. Even though terahertz pulses are useful in THz-TDS due to their broad spectral bandwidth, terahertz sources with higher output power are required in applications to sensing, security, and environmental monitoring. The size of the terahertz source is also important since, in the emission of THz-pulses, fs-lasers (e.g., Ti : sapphire lasers) are very large and the use of these system is largely limited to the laboratory. Furthermore, continuous-wave operation of the sources is necessary in applications to high-speed wireless communication and ultrafast computing. We focused on semi1612

Proceedings of the IEEE | Vol. 95, No. 8, August 2007

Fig. 1. Conduction band profile of a GaAs/AlGaAs THz QCL. Beginning with the left injection barrier in the dotted box, the layer thicknesses in A˚ are 54/78/24/65/38/149/30/95. The 149 A˚ well is doped at n ¼ 1:9  1016 cm3 [6]. The THz emission occurs though the radiative transition between state 3 and state 2. Since E is set to be ELO (ELO : LO phonon energy), electrons in state 2 rapidly relax to state 1 by LO phonon scattering.

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

of the electron and the density of state (DOS) becomes like a step function. In intersubband transitions, however, since the curvature has the same sign between sublevels, the transition energy is independent of k// and the DOS becomes like a delta function (in fact, due to the effect of nonparabolicity, the transition energy slightly depends on k//). Therefore, a narrow linewidth and high optical gain are expected. We describe the GaAs/AlGaAs THz-QCL fabricated by our group at NICT in the following. The structure of the sample was based on that of the MIT group, i.e., the resonant phonon depopulation type [6]. The QC module consists of a gain (states 3 and 2) and a longitudinal optical (LO)-phonon depopulation parts (states 2 and 1). Since the energy separation of the latter part ðEÞ was set at 9 36 meV, which corresponds to the GaAs LO phonon energy, the electrons in the lower laser state were rapidly extracted by LO phonon scattering. This led to efficient population inversion, which is required to achieve lasing action. We prepared a QCL sample with 480 modules. The sample was grown on a semi-insulating GaAs substrate by molecular beam epitaxy (MBE). The growth rate was set at  1 m/h. We evaluated the layer thickness in the active region by X-ray diffraction (XRD) spectroscopy. We confirmed the layer thickness was less than 1% thinner than the designed value by comparing the peak positions between the measured results and the spectrum calculated for the designed QC structure. The laser chip was fabricated with the standard wet process. We adopted a semiinsulating surface-plasmon waveguide structure [7] for the terahertz waveguide. The mesa was  200-m wide and the cavity was 3 mm long. The front mirror was a cleaved facet and the back facet was coated with an SiO2 =Ti/Au multilayer to attain high reflectivity. Fig. 2(a) plots the light output-current ðILÞ and voltage-current ðIVÞ curves measured for the QCL. These characteristics were obtained at 39 K during pulsed operation. The repetition rate was set at 460 Hz and the width of the current pulse was set at 200 ns. As can be seen, the threshold current is 5 A and the threshold current density is less than 1 kA/cm2 . Moreover, we observed a high peak-output power of around 30 mW. We found a dip in the IL curve at a bias current of 6 A. This originated from absorption in the air since the gain region in this QC structure has a diagonal transition nature and the transition energy shifts with increasing bias current due to the Stark effect. Fig. 2(b) shows the lasing spectrum for a sample measured just above the threshold current by FTIR. As can be seen, this lases in a single mode at low bias currents and the lasing frequency is 3.1 THz. We further measured the dependence of the threshold current density (not shown) on temperature ðTÞ. We confirmed lasing up to 123 K in pulsed operation. Even though the T-dependence is poorly fit by expðT=T0 Þ with single T0 (i.e., characteristic temperature), we obtained a

Fig. 2. Fundamental characteristics of fabricated THz QCLs. (a) Pulsed current-voltage (IV, black) and current-peak output power (IL, red) curves. The dip in the IL curve originates from the absorption in the air. (b) Emission spectrum just above threshold current at the same temperature as (a). The emission frequency corresponds to 3.1 THz.

T0 of 113 K for the worst case in the high-temperature region.

B. Terahertz-Range Quantum Well Photodetector There have been continuing efforts in recent years to develop practical sources and detectors of terahertz radiation [6], [8]–[11]. Most existing and emerging applications will utilize direct or heterodyne techniques of detection for terahertz imaging, spectroscopy and communications, and their successful implementation will depend on the availability of fast and sensitive multielement detectors. There have traditionally been very limited choices of detectors with suitable characteristics in the Vol. 95, No. 8, August 2007 | Proceedings of the IEEE

1613

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

terahertz region of the spectrum. Significant improvements have been achieved in the performance of mixers for heterodyne receivers below 1–2 THz, specifically, superconductor–insulator–superconductor (SIS) tunnel junctions and superconducting hot electron bolometers (HEBs). A number of photon detectors on the higher frequency side based on extrinsic semiconductors provide a viable option for some applications. Extrinsic Ge : Ga photodetectors, for example, offer sensitivities at least orders of magnitude better in the direct mode than superconducting bolometers operating at the same temperatures. Despite this progress, the significance of the terahertz region for practical imaging applications can only be fully attained with large-format multielement detectors. It seems plausible, however, that evolutionary improvements and scaling of available technologies may not meet these requirements or could be prohibitively expensive, which would make alternative device concepts and materials worth considering. One possible candidate that we are currently investigating is the GaAs/AlGaAs multi-quantum-well (QW) detector based on intersubband absorption in the quantum well. A suitable band structure is achieved with an appropriate width for the QW layer and a suitable composition for the alloy in the barrier layers. Compound semiconductors, such as GaAs/AlGaAs, provide reliable and well-established material systems for designing quantum-well photodetectors in a targeted spectral range. These were primarily used for near- and middle-infrared devices until recently. The processing technology is fully mature and large-format infrared arrays with up to 1024  1024 pixels have been demonstrated [12]. The optimal design for a QW photodetector with a minimal level of dark current corresponds to a situation where the first excited state in the quantum well is aligned with the top of the barrier (bound to a quasi-bound configuration). If we apply a similar methodology to the range between 1 and 8 THz, the parameters of the QW structure will correspond to low aluminum fractions of a few percent (1%–5%) and QW widths between 10 and 30 nm [13], [14]. We have designed a GaAs/AlGaAs detector with a targeted peak frequency of 3 THz (100 m) in an effort to extend the successful implementation of infrared QW arrays to the terahertz range. A simple multilayer structure has 18-nm GaAs QWs sandwiched between AlGaAs barriers with an Al alloy fraction of 2%. The effect of different barrier widths and doping concentrations on the expected dark current and spectral response of the structure were numerically simulated, and several samples with various in-well doping concentrations ð5  1016  2  1017 cm3 Þ and barrier widths (60 and 80 nm) were grown by MBE on semi-insulating GaAs substrates. X-ray diffraction, scanning electron microscopy/energy dispersive spectrometer (SEM/EDS), and photoluminescence (PL) measurements were used to verify the composition, 1614

