Selected NIST/EEEL Accomplishments in Nanoscale Science, Engineering, and Technology in 2003
NIST/EEEL researchers demonstrate a new type of single-electron charge source
In collaboration with workers at NTT in Japan, NIST/EEEL researchers have demonstrated operation of a silicon-based single-electron turnstile. This device, along with similar classes of devices (pumps and CCDs) which we are also pursuing, offers the possibility of developing a current standard based on the charge of the electron.
Single-electron tunneling devices based on the Coulomb blockade have been used as charge sources for a capacitance standard at NIST. We also desire a direct current standard, but the amount of current available from the pumps is too small to be useful. One possible alternative is Si-based SET devices, for two reasons: one is that the capacitance is much smaller in these devices, affording the ability to run them at much higher frequency. The second is the lack of the charge offset drift allows the potential to parallelize many devices together, thus increasing the total current by a large factor.
The advance over previous work is that the turnstiles are easier to operate than the Si-based pumps, and also because at least one of the error mechanisms is easier to analyze in the turnstiles. Although detailed error measurements have not yet been made, at a rough level we have shown flatness of the plateau and an absolute value which is correct within about 1 percent, I = ef.
Yukinori Ono, Neil N. Zimmerman, Kenji Yamazaki, and Yasuo Takahashi, to be published in Japanese Journal of Applied Physics.
Contact: Neil Zimmerman, neil.zimmerman@nist.gov, 301-975-5887
Second NIST Nanomagnetodynamics Workshop
On June 26, 2003, NIST held its second Nanomagnetodynamics Workshop in Monterey, California, in conjunction with the annual meeting of the Information Storage Industry Consortium (INSIC). INSIC is one of the industry's main forums for pre-competitive research. The program featured research reports from seven universities, all sponsored under the NIST Nanotechnology Initiative, along with research highlights from two NIST Laboratories.
The presentations were in programmatic areas of immediate interest to the magnetic data storage industry: (1) new composites and nanocrystalline alloys with high saturation magnetization and high resistivity; (2) rare-earth doping of nickel-iron alloys to control magnetic precession and damping; (3) magnetization reversal in high coercivity, nanoclustered and multilayered films; (4) nonlinear spin-wave dynamics, ferromagnetic resonance, and mechanisms of magnetization reversal; (5) dynamic excitations in magnetic nanostructures; and (6) ultrasensitive magnetic measurements.
The objective of the NIST nanomagnetodynamics program is to understand and control magnetic switching and damping, especially in small structures, with the aim of increasing data transfer rates during reading and writing. Target bandwidths are greater than 1 gigahertz, with switching times below 1 nanosecond.
Beach, G.S.D, Silva, T.J., Parker, F.T., Berkowitz, A.E., "High frequency characteristics of metal/native oxide multilayers," IEEE Trans. Magn., 39, 2669 (September 2003).
Contact: Ron Goldfarb, goldfarb@boulder.nist.gov, 303-497-3650
 |
 |
High-Frequency Noise Used by NIST to Characterize Commercial Recording Heads
In collaboration with two U.S. companies involved in the manufacture of disk drives and recording heads, NIST/EEEL completed a study of high-frequency magnetic noise in commercial magnetoresistive recording heads. The magnetic noise in the read sensors was measured over a frequency range of 0.1 to 8 gigahertz. Magnetic noise is due to thermal fluctuations of the magnetic layers used to sense the fields from data bits. These fluctuations scale inversely with the sensor size; as the sensor dimensions shrink below 100 nanometers, they can become the dominant source of noise. The peak magnetic noise level in 200 nanometer devices in modern disk drives is 0.2 nanovolts per root hertz, which is on the same order as the Johnson (thermal electron) noise.
The noise was measured as a function of both current through the sensor and magnetic field applied in a direction to simulate the bit fields. The noise spectra show several resonant peaks. The largest, near 5 gigahertz, is due to the uniform rotation of the magnetization near the center of the devices. Additional peaks are observed due to magnetic fluctuations at the ends of the devices, which are more strongly pinned by abutted permanent-magnet thin films. Other, higher-frequency peaks may be due to nonuniform modes in the center of the device. A low-frequency noise component, indicative of non-ideal magnetization structure and domain-wall motion, varied considerably from head to head. The strength of the peaks and, to a lesser extent, the frequency positions of the peaks, also varied from head to head. When the noise spectra are fully understood they will provide a diagnostic of the magnetic structure of the sensors.
