This effort includes synergistic work crucial to the development of future terahertz (THz, 1012 Hz) nano-size circuits, employing Metallic Single-Wall Carbon Nanotubes (SWCNTs) ¡V unique 1D nanoscale materials: I) We will enhance the perfomance we have demonstrated for SWCNTs as direct detectors at THz frequencies and demonstrate these as novel heterodyne detectors that can constitute sensitive elements for THz focal plane arrays (FPAs) that require only moderate cooling (?î 100 K); II) Development of compact THz sources in which heated CNTs produce broadband THz radiation; III) Integration of the detectors and sources into THz integrated circuits. We will further deepen our understanding of CNT devices at THz through comprehensive time-dependent quantum simulations that characterize plasmonic resonances, AC transport and contact coupling in a variety of CNT configurations
INTELLECTUAL MERIT. Unique measurements will test theories of how electromagnetic waves interact with and couple to electrons in CNTs at THz frequencies, including slowly propagating plasmons and resonances due to such plasmons, while also obtaining values for small bandgaps. The time-dependent quantum simulations will utilize several unique approaches and a novel eigenvalue solver (¡§FEAST¡¨) developed by the Co-PI.
BROADER IMPACT. The results will provide the community a broad range of new understanding of the basic properties of 1-D conductors and their contacts at very high frequencies, while involving students at all levels. The work can potentially benefit society at large, especially in terms of enabling very compact THz cancer imagers, imagers for security applications, and a new paradigm for designing THz integrated circuits.
Terahertz waves are electromagnetic waves with frequencies about three orders-of-magnitude higher than microwaves which are, for example, employed in cell phone technology. That type of communication applications can not be realized at terahertz (THz) frequencies at present, however. Terahertz sources such as lasers exist but are extremely bulky and expensive, dissipate a lot of power, and/or have to be operated at cryogenic temperatures. Nanotechnology using materials such as carbon nanotubes is a new field that shows promise for realizing compact THz electronic devices. On this grant we have developed a simple, ultra-compact THz source that consists of a network of carbon nanotubes powered by a simple low voltage DC power source. Carbon nanotubes have typical diameters of 1 to 2 nanometers. The carbon nanotubes in our devices are deposited as a network on a small THz antenna that we fabricate with microelectronic lithography techniques on a thin substrate of silicon, the same material that is used in most integrated circuits for lower frequencies (Image 1). The THz radiation is collected by a small silicon lens that re-radiates it in a concentrated beam. The device is packaged in a cubic metal block that is about 2 cm on each side, see Image 2. The silicon lens is in the center of the block. The THz radiation is not coherent like that of a laser, but incoherent more like that of a flash light. While the output power is at most a millionth of a watt it is sufficient for it to act as a THz source for short distances. The frequency range radiated can be varied by changing the dimensions of the antenna. The actual active element, the carbon nanotube network, is only a few micrometers in size. This makes it possible to further miniaturize the CNT THz source. We have developed a THz integrated circuit (TIC), also fabricated on a silicon substrate, schematically shown in Image 3. The length of this circuit is only about 100 micrometers (0.004"). The CNTs are deposited to the left in that image and are powered by the DC source shown at the top. The THz waves propagate along the wire (we use microstrips or coplanar waveguides; gold color in Image 3), past the "sample". The THz waves are then radiated from an on-chip antenna, through a silicon lens on the opposite side of the substrate. We have demonstrated how we can record THz spectra using this TIC by sprinkling powders of different materials on top of it. The spectra were recorded in a Fourier Transform Spectrometer ("FTS"), a standard instrument for THz use. The TIC used as a spectrometer has identified minute quantities of materials through their THz spectral "finger print". As an example, Image 4 shows the vibrational spectrum of the polymer polyhydoxybutyrate (PHB). Future extensions of this concept include (1) integrating the CNTs in a standard silicon CMOS chip and employing the CMOS transistors, realizing a THz "spectrometer on a chip", (2) integrating nano/micro-fluidic channels in the silicon chip just below the TIC and measuring spectra of liquids flowing through such channels and (3) integrating the CNT sources in silicon chips to develop THz cameras with built-in illumination for medical imaging in order to distinguish cancer tissue from normal tissue. All these applications are basically possible because of the very small size and easy integrability of the CNT source. We have submitted an invention disclosure for the TIC spectrometer concept and UMass/Amherst is applying for a patent. Thus, there is potential for commercial development of the different versions of the TIC. Furthermore, much fundamental knowledge has been gained during the grant period regarding how CNTs behave as they are powered by the DC source and radiate THz. Numerical modeling and simulations have become essential to supplement the experimental research, and provide fundamental insights for understanding the dynamics of electrons in novel nanoscale carbon-based materials. In this project, we have developed a fundamental and state-of-the-art 3D real-space and real-time numerical model using the time-dependent density functional theory (TDFFT). These TDDFT simulations have been used by the PIs to study plasmon phenomena in 1D Carbon-based nanomaterials (CNTs, narrow nanoribbons) and their capacity to operate at THz frequencies. In this work we model all the individual atoms of the CNT (110 carbon atoms; see image 5), as well as all relevant electrons and their interactions, avoiding the approximations used in previous simulations. The simulation methods developed will in the future benefit many other areas such as nanoscience in general and molecular chemistry. The Co-PI maintains a public domain web site devoted to the FEAST eigen-value solver v2.1, (Feb. 2013), www.feast-solver.org. This solver is one of the crucial components that allows us to perform the very large (computationally) simulation problems described above.