Proceedings of the IEEE | Vol. 95, No. 8, August 2007

period, and energy-level structure of the samples. Only minor deviations from the designed parameters were observed. The samples were processed into single-element square-shaped mesas with top and bottom ohmic contacts of different sizes using standard photolithography, wet etching, and thermal deposition. A simple grating coupler was implemented on top of the mesas to allow front-side illumination. Despite the low Al content (2%), we obtained consistent results with MBE growth and could control the level of dark current and the impedance of fabricated devices within design-specific requirements. The level of dark current in optimized samples was a few A/cm2 (Fig. 3) and the detector observed response close to the designed detection frequency (Fig. 4). Discrepancy between the expected spectrum and the experiment can be traced to a limitation of the model. Broad feature in the spectrum is caused by bound-tocontinuum transitions, which were not considered in the present model. The responsivity of the detector was measured with a calibrated blackbody source. A responsivity of about 10 mA/W at an electric bias of 40 mV and an operating temperature of 4 K was obtained by comparing current–voltage characteristics under different photon flux conditions. We believe that optimizing the in-well doping concentration further and improving the design of the grating coupler can produce even better performance. We are currently evaluating the viability of this type of detector for practical applications with a prototype 32-element array.

I II . WIDEBAND THz-TDS Radiation in the terahertz range is roughly defined from 0.1 to 10 THz, and has been an Bundeveloped gap[ between electronics and optics. The technique of terahertz spectroscopy is very attractive because many materials

Fig. 3. Dark current characteristic at various temperatures. Below 8 K the current flattens out to about 1 A/cm2 .

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

Fig. 5. Example of standard THz-TDS system with PC antenna emitter Fig. 4. Spectrum of photoresponse compared with the result of numerical simulations. Detector observed response close to the designed detection frequency.

have an absorption band, which is called a fingerprint, and this has been focused on in the research field of materials analysis [15], [16]. We have to find the fingerprint spectra of target materials and the adjacent materials in security and/or biological uses. This is why broadband terahertz spectroscopy is needed for basic research. Fourier transform infrared (FTIR) spectroscopy was used to measure specimens in the terahertz region for a long time. Coolant is needed to measure low-frequency and/or weak signals. A powerful new tool for measuring in the terahertz region called THz-TDS was developed in the past decade. Electric waveforms of terahertz radiation are generated and measured by gated detection with a short pulse laser [17]. THz-TDS is applied to various applications, for example, superconductors [18], elementary excitation in solid-state materials [19], imaging [20], [21], DNA [22], thin film [23], terahertz tomography [24], and so on. THz-TDS measurements have three main advantages over FTIR. 1) The waveform of transmitted or reflected terahertz waves are obtained using THz-TDS measurements. Accordingly, the refractive index and absorption coefficient of the sample are calculated from the phase and amplitude of the waveform. 2) THz-TDS can be applied to investigate multilayer specimens by using information in the time domain waveforms. 3) There is no need to use coolant for the detector. However, as most THz-TDS systems cover a frequency range from 0.1 to 3 THz, a broadband system is necessary for application to security and biological uses. Electrooptic (EO) crystal and/or photoconductive (PC) antennas with short-optical-pulse lasers are used to generate and detect terahertz waves in THz-TDS. Fig. 5 shows an

and detector. Silicon lens are attached on the biased emitter and detection antenna in order to reduce the reflection of terahertz wave.

example of a standard THz-TDS system with PC antennas. There is a silicon semispherical lens attached to the biased emitter and detection antenna to reduce the reflection of terahertz waves by the boundary of the substrate of emitter and air. There is an insensitive region in measuring terahertz waves, because of the reststrahlen band, which is typically 5–10 THz. Several groups have developed the broadband THz-TDS system and applied the system to spectroscopy aggressively. Thin EO crystals and/or PC antennas with ultarashort pulse laser are used in order to generate and detect the broadband terahertz wave [25]–[32]. We have developed the broadband THz-TDS system outlined in Fig. 6. The system is based on the result of our long years of research [29]–[32]. A frequency region of 0.1–15 THz is covered by this system. The

Fig. 6. Developed broadband THz-TDS system. The biased emitter was arranged as reflection geometry and terahertz radiaiton and optical-gate pulse were incident onto the fabricated surface of detection antenna.

Vol. 95, No. 8, August 2007 | Proceedings of the IEEE

1615

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

details of the system are as follows. We used a modelocked Ti : sapphire laser as the light source to generate and detect terahertz waves. It had a center wavelength of 800 nm, a spectral width of 100 nm at a repetition rate of 78 MHz, and an average power of 450 mW. The output pulse was split into two parts: the first was focused onto the surface of an ac-biased low temperature grown GaAs (LT-GaAs) PC antenna used as the terahertz radiation source and the second was focused onto another antenna used as the detector. The autocorrelation width on the PC antennas’ surface was 20 fs using a dispersion-compensation mirror and keeping dispersive optics to a minimum. The generated terahertz radiation was collimated with an off axis parabolic mirror and then focused onto the sample with another parabolic mirror. Transmitted terahertz radiation was collimated and focused onto the PC antenna for detection by two off axis parabolic mirrors. The optical-gate pulse was merged into the terahertzradiation path by a high-resistivity silicon wafer inserted between the two off axis parabolic mirrors. Silicon has a high THz refractive index and will there fore reflect a significant portion (about 40%) of the THz radiation. Terahertz radiation from the emitter was collected by reflection geometry and transmitted terahertz radiation from the sample and gate pulse were incident onto the fabricated surface of the PC antenna to avoid absorption by the GaAs antenna substrate. The terahertz-wave emitter had a strip line, and was ac biased at a frequency of about 1.7 kHz. The antenna was a 30-m dipole antenna with a 5-m gap in the center. A silicon semispherical lens was attached to the backside of PC antenna to minimize reflection caused by the boundary between the GaAs substrate and air. The photocurrent from the PC antenna was preamplified and then detected with a lock-in amplifier. The signal to noise (S/N) ratio for the system was 1000 : 1. The terahertz-radiation path was enclosed in a vacuumtight box, which was purged with nitrogen gas to reduce the effect of water-vapor absorption. The measurements were done at room temperature. We measured spectrum of nitrogen gas and of air which length was about 80 cm. And transmission spectrum of air was obtained and it is shown in Fig. 7. Many absorption lines are observed. We developed a broadband THz-TDS system and the experimental results suggest that it can effectively be applied to security and biological fields. We developed a broadband THz-TDS system and the experimental results suggest that it can effectively be applied to security and biological fields.