Russek, S.E., Kaka, S.F., McMichael, R.D., Donahue, M.J., "High-speed switching and rotational dynamics in small magnetic thin-film devices," in Spin Dynamics II, Spinger-Verlag, pp. 93-154 (January 2003).
Stutzke, N.A., Burkett, S.L., Russek, S.E., "Temperature and field dependence of high-frequency magnetic noise in spin valve devices," Appl. Phys. Lett., 82, 91 (January 2003).
Kaka, S.F., Nibarger, J.P., Russek, S.E., Stutzke, N.A., Burkett, S.L., "High-frequency measurements of spin valve films and devices," J. Appl. Phys., 93, 7539 (May 2003).
Stutzke, N.A., Burkett, S.L., Russek, S.E., "High-frequency noise measurements in spin-valve devices," J. Vacuum Sci. Technol. B, 21, 1167 (July 2003).
Contact: Steve Russek, russek@boulder.nist.gov, 303-497-5097
Cryogenic Capability Added to NIST Pulsed Inductive Microwave Magnetometer
NIST/EEEL's recently developed pulsed inductive microwave magnetometer (PIMM) has been enhanced with variable temperature capability. The new instrument can measure the magnetodynamic response of magnetically soft, thin-film materials at temperatures from 25 to 325 kelvins. In addition, the CryoPIMM has been augmented with high-field magnets that can apply dc bias fields as large as 45 millitesla, permitting the study of materials with high anisotropy, such as single-crystal films of iron and nickel.
The CryoPIMM is a powerful new tool to investigate the fundamental origins of precessional damping in thin metallic films. The origin of the oscillatory response stems from the gyromagnetic properties inherent in all ferromagnets. When a torque is applied to the magnetization, the intrinsic response of the electron moment is precession, similar to how a gyroscope precesses under the influence of the Earth's gravitational field. However, the angular momentum of the electron spin precesses at megahertz to gigahertz frequencies.
Since the spins are coupled to the atomic lattice, the precession is eventually damped. Nevertheless, the resulting oscillations of the magnetic moment can be deleterious in practical applications, such as magnetic data storage. For example, the data transfer rate in commercial disk drives is now approaching 1 gigabit per second. Disk drive engineers must be careful to avoid effects stemming from gyromagnetic precession at these frequencies. Most importantly, there is a need to determine sources of damping, with the goal of controlling the damping as a materials design parameter. There are multiple conflicting theories for damping in metallic thin films. Observation of a temperature dependence in the damping would be confirming evidence for some of the theories.
Kos, A.B., Nibarger, J.P., Lopusnik, R., Silva, T.J., Celinski, Z.J., "Cyrogenic pulsed inductive microwave magnetometer," J. Appl. Phys., 93, 7068 (May 2003).
Nibarger, J.P., Lopusnik, R., Silva, T.J., "Damping as a function of pulsed field amplitude and bias field in thin film Permalloy," Appl. Phys. Lett., 82, 2112 (Mar 2003).
Nibarger, J.P., Lopusnik, R., Silva, T.J., Celinski, Z.J., "Variation of the effective magnetization and the Landé g factor with thickness in thin Permalloy films," Appl. Phys. Lett., 83, 93 (July 2003).
Lopusnik, R., Nibarger, J.P., Silva, T.J., Celinski, Z.J., "Different dynamic and static magnetic anisotropy in thin Permalloy films," Appl. Phys. Lett., 83, 96 (July 2003).
Contacts: Tom Silva, silva@boulder.nist.gov, 303-497-7826
Tony Kos, kos@boulder.nist.gov, 303-497-5333
Spin-Momentum Transfer Is Ubiquitous in Multilayer Magnetic Films
To enable future applications in spin electronics, such as ultra-high-frequency oscillators, NIST is working to measure coherent spin dynamics in metal/semiconductor heterostructures. Spin-momentum transfer (SMT) is being investigated using mechanical point-contact spectroscopy to detect current-induced excitations in multilayer films.
It is known that, for sufficiently high current densities and applied magnetic fields, there is an abrupt increase in the resistance of a point-contact junction. The resistance step is attributed to the generation of magnons (spin waves) by the spin-momentum transfer effect. The NIST team has found that SMT is a generic effect, occurring for a wide range of experimental conditions: for both in-plane and out-of-plane fields, for multilayers grown at the both the first and second maxima in giant magnetoresistance (GMR), and for ferromagnetically coupled multilayers.
NIST has observed SMT in a number of different and previously unexplored alloys of Co, Fe and Ni in multilayer structures: Cu/CoFe, Cu/Fe and Cu/NiFe. These results indicate that SMT, which gives rise to resistance steps in point contact measurements, is a general property of magnetic multilayers and is not specific to Cu/Co multilayers. The multilayer films that exhibit SMT features need not even exhibit a measurable GMR signal: no GMR was measured for the Cu/Fe multilayers.
An estimate of the spin momentum transfer efficiency from a polarized conduction current was obtained from point-contact data for Cu/Co multilayers. The analysis uses the theory of IBM researcher John Slonczewski, who first predicted the SMT effect in 1996. From this theory, the critical current at which the point contact resistance experiences a sudden jump can be used to determine the SMT efficiency in an experimental geometry with an applied field perpendicular to the magnetic multilayer. Estimates of SMT efficiency for point-contact data give values from 25 to 35 percent, close to the maximum expected values calculated by Slonczewski for Co-based multilayers.
Rippard, W., Pufall, M., Silva, T.J., "Quantitative studies of spin-momentum-transfer-induced excitations in Co/Cu multilayer films using point-contact spectroscopy," Appl. Phys. Lett., 82, 1260 (February 2003).
Pufall, M., Rippard, W., Silva, T.J., "Materials dependence of the spin-momentum transfer efficiency and critical current in ferromagnetic metal/Cu multilayers," Appl. Phys. Lett., 83, 323 (July 2003).
Contact: Tom Silva, silva@boulder.nist.gov, 303-497-7826
NIST Ultra-Sensitive Microcantilevers Measure Magnetic Films with Sub-Atomic-Layer Resolution
Work at NIST/EEEL to develop ultra-sensitive torque magnetometers has lead to quantitative measurements of micrometer-scale magnetic dots. Measurements are based on the detection of mechanical torques on thin films deposited onto microcantilevers. A main challenge of these techniques is getting well-defined micromagnetic samples onto the cantilevers. To this end, NIST scientists developed a wafer-level microfabrication process in which the film deposition and patterning are combined with the cantilever micromachining process. This allows magnetic measurements of samples with a total magnetic moment smaller than that detectable with conventional magnetometers.
Cantilevers with low spring constants and high mechanical quality factors are essential for these measurements. The NIST cantilevers are double torsional oscillators made from single-crystal silicon with resonant frequencies of 120 kilohertz and mechanical quality factors of 12,000 or more. In the dynamic deflection method, the cantilever and magnetic film are placed in an external magnetic field. A small orthogonal ac torque field is applied at the cantilever resonant frequency, and the resulting torque is measured as a function of external field. A Ni-Fe square, 5 micrometers x 5 micrometers x 30 nanometers thick, had a measured magnetic moment of 0.51 picojoule per tesla. The cantilevers have a torque resolution of 0.84 attonewton-meter, corresponding to a magnetic moment noise level of 6.7 femtojoules per tesla. Thus, magnetic moments can be measured with a resolution below that of a complete atomic layer of Ni-Fe. With improvements in microfabrication, the target sensitivity is 100,000 Bohr magnetons (1 attojoule per tesla) at room temperature.
Micrometer and sub-micrometer scale magnetic measurements have proven to be a challenge for conventional magnetometers. Conventional measurements are made on arrays of micromagnetic dots. However, due to fabrication limitations, the results are clouded by statistical variations in dot shape, size, and spacing. Thus, more sensitive detectors are needed that can measure magnetic properties on individual dots. One way to understand spin damping in ferromagnetic systems is to investigate size effects as magnetic devices are reduced to sub-micrometer dimensions. Studies of magnetic nanodots will give a better understanding of spin damping and aid in the development of faster disk drives.
Moreland, J.M., "Micromechanical instruments for ferromagnetic measurements," J. Phys. D, 36, 39 (January 2003); Chabot, M.D., Moreland, J.M., "Micrometer-scale magnetometry of thin Ni80Fe20 films using ultra-sensitive microcantilevers," J. Appl. Phys. 93, 7897 (May 2003).