IV. MATERI ALS DATAB ASE IN TERAHERTZ RANGE There are various ongoing projects that apply terahertz spectroscopy and its imaging systems [33]–[35] to industrial applications. Most active fields involve security issues, such 1616

Proceedings of the IEEE | Vol. 95, No. 8, August 2007

Fig. 7. Transmission spectrum of air measured by the broadband THz-TDS system illustrated in Fig. 6.

as the detection of hidden weapons, explosives, and illegal drugs [36]–[38]. Biological, medical, and pharmaceutical applications are also promising [39], [40]. Any kind of spectroscopy requires spectra databases for practical applications. The spectrum of an unknown material is identified by comparing it against Bknown[ materials in a spectral database. Limited number of spectra is sufficient when the target materials are specific, such as illegal drugs. Various databases are needed to enlarge the range of applications to make terahertz spectroscopy a popular measure of materials. There is little need to discuss repeatability or reliability of the system itself in common midinfrared spectroscopy because the technology is already mature. Thousands of databases and standard materials are commercially available, and it is also common for user groups to collect and share data. For example, the spectra database of a large number of pigments and binders is supported by various museums and universities around the world [41], [42]. Terahertz spectra, on the other hand, have mainly been collected to investigate atmospheric transmissions, and spectra databases for solids or liquids are therefore limited to a few research fields. In addition, the technology for general terahertz spectroscopy is still progressing, and its performance greatly depends on the system. FTIR spectrometers provide data from 20 to 670 cm1 (or 0.6 to 20 THz), and most of the THz-TDS systems with femtosecond pulse lasers are designed for a frequency range from 0.2 to 3 THz (or 10 to 100 cm1 ). The frequency resolution and sensitivity vary among measurement systems. Therefore, intercomparison tests by organizations should be carried out to establish a reliable database. There are several open databases for terahertz spectroscopy. The European project BTHz-Bridge: Terahertz radiation in biological research, investigations on diagnostics, and study on potential genotoxic effects[ compared the spectra of the same synthetic materials obtained by different organizations, and the database provided

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

spectra from 100 to 300 GHz, 300 GHz to 3 THz, and 3 THz to 20 THz, with the organizations’ names [39]. Since the project was for biological issues, the database provided various membranes, proteins, and other biological materials. The National Institute of Standard and Technology (NIST) provides 36 spectra for foods and medicines and various others on their website, prepared from pellets with polyethylene spectrograde powder. Sugar and ground sugar have large differences in spectra as well as salt and ground salt. The frequency range is from 40 to 400 cm1 (or 1.2 to 12 THz) [43]. The Terahertz Science and Technology Network research group has provided some images of spectra as images in their terahertz image gallery [44]. There is a government funded project in Japan, and it could be opened to approved users [45]. A database for developing a terahertz remote-sensing system in emergencies is being established at NICT. The material data is for buildings, clothes, and human bodies. Toxic gases are also important, and are discussed in Section V. NICT, as a national institution, should contribute to standards and regulations for measurement systems using terahertz. It should prepare a standard procedure to ensure the efficiency of various terahertz spectroscopy systems, starting from proper preparation of standard specimens that can be applied to all systems. The first-ever terahertz spectra database for materials of fine art has been established as a new application of terahertz spectroscopy by using the wideband THz-TDS system previously described and the classical FT-THz system [46]. The frequency range is from 0.5 to 15 THz. It will be open to the public by the end of 2007. The materials are pigments, binders, varnishes, and adhesives used in historical paintings and in present conservation techniques [47], including more than 100 spectral data on organic and inorganic pigments, natural resins, and synthetic polymers. Fig. 8(a) shows spectra of blue pigments: natural ultramarine (lapis lazuli), artificial ultramarine, azurite, cobalt blue deep, and organic natural indigo, which were used in a stained glass like sample in Fig. 8(b). These painted pigments having the same blue can be distinguished by their transmission values at certain frequencies. At 2.3 THz, for example, the blue part was displayed as a gray scale as shown in this figure. THz spectroscopy can also distinguish binders. Fig. 9 has examples of the transmission terahertz spectra of common binders. Beeswax has been used as a binder since the civilization of Pompeii and is commonly used as a wax in medicines. It has a slightly broad absorption peak at 6.4 THz. Most balsams have similar spectra with several absorption peaks above 12 THz. Colorless transparent binders can be displayed as a gray scale of their transmission values at certain frequencies. Fig. 10(a) shows cobalt blue medium plates painted with four binders; petrol, beeswax, acrylic resin, and natural wax. The painted binders appear as in Fig. 10(b) at 6 THz.

Fig. 8. Example of spectra of pigments and THz false colour image. (a) Spectra and gray scale chart base on the spectra of blue pigments. (b) Application of THz spectroscopy for determining pigments in a painting. [The frequency for determining the gray scale of each pigments is now clearly shown as 2.3 THz in Fig. 8 (a)].

Most spectra for the pigments are sharp and appear between 2 and 15 THz, so that terahertz spectroscopy can be used to individually determine pigments and binders from the database. Since materials used in fine art are made of natural resources and are used in various industries, our database can contribute to earth science, Vol. 95, No. 8, August 2007 | Proceedings of the IEEE

1617

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

Fig. 9. Example of spectra of binders.

Fig. 11. Overview of the NICT THz remote sensing.

environmental studies, food and pharmaceutical industries, and related fields of research.

V. DEVELOPMENT OF NICT TERAHERTZ-WAVE PROPAGATION MODEL We are developing an NICT THz-wave remote sensing system to observe Earth’s atmosphere. Our target frequency region is 0.1–33 THz. Three research projects are currently in progress: 1) development of the NICT THz-wave radiative transfer model; a) construction of the radiative transfer model including atmospheric absorption, emission, and scattering; b) laboratory experiment for spectroscopic parameters, such as pressure broadening parameters of molecular line transitions and continua in the THz region; 2) feasibility study of new satellite measurement in THz frequency region; 3) development of THz observation system to monitor city air, such as humidity and air quality.

An overview of the THz remote sensing is given in Fig. 11. In this paper, we describe the recent progress towards a THz-wave propagation model including the development of an atmospheric radiative transfer model as well as the derivation of accurate spectroscopic parameters from laboratory experiments. Several radiative transfer models exist in the radiowave region as well as in the infrared region, but there are few considerations the THz frequency region itself. Fig. 12 illustrates the discrepancy between radiowave and infrared wave propagation models. The radiowave model was calculated using the JPL line catalog [48] for the absorption by atmospheric molecular transition and Liebe model [49] for atmospheric continuum absorption, which

Fig. 12. The discrepancy between radiowave and infrared wave

Fig. 10. Visible and THz false color image of cobalt blue plates with four binders. Clockwise from the top right: oil, beeswax, acrylic resin, and natural balsam. (a) Visible and (b) THz false color image at 6 THz.

1618

Proceedings of the IEEE | Vol. 95, No. 8, August 2007

propagation models. The radiowave model was calculated using JPL line catalog [48] for the absorption by atmospheric molecules and Liebe model [49] for atmospheric continuum absorption. The infrared model was calculated on the basis of the HITRAN line catalog [51] and the CKD atmospheric continuum model [52].