Contact: John Moreland, moreland@boulder.nist.gov, 303-497-3641
NIST Advances Quantum Dot Growth and Characterization to Demonstrate Single Photon Turnstile
NIST is developing single photon emitters using semiconductor quantum dots (QDs) for precision radiometry and quantum cryptography applications. They have demonstrated photon anti-bunching in the emission from an optically pumped single InGaAs QD, as indicated by the histogram of start-stop times from photon counting detectors in the two arms of an interferometer. The small peak area at zero time delay compared to that at non-zero delays was the signature that photons were being emitted one by one. Single photon emission has been observed up to 120 K, which represents a record high temperature for such a demonstration in InGaAs QDs. To produce such photon turnstiles requires control, characterization and modeling of QD structural, electrical and optical properties. Towards these goals, NIST researchers determined how InAs epilayers evolve during temperature quenching after deposition of the initial layer. The results may have implications on the accuracy of measurements of the shapes and sizes of QDs formed at growth temperature. NIST has also developed techniques for electrical and optical characterization of QDs. Using photoluminescence and capacitance spectroscopies, NIST researchers identified a bimodal distribution of InGaAs QDs produced under specific growth conditions. Further measurements with AFM and TEM microscopies indicated that the shapes of the large and small QDs were different, as shown in the photo. Combining these measurements with theoretical calculations and modeling, the researchers constructed a consistent picture for the energy level diagrams of both kinds of QDs. NIST experience with ensembles of QDs led to the spectroscopic measurement of an individual QD. Through growth of a sufficiently low density of small QDs and isolation by lithography and etching, the unique spectral signature of emission from a single QD was identified. This was a critical step toward the generation of countable photons on demand.
Mirin, R.P., Roshko, A., van der Puijl, M., Norman, A.G., "Formation of InAs/GaAs quantum dots by dewetting during cooling," J. Vac. Sci. Technol. B 20, 1489 (2002);
Anders, S., Kim, C.S., Sonnenberg-Klein, B., Keller, M.W., Mirin, R.P., Norman, A.G., "Bimodal size distribution of self-assembled InGaAs quantum dots," Phys. Rev. B 66, 125309 (2002); additional publications in preparation.
Contact: Rich Mirin, mirin@boulder.nit.gov, 303-497-7955
NIST Researchers Measure Quantum Dot Shapes and Densities
NIST researchers are measuring the densities and shapes of self-assembled quantum dots (QDs) and determining the uniformity of these features. Many device applications for QDs, for example lasers and photodetectors, require high dot densities and controlled dot sizes. The height and density distributions of InGaAs/GaAs QDs in samples without capping have been measured by atomic force microscopy. Samples from two different molecular beam epitaxy systems and one organo-metallic vapor-phase epitaxy system have been used in the study. The areal QD density was found to vary up to 50 % over the central 2 x 2 cm2 section of two-inch wafers and by as much as 23 % on a length scale of micrometers. Further work is under way to identify the cause of these variations and to find means of minimizing them. The heights of the dots were determined to have a roughly Gaussian distribution for smaller dots. This changed to a bimodal distribution of small and large dots in samples where more InGaAs was deposited. From cross-sectional high-resolution transmission electron microscopy (HRTEM) imaging, NIST scientists determined that smaller dots are pyramidal, while larger dots have multiple side facets. Multiple faceting was observed only in dots more than 8.5 nm in height and allows increased dot volume without a substantial increase in base area. Although studies of size and shape are more easily and commonly performed on uncapped dots, in devices QDs are invariably capped. Plan-view HRTEM was used to measure the dot diameters of both capped and uncapped dots (see photo). The base diameters of the capped QDs were 100 to 200 nm - 3 to 5 times larger than the bases of uncapped QDs. These important results are contributing to the understanding of QD growth and the development of three-dimensional arrays for device applications.
S. Y. Lehman, A. Roshko, R. P. Mirin, and J. E. Bonevich, "Investigation of the Shape of InGaAs/GaAs Quantum Dots," pp. E13.40.1-6 in Physics and Technology of Semiconductor Quantum Dots, eds. A.L. Efros, D.J. Norris, P.M. Petroff, A. Zrenner, The Materials Research Society, Pittsburgh PA (2003).