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

are recommended to be used in the 0–1 THz by the Radio Communication Sector of the International Telecommunications Union (ITU-R.P.676-6, 2007) [50]. The infrared model was calculated on the basis of the HITRAN line catalog [51] and the Clough, Kneizys, and Davies (CKD) atmospheric continuum model [52]. HITRAN catalog and CKD model are widely used, e.g., in the LBLRTM model [53]–[55]. The U.S. standard midlatitude atmospheric scenario at the ground (0 km) was used for the calculation. What is the biggest difference of the radiative transfer between THz-wave and other frequency regions? The THz region is characterized by: 1) dense water vapor line absorption and 2) large atmospheric continuum absorption due to collision complexes with N2 , O2 , and H2 O. In the past, it has been difficult to measure parameters describing these two phenomena accurately in the THz region due to the instrumental BTHz gap[ problem. We are performing laboratory experiments to measure both pressure broadening parameters of water vapor lines and background continua in THz region.

A. Radiative Transfer Model Within the framework of NICT’s THz project a new radiative transfer (RT) model is developed for microwave to midIR wavelengths (0.1–33.0 THz). The purpose is to provide a flexible tool that is suitable to study topics related to remote sensing, monitoring of air quality, determination of the global energy budget, as well as communication system development. So far, several models are currently used in the team: the SMILES Observation retrieval COde (SMOCO) [56], the Microwave Observation LIne Estimation and REtrieval (MOLIERE) [57] and the Approximate Spherical Atmospheric Radiative TRansfEr model (SARTre) [58]. Both SMOCO and MOLIERE have been applied for theoretical studies and retrieval of trace gas and temperature profiles from measurements of several microwave and sub-mm instruments [59]–[61], while the derivation of cloud properties from MIPAS data [62] has been demonstrated using SARTre [63]. Recently, MOLIERE and SARTre have jointly been used for preliminary calculations in the feasibility study towards a THz satellite instrument. The capabilities of THz observations concerning cloud, humidity and trace gas remote sensing are examined within this study. Effects of high altitude ice clouds on measurements around selected water vapor lines in the THz region feasible to be used in cloud retrievals are illustrated in Fig. 13. Common to the previously mentioned models is the consideration of a horizontally homogeneous atmosphere, i.e., the assumption of a 1-D atmosphere. Among them, only SARTre accounts for effects of aerosols and clouds, including emission and absorption as well as scattering contributions. On the other hand, both SMOCO and MOLIERE provide fast modules to estimate weighting functions with respect to atmospheric parameters, e.g., temperature and trace gas abundance, as well as an

Fig. 13. Effects of high altitude ice clouds on nadir observations around several water vapor lines. The cloud (ice water path IWP ¼ 10 g/m  2, effective particle size De ¼ 12 m) is located between 15 and 16 km altitude. While the lower frequency measurements are sensitive to the atmospheric state down to the lower troposphere and the surface (compare brightness temperatures outside the different lines, that correspond to the physical temperature of the lowest sounded altitude), the higher frequency observations contain information about the ice clouds.

inversion procedure based on the optimal estimation method [64]. The new RT model will combine capabilities of these models and extend them, e.g., to take into account the horizontal inhomogeneity of atmospheric parameters by assuming a 2-D atmosphere. The main characteristics are as follows: • refracted ray path in a spherical 2-D atmosphere; • emission, absorption, and scattering of radiation by the planetary surface and the atmosphere (considering both gas phase and particulate matter), assuming local thermodynamic equilibrium; • line-by-line calculation using standard spectroscopic line catalogs (HITRAN [51], JPL [48]) and semiempirical continuum models for dry air and water vapor (Liebe [49], CKD [54]); • optical properties of aerosols and clouds from external calculations and databases (e.g., [65], [66]); • scattering source term using DISORT [67] assuming a locally plane-parallel atmosphere; • analytical weighting functions for atmospheric parameters; • instrumental function of the detector; • inversion modules for 1-D and 2-D retrieval. In addition to standard spectroscopic line catalogs and continuum models, spectroscopic parameters derived from ongoing measurements at the NICT laboratory (see Section V-B) will be used. For the calculation of the weighting functions analytical expressions will be derived for nonscattering cases while algorithms to handle scattering cases will be investigated in the future.

B. Laboratory Experiment for Provision of Spectroscopic Parameters Atmospheric absorption of THz waves is commonly described by the two components. One is the line Vol. 95, No. 8, August 2007 | Proceedings of the IEEE

1619

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

contribution, which arises mainly from the resonance absorption of water vapor. The calculation of atmospheric transmission requires line positions, intensities, pressure broadening coefficients, and pressure shift coefficients. These parameters need to be carefully determined by spectroscopic laboratory measurements. The accuracy of atmospheric parameters derived based on the model directly depend on the accuracy of the laboratory measurements. A large amount of water vapor spectroscopic data has been collected in the JPL [48] and HITRAN [51] databases, widely used for the computation of atmospheric transmission. However, in the THz and far infrared region, only a small number of experimental studies have been reported, especially for pressure broadening and pressure shifting for pure rotational spectra of water vapor and its isotopomers [68], [69]. The other component of atmospheric absorption is the continuum, which in practice is defined as the excess of absorption unaccounted for by the resonance water vapor spectrum. When atmospheric spectra are simulated by a collection of lines, the results are in poor agreement with long path observations. This excess was found to affect the atmospheric wave propagation over the broad spectral range from the microwave to the infrared [52], [70], [71]. This so-called continuum absorption is not yet completely understood. Anomalous far-wing absorption, absorption by water vapor dimers or larger clusters, and absorption induced by collisions between atmospheric molecules have been proposed to explain the continuum absorption [72]–[74]. Because the exact physical nature of the continuum is still not understood, large uncertainties remain related to the prediction of far-wing absorption. Clough, Kneizys, and Davies (CKD) have devised a widely used semiempirical line shape function for water vapor adjusted to give reasonable agreement with laboratory experiments and atmospheric observation [52], [54]. Tipping and Ma have developed a collision theory to account for the continuum and their model is in good agreement with the available data [74]. However, experimental data is required. We aim at the development of a THz wave propagation model based on accurate spectroscopic parameters of the line and continuum parts. To develop a reliable model, the inclusion of broadband data is essential. Large numbers of carefully recorded spectra by different experimental apparatus are required for that purpose. Meshkov et al. developed an instrument for the measurement of the continuum in the millimeter and submillimeter spectral region for mixtures of gases [75]. Their instrument is based upon a fast scanning cavity ringdown approach that makes possible the measurement of continuum and broad line absorption at 6000 distinct cavity frequencies from 0.17 to 0.26 THz [75]. Fourier transform spectrometers (FTSs) with a mercury lamp and a liquid He cooled bolometer detector are widely used for the far infrared broadband measurements. 1620

Proceedings of the IEEE | Vol. 95, No. 8, August 2007

Gasster et al. measured the pure rotational spectrum of water vapor in the spectral range 0.75–3.3 THz using FTS and extracted pressure broadening coefficients [68]. Podobedov et al. conducted laboratory absorption measurements of the water vapor continuum in the far infrared region from 0.4–1.8 THz using a multipass absorption cell and FTS [76]. Due to the development of femtosecond lasers, recently terahertz time-domain spectroscopy (THz-TDS) has become popular in the spectroscopy in THz region [72]. Grischkowsky et al. measured THz spectra of some gas molecules by THz-TDS [72], [77]. The development of such experimental techniques is required for the derivation of precise spectroscopic parameters. Within the THz project, pressure broadening coefficients of water vapor have been derived from laboratory measurements. We have measured N2 - and O2 -broadened coefficients of water vapor in the range of 0.5 to 7 THz using both FTS and THz-TDS [78], [79]. Han et al. performed a comprehensive comparison between THzTDS and far infrared FTS showing that in terms of signalto-noise ratio, THz-TDS is advantageous at frequencies below 3 THz, while FTS works better at frequencies above 5 THz [80]. The combination of these techniques allows us to measure broadening parameters in a wide frequency range. Furthermore, pressure broadening parameters of water vapor isotopomers like HDO and D2 O have been measured. The measured data may improve the accuracy of atmospheric parameters required for applications in climate modeling, radio astronomy, and satellite remote sensing.

VI. SUMMARY The National Institute of Information and Communications Technology (NICT, Japan) has started the Terahertz Project. Its fundamental purpose is to enable technical infrastructure to be created for diverse applications of terahertz technology. The technical infrastructure includes the terahertz semiconductor devices, the wideband THz-TDS, the modeling and measurement of atmospheric propagations, and the establishment of a framework to construct materials databases in the terahertz range. These are common technical infrastructure even in any terahertz systems. h

Acknowledgment The authors would like to thank to Prof. M. Tonouchi, Prof. N. Hiromoto, and Dr. K. Sakai for their support and continuing encouragement. The authors would also like to thank Prof. K. Hirakawa for his encouragement. The authors would also like to thank Istituto per il restauro L’Ambiente for preparing a stained-glass like specimen with materials supplied by Zecchi, and for useful discussions on conservation science.

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

REFERENCES [1] M. Tonouchi, Terahertz Technology. Tokyo, Japan: Ohmsha, 2006. [2] M. Tonouchi, BCutting-edge terahertz technology,[ Nature Photonics, vol. 1, pp. 97–105, 2007. [3] R. F. Kazarinov and R. A. Suris, BPossibility of amplification of electromagnetic waves in a semiconductor with superlattice,[ Sov. Phys. Semicond., vol. 5, pp. 707–709, 1971. [4] J. Faist, F. Capasso, D. L. Sivco, C. Sirtori, A. L. Hutchinson, and A. Y. Cho, BQuantum cascade lasers,[ Science, vol. 264, pp. 553–556, 1994. [5] R. Ko¨hler, A. Tredicucci, F. Beltram, H. Beere, E. Linfield, A. Davies, D. Ritchie, R. Lotti, and F. Rossi, BTerahertz semiconductor-heterostructure laser,[ Nature, vol. 417, pp. 156–159, 2002. [6] B. S. Williams, S. Kumar, H. Callebaut, and Q. Hu, BTerahertz quantum-cascade laser operating up to 137 K,[ Appl. Phys. Lett., vol. 83, no. 25, pp. 5142–5144, 2003. [7] B. S. Williams, S. Kumar, H. Callebaut, and Q. Hu, BTerahertz quantum-cascade laser at   100 m using metal waveguide for mode confinement,[ Appl. Phys. Lett., vol. 83, no. 11, pp. 2124–2126, 2003. [8] B. S. Williams, S. Kumar, Q. Hu, and J. L. Reno, BOperation of terahertz quantumcascade lasers at 164 K in pulsed mode and at 117 K in continuous-wave mode,[ Opt. Express, vol. 13, pp. 3331–3339, 2005. [9] L. Ajili, G. Scalari, N. Hoyler, M. Giovannini, and J. Faist, BInGaAs-AlInAs/InP terahertz quantum cascade laser,[ Appl. Phys. Lett., vol. 87, no. 141 107, 2005. [10] H. Yasuda, I. Hosako, S. Miyashita, and M. Patrashin, BTerahertz electroluminescence from GaSb/AlAs quantum cascade laser,[ Electron. Lett., vol. 41, pp. 1062–1063, 2005. [11] M. Fujiwara, T. Hirao, M. Kawada, H. Shibai, S. Matsuura, H. Kaneda, M. Patrashin, and T. Nakagawa, BDevelopment of a gallium-doped germanium far-infrared photoconductor direct hybrid two-dimensional array,[ Appl. Opt., vol. 42, no. 12, pp. 2166–2173, 2003. [12] S. D. Gunapala, S. V. Bandara, J. K. Liu, C. J. Hill, S. B. Rafol, J. M. Mumolo, J. T. Trinh, M. Z. Tidrow, and P. D. LeVan, B1024  1024 pixel MWIR and LWIR QWIP focal plane arrays,[ Proc. SPIE Int. Soc. Opt. Eng., vol. 5783, pp. 789–803, 2005. [13] H. C. Liu, C. Y. Song, A. J. SpringThorpe, and J. C. Cao, BTerahertz quantum-well photodetetor,[ Appl. Phys. Lett., vol. 84, pp. 4068–4070, 2004. [14] H. Luo, H. C. Liu, C. Y. Song, and Z. R. Wasilewski, BBackground-limited quantum-well photodetetor,[ Appl. Phys. Lett., vol. 86, no. 231 103, 2005. [15] B. Ferguson and X. C. Zhang, BMaterials for terahertz science and technology,[ Nature Mater., vol. 1, pp. 26–33, Sep. 2002. [16] K. Sakai, Ed., Terahertz Optoelectronics. Heidelberg, Germany: Springer Berlin, 2005. [17] D. H. Auston, K. P. Cheung, and P. R. Smith, BPicosecond photoconducting Hertzian dipoles,[ Appl. Phys. Lett., vol. 45, pp. 284–286, Aug. 1984. [18] R. A. Kaindl, M. A. Carnahan, J. Orenstein, and D. S. Chemla, BFar-infrared optical conductivity gap in superconducting MgB2 films,[ Phys. Rev. Lett., vol. 88, pp. 027 003-1–027 003-3, Jan. 2002. [19] R. Huber, B. A. Schmid, Y. R. Shen, D. S. Chemla, and R. A. Kaindl, BStimulated

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33] [34]