Contact: Alexana Roshko, roshko@boulder.nist.gov, 303-497-5420
Rich Mirin, mirin@boulder.nit.gov, 303-497-7955
NIST Researchers Measure Nanoscale Strain Adjacent Buried AlGaAs-Oxide Layers
NIST is developing a method for measuring elastic strain in compound semiconductors with high spatial resolution (order of 10 nm). Strain is a major factor in the compound semiconductor photonics industry. It can be beneficial, as in strain driven self-assembly of quantum dots. It can also be detrimental, as in the strain associated with oxidation of buried AlGaAs layers used to define apertures in vertical cavity surface emitting lasers (VCSELs). In this case strain can lead to delamination and device failure. High resolution measurements of strain are needed to understand and control these phenomena. NIST researchers are developing such a technique using electron backscatter diffraction (EBSD). The diffuseness of the EBSD patterns is used as a direct measure of the elastic strain present in the volume of material sampled. With a beam diameter of 20 nm, the sampling volume (which also depends on electron scattering in the material) is estimated to be approximately 30 nm for many compound semiconductors. Measurements have been made of strain in GaAs layers adjacent buried AlGaAs-oxide layers in VCSEL structure samples. In some specimens the largest strain field was found behind the oxide - AlGaAs interface, that is, adjacent the oxidized layer, while in other specimens the largest strain field was in front of the interface, adjacent the unoxidized AlGaAs layer. In both cases the strain was not symmetrically distributed around the oxidized layer but appeared instead to be concentrated on the side closest to the substrate, probably due to the relatively thick substrate constraining the structure more than the thin structure above the oxide layer. These measurements will enable identification of processing conditions that minimize strain, thereby contributing to increased device yields and lifetimes.
R.H. Geiss, A. Roshko, K.A. Bertness, and R.R. Keller, "Electron Backscattering Diffraction for Studies of Localized Deformation," pp. 329-336 in Electron Microscopy: Its Role in Materials Science, eds. J.R. Weertman, M. Fine, K. Faber, W. King, and P. Liaw, TMS Conference Proceedings, Warrendale, PA, (2003);
R.R. Keller, A. Roshko, R.H. Geiss, K.A. Bertness, and T.P. Quinn, "EBSD Measurement of Strains in GaAs due to Oxidation of Buried AlGaAs Layers," to appear in Microelectronic Engineering; additional publications in preparation.
Contacts: Alexana Roshko, roshko@boulder.nist.gov, 303-497-5420
Kris Bertness, bertness@boulder.nist.gov, 303-497-5069
NIST Researchers Measure Dipole Moment of Semiconductor Quantum Dots
NIST researchers are measuring the fundamental properties of semiconductor quantum dots (QDs) in support of the development of devices for quantum computing and communication applications. Little detailed knowledge of the transition dipole moment of QDs has been available to date, although it is important in determining gain for laser design and in implementing the coherent manipulation of the QD state in QD-based quantum logic gates. NIST scientists have demonstrated a technique to directly measure the dipole moment of QDs and have applied it to self-assembled InGaAs/GaAs dots. It relies on the measure of pulses of light that exit a waveguide containing the QDs after reflecting multiple times from the waveguide facets. Light that is coupled out of the waveguide is mixed with a delayed gating pulse in a nonlinear crystal to time resolve the output. The absorption coefficient is determined by comparing the areas of adjacent pulses and taking into account the measured waveguide facet reflectivities. The dipole moment is derived from the ground state absorption and the QD areal density, as determined from transmission electron micrographs. NIST researchers determined that the largest source of uncertainty (15%) was from the QD areal density. The measurement technique overcomes the larger uncertainties in previous measurements that were based on threshold currents of laser diodes or that had large background material absorption or difficulty estimating coupling efficiency into and out of the QD region. Techniques are now being developed to measure the absorption of a single, isolated quantum dot directly.
Silverman, K.L., Mirin, R.P., Cundiff, S.T., Norman, A.G., "Direct measurement of polarization resolved transition dipole moment in InGaAs/GaAs quantum dots," Appl. Phys. Lett. 82, 4552 (2003)
Contact: Rich Mirin, mirin@boulder.nit.gov, 303-497-7955
NIST/EEEL Researchers Develop Single Photon Counting Detectors
NIST researchers have designed and built the world's lowest noise photon detector. With this device they have demonstrated the first system capable of counting photons in a pulse of light at 830 nm, 1.3 µm and 1.55 µm and have achieved quantum-limited detection of photons from 200 nm (UV) to 2 m (IR). End-to-end optical efficiency of ~20% has been achieved and work is underway to increase this to > 80%. Projects are in progress with collaborators at LANL, NASA, Boston University and Stanford University to incorporate these detectors in quantum key distribution networks, and quantum information technology testbeds and to characterize advanced single photon sources.