[35]

terahertz emission from intraexcitonic transitions in Cu2 O,[ Phys. Rev. Lett., vol. 96, pp. 017 402-1–017 402-4, Jan. 2006. B. B. Hu and M. C. Nuss, BImaging with terahertz waves,[ Opt. Lett., vol. 20, no. 16, pp. 1716–1718, Aug. 1995. Z. Jiang, X. G. Xu, and X.-C. Zhang, BImprovement of terahertz imaging with a dynamic subtraction technique,[ Appl. Opt., vol. 39, pp. 2982–2987, Jun. 2000. M. Brucherseifer, M. Nagel, P. H. Bolivar, H. Kurz, A. Bosserhoff, and R. Bu¨ttner, BLabel-free probing of the binding state of DNA by time-domain terahertz sensing,[ Appl. Phys. Lett., vol. 77, pp. 4049–4052, Dec. 2000. Z. Jiang, M. Li, and X.-C. Zhang, BDielectric constant measurement of thin films by differential time-domain spectroscopy,[ Appl. Phys. Lett., vol. 76, pp. 3221–3223, May 2000. B. Ferguson, S. Wang, D. Gray, D. Abbott, and X.-C. Zhang, BT-ray computed tomography,[ Opt. Lett., vol. 27, pp. 1312–1314, Aug. 2002. C. Ku¨bler, R. Huber, S. Tu¨bel, and A. Leitenstorfer, BUltrabroadband detection of multi-terahertz field transients with GaSe electro-optic sensors: Approaching the near infrared,[ Appl. Phys. Lett., vol. 85, pp. 3360–3362, Oct. 2004. R. Huber, F. Tauser, A. Brodschelm, M. Bichler, G. Abstreiter, and A. Leitenstorfer, BHow many-particle ineractions develop after ultrafast exciaion of an electron–hole plasma,[ Nature, vol. 414, pp. 286–289, 2001. Y. C. Shen, P. C. Upadhya, E. H. Linfield, H. E. Beere, and A. G. Davies, BUltrabroadband terahertz radiation from low-temperature-grown GaAs photoconductive emitters,[ Appl. Phys. Lett., vol. 83, pp. 3117–3119, Oct. 2003. Y. C. Shen, P. C. Upadhya, H. E. Beere, E. H. Linfield, A. G. Davies, I. S. Gregory, C. Baker, W. R. Tribe, and M. J. Evans, BGeneration and detection of ultrabroadband terahertz radiation using photoconductive emitters and receivers,[ Appl. Phys. Lett., vol. 85, pp. 164–166, Jul. 2004. S. Kono, M. Tani, P. Gu, and K. Sakai, BDetection of up to 20 THz with a low-temperature-grown GaAs photoconductive antenna gated with 15 fs light pulses,[ Appl. Phys. Lett., vol. 77, pp. 4104–4106, Dec. 2000. S. Kono, M. Tani, and K. Sakai, BUltrabroadband photoconductive detection: Comparison with free-space electro-optic sampling,[ Appl. Phys. Lett., vol. 79, pp. 898–900, Oct. 2001. S. Kono, M. Tani, and K. Sakai, BCoherent detection of midinfrared radiation up to 60 THz with an LT-GaAs photoconductive antenna,[ IEE Proc.VOptoelectron., vol. 149, pp. 105–109, 2002. H. Shimosato, S. Saito, M. Ashida, T. Itoh, and K. Sakai, BUltrabroadband detection of terahertz radiation from 0.1 to 100 THz with photoconductive antenna,[ Ultrafast Optics V (Springer Series in Optical Sciences), pp. 317–323, Jul. 2007. D. Mittleman, Sensing With Terahertz Radiation. Berlin, Germany: Springer, 2005. D. L. Woolard, E. R. Brown, M. Pepper, and M. Kemp, BTerahertz frequency sensing and imaging: A time of reckoning future applications?[ Proc. IEEE, vol. 93, no. 10, pp. 1722–1743, Oct. 2005. A. W. M. Lee, B. S. Williams, S. Kumar, Q. Hu, and J. L. Reno, BReal-time imaging using a 4.3-THz quantum cascade laser and

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43] [44]

[45]

[46]

[47]

[48]

[49]

[50] [51]

[52]

[53]

[54]

[55]

a 320  240 microbolometer focal-plane array,[ IEEE Photon. Technol. Lett., vol. 18, no. 13, pp. 1415–1417, Jul. 2006. K. Kawase, Y. Ogawa, Y. Watanabe, and H. Inoue, BNon-destructive terahertz imaging of illicit drugs using spectral fingerprints,[ Opt. Express, vol. 11, pp. 2549–2554, 2003. Y. C. Shen, T. Lo, P. F. Tady, B. E. Cole, W. R. Tribe, and M. Kemp, BDetection and identification of explosives using terahertz pulsed spectroscopic imaging,[ Appl. Phys. Lett., vol. 86, no. 241 116, 2005. J. F. Federici, B. Schulkin, F. Huang, D. Gary, R. Barat, F. Oliveira, and D. Zimdars, BTHz imaging and sensing for security applications, explosives, weapons and drugs,[ Semicond. Sci. Technol., vol. 20, pp. 5266–5280, 2005. G. P. Gallerano. (2004). THz-RIDGE Final Report. [Online]. Available: http://www. frascati.enea.it/THz-BRIDGE/ T. Globus, M. Bykhovskaia, D. Woolard, and B. Gelmont, BSub-millimetre wave absorption spectra of artificial RNA molecules,[ J. Phys. D., vol. 36, pp. 1314–1322, 2003. M. R. Derrick, D. Stulik, and J. M. Landry, Infrared Spectroscopy in Conservation Science. Los Angeles: The Getty Conservation Institute, 1999. IRUG spectral database edition 2000 search engine. [Online]. Available: http://www.irug. org/ed2k/search.asp THz spectral database. [Online]. Available: http://webbook.nist.gov/chemistry/thz-ir/ THz image gallery. [Online]. Available: http://www.thznetwork.org/wordpress/index. php/thz-images/ THz database. [Online]. Available: http:// www.riken.go.jp/lab-www/tera/TP_HP/ RIKENDatabase/index.html K. Fukunaga, Y. Ogawa, S. Hayashi, and I. Hosako, BTerahertz spectroscopy for art conservation,[ IEICE Electron. Express, 2007. A. Aldrovandi, M. L. Altamura, M. T. Cianfanelli, and P. Riitano, BPictorial materials: The creation of a sample charge for the characterization of materials by means of multispectral analysis,[ OPD Restauro, vol. 8, pp. 191–210, 1996. H. Pickett, R. Poynter, E. Cohen, M. Delitsky, J. Pearson, and H. Mueller, BSubmillimeter, millimeter, and microwave spectral line catalog,[ J. Quant. Spectrosc. Radiat. Transf., vol. 60, pp. 883–890, 1998. H. J. Liebe, BMPMVAn atmospheric millimeter-wave propagation model,[ Int. J. Infrared Millim. Waves, vol. 10, pp. 631–650, 1989. ITU-R. [Online]. Available: http://www.itu. int/ITU-R/ L. S. Rothman et al., BThe HITRAN 2004 molecular spectroscopic database,[ J. Quant. Spectrosc. Radiat. Transf., vol. 96, pp. 139–204, 2005. S. A. Clough, F. X. Kneizys, and R. W. Davies, BLine shape and the water vapor continuum,[ Atmos. Res., vol. 23, pp. 229–241, 1989. S. A. Clough, F. X. Kneizys, L. S. Rothman, and W. O. Gallery, BAtmospheric spectral transmittance and radiance: FASCOD1B,[ Proc. Soc. Photo. Opt. Instrum. Eng., vol. 277, pp. 152–166, 1981. S. A. Clough, M. J. Iacono, and J.-L. Moncet, BLine-by-line calculation of atmospheric fluxes and cooling rates: Application to water vapor,[ J. Geophys. Res., vol. 97, pp. 15 761–15 785, 1992. LBLRTM. [Online]. Available: http://rtweb. aer.com/lblrtm.html

Vol. 95, No. 8, August 2007 | Proceedings of the IEEE

1621

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

[56] Y. Kasai, C. Takahashi, S. Ochiai, P. Baron, J. Urban, T. Motoki, Y. Irimajiri, and A. Kleinboehl, BSMOCO: A retrieval code for Super-conductive Sub-MIllimeter Limb Emission Sounder (SMILES) planned for the International Space Station,[ J. Quant. Spectrosc. Radiat. Transf., in preparation. [57] J. Urban, P. Baron, N. Lautie´, N. Schneider, K. Dassas, P. Ricaud, and J. de La Noe¨, BMoliere (v5): A versatile forward- and inversion model for the millimeter and sub-millimeter wavelength range,[ J. Quant. Spectrosc. Radiat. Transf., vol. 83, pp. 529–554, 2004. [58] J. Mendrok, BThe SARTre model for radiative transfer in spherical atmospheres and its application to the derivation of cirrus cloud properties,[ Ph.D. dissertation, Freie Universita¨t, Berlin, Germany, 2006. [59] Y. J. Kasai, J. Urban, C. Takahashi, S. Hoshino, K. Takahashi, J. Inatani, M. Shiotani, and H. Masuko, BStratospheric ozone isotope enrichment studied by submillimeter wave heterodyne radiometry: The observation capabilities of SMILES,[ IEEE Trans. Geosci. Remote Sens., vol. 44, no. 3, pp. 676–693, Mar. 2006. [60] D. P. Murtagh, U. Frisk, F. Merino, M. Ridal, A. Jonsson, J. Stegman, G. Witt, P. Eriksson, C. Jime´nez, G. Me´gie, J. de La Noe¨, P. Ricaud, P. Baron, J. R. Pardo, A. Hauchecorne, E. J. Llewellyn, D. A. Degenstein, R. L. Gattinger, N. D. Lloyd, W. F. J. Evans, I. C. McDade, C. S. Haley, C. Sioris, C. von Savigny, B. H. Solheim, J. C. McConnell, K. Strong, E. H. Richardson, G. W. Leppelmeier, E. Kyro¨la¨, H. Auvinen, and L. Oikarinen, BAn overview of the Odin atmospheric mission,[ Can. J. Phys., vol. 80, pp. 309–318, 2002. [61] J. de La Noe¨, O. Lezeaux, G. Guillemin, R. Lauque´, P. Baron, and P. Ricaud, BA ground-based microwave radiometer dedicated to stratospheric ozone monitoring,[ J. Geophys. Res., vol. 103, pp. 22 147–22 161, 1998. [62] H. Fischer and H. Oelhaf, BRemote sensing of vertical profiles of

[63]

[64]

[65]

[66]

[67]

[68]

[69]

[70]

[71]

[72]

atmospheric trace constituents with MIPAS limb-emission spectrometers,[ Appl. Opt., vol. 35, pp. 2787–2796, 1996. J. Mendrok, F. Schreier, and M. Ho¨pfner, BEstimating cirrus cloud properties from MIPAS data,[ Geophys. Res. Lett. C. Rodgers, Inverse Methods for Atmospheric Sounding: Theory and Practise. Singapore: World Scientific, 2000. P. Yang, H. Wei, H.-L. Huang, B. A. Baum, Y. X. Hu, G. W. Kattawar, M. I. Mishchenko, and Q. Fu, BScattering and absorption property database for nonspherical ice particles in the near-through far-infrared spectral region,[ Appl. Opt., vol. 44, pp. 5512–5523, 2005. M. I. Mishchenko, L. D. Travis, and A. A. Lacis, Scattering, Absorption, and Emission of Light by Small Particles. Cambridge, U.K.: Cambridge Univ. Press, 2002. K. Stamnes, S.-C. Tsay, W. Wiscombe, and K. Jayaweera, BNumerically stable algorithm for discrete-ordinate-method radiative transfer in multiple scattering and emitting layered media,[ Appl. Opt., vol. 27, pp. 2502–2509, 1988. S. D. Gasster, C. H. Townes, D. Goorvitch, and F. P. J. Valeo, BForeign-gas collision broadening of the far-infrared spectrum of water vapor,[ J. Opt. Soc. Amer. B, vol. 5, pp. 593–601, 1988. V. B. Podobedov, D. F. Plusquellic, and G. T. Fraser, BTHz laser study of self-pressure and temperature broadening and shifts of water vapor lines for pressures up to 1.4 kPa,[ J. Quant. Spectrosc. Radiat. Transf., vol. 87, pp. 377–385, 2004. G. R. Davis, BThe far infrared continuum absorption of water vapor,[ J. Quant. Spectrosc. Radiat. Transf., vol. 50, pp. 673–694, 1993. P. W. Rozenkranz, BWater vapor microwave continuum absorption: A comparison of measurements and models,[ Radio. Sci., vol. 33, pp. 919–928, 1998. H. Harde, R. A. Cheville, and D. Grischkowsky, BTerahertz studies

[73]

[74]

[75]

[76]

[77]

[78]

[79]