A.J. Miller, S. Nam, J.M. Martinis, A.V. Sergeinko, "Demonstration of Low-Noise Near-Infrared Photon Counter With Multiphoton Discrimination", Applied Physics Letters, v. 83, 791-793 (2003).
Contact: Sae Woo Nam, nams@boulder.nist.gov, 303-497-3148
NIST/EEEL Demonstrates Spatially Resolved Measurement of Carrier Lifetime
NIST/EEEL researchers have combined optically induced transient analysis with a scanning capacitance microscope (SCM), originally developed at NIST for determining shallow dopant profiles of silicon transistors. The research demonstrates that it is possible to not only sense dopant distributions, but also to measure carrier generation and recombination lifetimes on nanosecond and nanometer scales with the SCM. The work also shows that stray light from the laser used in the atomic force microscope (AFM), part of the conventional SCM, fundamentally changes the signal of the SCM.
The optical transient measurement is implemented by precisely switching the AFM laser on or off and then measuring as a function of time the 1/e relaxation of the capacitance signal between the microscope tip and sample. In commercial combined AFM/SCM instruments, the tip deflection is monitored using a laser beam reflected from the cantilever to which the tip is attached. It was found that scattering of this light onto the sample's surface generates electron-hole pairs, introducing a large background in the capacitance, the primary signal of the SCM. Analytical models of excess carriers generated by a light pulse and MEDICI computer simulations confirm the observed response of the SCM to light. Use of the small area SCM probe to measure carrier lifetime means that the spatial resolution of the technique is limited only by the diffusion length of carriers in the semiconductor under test, the fundamental limit to which spatial variations in carrier lifetime can ever be resolved.
Laser sensors are an essential part of a common commercial implementation of the AFM, but when the same platform is used to measure capacitance with a conducting tip, stray light dominates the measurement. This background has been largely unnoticed by equipment vendors since compensating calibrations have typically masked the effect. However, the EEEL experiments demonstrate that the signal-to-noise of SCM measurements can be improved by an order-of-magnitude or more when acquired with the cantilever laser sensor turned off. This translates into a comparable sensitivity increase for dopant distributions in deep submicrometer transistor junctions.
A patent on the technique has been applied for (U.S. Provisional Patent Application Serial No. 60/443,631). The observation of the profound effect of the AFM laser on any electrical measurements made with the same probe tip could stimulate development of tip-control technologies that are independent of laser beams as platforms for future implementations of SCM and related microscopes.
Buh, G. H., Kopanski, J. J., "Atomic Force Microscope laser illumination effects on a sample and its application for transient spectroscopy," Appl. Phys. Lett., Vol. 83, No. 12, pp. 2486-2488 (22-SEP-2003)
Contact: Joe Kopanski, joseph.kopanski@nist.gov, 301-975-2089
EEEL/HP Research Progresses Toward Critical Molecular Electronics Measurements
NIST researchers in collaboration with Hewlett-Packard Laboratories, US, have developed a technique for measuring the behavior of molecular electronic devices. Two independent laboratories used the method to examine the electronic properties of a molecular-monolayer-based device and obtained almost identical results.
To make a test structure, the researchers sandwiched a monolayer of eicosanoic acid molecules between a series of perpendicular aluminum wires. The resulting alumina (AlOx) tunnel-junction-based crossbar device exhibited hysteretic switching behavior. Richter and colleagues believe this is the result of an interaction between the molecules and the wire electrodes. The scientists also tested crossbar structures without a layer of eicosanoic acid molecules to ensure that the electrical results were due to the presence of the molecular monolayer and not the test structure.
Reproducible electrical measurements of molecular electronic devices will speed the entry-to-market of this promising emerging technology and help lead to a fundamental understanding of how charges are transported in molecules.