[80]

of collision-broadened rotational lines,[ J. Phys. Chem. A, vol. 101, pp. 3646–3660, 1997. D. E. Burch and D. A. Gryvnak, Atmospheric Water Vapor, A. Deepak, T. D. Wilderson, and L. H. Ruhnke, Eds. New York: Academic, 1980. Q. Ma and R. H. Tipping, BThe averaged density matrix in the coordinate representation: Application to the calculation of the far-wing line shapes for H2 O,[ J. Chem. Phys., vol. 111, pp. 5909–5921, 1999. A. I. Meshkov and F. C. De Lucia, BBroadband absolute absorption measurements of atmospheric continua with millimeter wave cavity ringdown spectroscopy,[ Rev. Sci. Instrum., vol. 76, no. 083 103, 2005. V. B. Podobedov, D. F. Plusquellic, and G. T. Fraser, BInvestigation of the water-vapor continuum in the THz region using a multipass cell,[ J. Quant. Spectrosc. Radiat. Transf., vol. 91, pp. 287–295, 2005. D. Grischkowsky, S. Keiding, M. van Exter, and C. Fattinger, BFar-infrared time-domain spectroscopy with terahertz beams of dielectrics and semiconductors,[ J. Opt. Soc. Amer. B, vol. 7, p. 2006, 1990. R. A. Cheville and D. Grischkowsky, BFar-infrared foreign and self-broadened rotational linewidths of high-temperature water vapor,[ J. Opt. Soc. Amer. B, vol. 16, pp. 317–322, 1999. T. Seta, H. Hoshina, Y. Kasai, I. Hosako, C. Otani, S. LoQow, J. Urban, M. Ekstro¨m, P. Eriksson, and D. Murtagh, BPressure broadening coefficients of the water vapor lines at 556.936 and 752.033 GHz,[ J. Quant. Spectrosc. Radiat. Transf., submitted for publication. P. Y. Han, M. Tani, M. Usami, S. Kono, R. Kersting, and X. C. Zhang, BA direct comparison between terahertz time-domain spectroscopy and far-infrared Fourier transform spectroscopy,[ J. Appl. Phys., vol. 89, pp. 2357–2359, 2001.

ABOUT THE AUTHORS Iwao Hosako received the B.S., M.S., and Ph.D. degrees from the University of Tokyo, Japan, in 1988, 1990, and 1993, respectively. After two years with NKK Corp’s ULSI Laboratory from 1993 to 1994, he joined Communications Research Laboratory (former name of NICT). He is currently a Project Leader (research Manager) of the Terahertz Project at the National Institute of Information and Communications (NICT), Tokyo. His research during 1995–1998 focused on cryogenic readout circuits with emphasis on ultralow 1/f-noise transistors for a far-infrared detector system. His research interests include nearly all aspects of terahertz-device technologies such as semiconductor emitters, detectors, and optical thin films. He is serving on several research committees for terahertz technologies.

1622

Proceedings of the IEEE | Vol. 95, No. 8, August 2007

Norihiko Sekine (Member, IEEE) received the B.S., M.S., and Ph.D. degrees in electronic engineering from the University of Tokyo, Japan in 1994, 1996, and 1999, respectively. After experience in industry, he became a researcher at the National Institute of Information and Communications Technology (NICT), Tokyo. His research interests include the physical properties of semiconductor nanostructures in the terahertz regime and their application to terahertz devices.

Hosako et al.: At the Dawn of a New Era in Terahertz Technology

Mikhail Patrashin received the Ph.D. degree in semiconductor device physics from the Institute of Radio-Engineering and Electronics, Russia, in 1995. He is a member of the research staff at the National Institute of Information and Communications Technology (NICT), Tokyo, Japan. He is currently working on the development of semiconductor devices and related technologies for various applications in the terahertz range.

Shingo Saito received the B.S., M.S., and Ph.D. degrees from Tohoku University, Sendai, Japan, in 1990, 1992, and 1995, respectively. From 1995 to 2002, he was a Research Associate at Institute for Solid State Physics at the University of Tokyo, Japan. He is currently a Senior Researcher at the National Institute of Information and Communications Technology, Kobe, Japan, where he currently is engaged in research on solid-state spectroscopy, especially THz wave region, using ultrashort pulse lasers for information and communication technology. Dr. Saito is a member of the Physical Society of Japan and the Japan Society of Applied Physics.

Kaori Fukunaga (Member, IEEE) received the Ph.D. degree in electrical engineering from Tokyo Denki University, Japan, in 1993. She worked on high-voltage insulation systems at Fujikura, Ltd from 1989 to 1993. She joined the National Institute of Information and Communications Technology (NICT), Tokyo, in 1995, and is a Senior Researcher at the Applied Electromagnetic Research Centre. She has been involved in material analyses using electromagnetic and electroacoustic techniques. Dr. Fukunaga is a member of the Institute of Physics, IEE of Japan, the Japan Institute of Electronics Packaging, and the Japan Society for the Conservation of Cultural Property.

Yasuko Kasai received the Ph.D. degree from the Chemistry Department of Tokyo Institute of Technology, Japan, in 1995 with microwave molecular spectroscopy and radio astronomical observations. She is a Senior Researcher in Environmental Sensing and Network Group of Applied Electromagnetic Research Center of National Institute of Information and Communications Technology (NICT), Tokyo. Her scientific research interests are on satellite observation of the Earth’s atmosphere in the frequency region from the millimeter-wave to infrared.

Philippe Baron received the Ph.D. degree in physics from Bordeaux University, France, in 1999. He is working at the National Institute of Information and Communications Technology, Tokyo, Japan, with a JSPS fellowship. His research subjects are radiative transfer modeling and retrieval, as well as atmospheric chemistry and dynamics.

Takamasa Seta received the Ph.D. degree from the University of Tokyo, Japan, in 2006, with work in chemical reactions of environmental interests. He has been a Postdoctoral Fellow at the National Institute of Information and Communications Technology, Tokyo, since 2006. His research interests are laboratory spectroscopic measurements for the satellite observations and the terahertz-wave propagation model.

Jana Mendrok received the Dipl.Ing. degree in geodesy from Technical University of Dresden, Germany, in 2002 and the Ph.D. degree in natural sciences from Free University Berlin, Germany in 2006. She was with the Remote Sensing Technology Institute of the German Aerospace Center from 2002 to 2006. Currently, she is a Postdoctoral Researcher with the Applied Electromagnetic Research Center, National Institute of Information and Communications Technology, Tokyo, Japan. Her research interests are in radiative transfer modeling and passive remote sensing with focus on the influence of clouds on trace gas retrievals and cloud property retrievals from infrared to microwave measurements.

Satoshi Ochiai received the B.E. and M.E. degrees in chemical engineering from Kyoto University, Japan, in 1986 and 1988, respectively. In 1988, he joined the Communications Research Laboratory, which was reorganized into the National Institute of Information and Communications Technology (NICT), Tokyo, Japan, in 2004. Thereafter, he has been engaged in research on millimeter- and submillimeter-wave remote sensing.

Hiroaki Yasuda received the B.E. and M.E. degrees in applied physics all from the University of Tokyo, Japan, in 1989 and 1991, respectively. After six years with Toshiba Corp’s ULSI Laboratory from 1991 to 1996, he joined the Communications Research Laboratory (former name of NICT). He is a Senior Researcher at the National Institute of Information and Communications Technology (NICT), Tokyo. His research interests include the physical properties of semiconductor nanostructures in the millimeter to terahertz regime and their application to high-frequency devices.

Vol. 95, No. 8, August 2007 | Proceedings of the IEEE

1623