C. A. Richter, D. R. Stewart, "Metrology for Molecular Electronics," GOMAC Digest of Technical Papers, GOMACTech 2003, Mar 31, 2003 to Apr 03, 2003, Tampa, Florida, Vol. 28, pp. 281-284 (31-MAR-2003)
Contact: Curt Richter, curt.richter@nist.gov, 301-975-2082
 |
 |
NIST Researchers Developing CD Reference Materials to Calibrate Linewidth Metrology Tools
A team of scientists at NIST/EEEL and International SEMATECH, in conjunction with industry partner VLSI Standards, Inc., is developing critical dimension (CD) reference materials for calibrating linewidth-metrology tools that are used in the advanced semiconductor manufacturing. This application requires reference features having linewidths less than 100 nm and CD uncertainties in the range 2 nm to 5 nm. The approach to reaching this target is to pattern the reference features so that their sidewalls are aligned to specific lattice planes in mono-crystalline silicon films whose surfaces have a specific orientation to the films' surfaces.
A large set of candidate reference features are all subjected to a transfer-calibration metrology having better than nanometer-level precision. Primary calibration is accomplished by applying high-resolution transmission electron microscopy (HRTEM) to a sub-set of the features. The resultant phase-contrast imaging of the near-atomically-rectangular cross-section enables expressing each feature's width as a number of lattice-plane spacings whose known absolute dimensions are referenced to the International Standard Meter. While the features of the subset are unavoidably destroyed during HRTEM imaging, their linewidths, as determined by HRTEM and the transfer metrology, are reconciled in a calibration curve allowing conversion of the transfer-metrology linewidth measurements of the remaining features to their "true" linewidths. Transfer calibration based on electrical test structure metrology techniques has been reported by the project previously. The unacceptable magnitude of the uncertainties is now known to result from electrical test structure metrology's requiring feature lengths of approximately 10 micrometers. Not only are such lengths much greater than those needed for the application, it is now realized that it is impossible to replicate such long features with end-to-end CD-uniformity sufficient to satisfy the targeted uncertainty levels.
By changing the transfer metrology to atomic force microscopy (AFM) performed by NIST staff using the world's foremost AFM facilities at SEMATECH, we have recently projected major reductions in uncertainty that are based on extensive analyses conducted in conjunction with the staff of the Statistical Engineering and the Precision Engineering Divisions at NIST. Most recent projections of uncertainty now indicate that it has dropped from the former 15-nm level to close to 5 nm with nominal feature linewidths close to the targeted 100 nm.
R.A. Allen et al, "Test Structures for Referencing Electrical Linewidth Measurements to Silicon Lattice Parameters Using HRTEM," IEEE Transactions on Semiconductor Manufacturing, Vol. 16, No. 2, pp. 239-248 (01-MAY-2003)
Contacts: Michael Cresswell, michael.cresswell@nist.gov, 301-975-2072
Richard Allen, richard.allen@nist.gov, 301-975-5026
 |
| A recently-patterned reference feature, left, with estimated nominal linewidth of approx. 140 nm and CD uniformity consistent with a projected uncertainty of approx. 5 nm. |
NIST/EEEL Researchers Measure Electrically Active Defects in Nanoscale Gate Dielectrics
In support of the semiconductor industry, NIST researchers, in collaboration with the IBM Semiconductor Research and Development Center, have measured electrically active defects of a HfO2 gate dielectric using a technique called charge pumping. These defects may be responsible for observed drive current degradation in transistors with HfO2.
As the feature sizes of transistors are scaled downward, the gate dielectric capacitance must be increased to reduce short-channel effects and keep at least the same drive current. Historically, this has been accomplished by reducing the thickness of SiO2. However, as the thickness of SiO2 drops well below 2 nm, fundamental limitations associated with tunneling current leaking through the dielectric will prohibit the use of SiO2. This has caused an increased intensity in the research and development of new high permittivity gate dielectrics (e.g., HfO2, ZrO2) with lower leakage current. However, these dielectrics have a large number of technological problems perhaps the worst of which is a generally poor interface with silicon. Measuring defects possibly responsible for mobility and, hence, drive current degradation, is a first step in improving the interface.
J.-P. Han, E. M. Vogel, E.P. Gusev, C. D'Emic, C.A. Richter, D. W. Heh, J. Suehle, "Energy Distribution of Interface Traps in High-k Gated MOSFETs," 2003 Symposium on VLSI Tech. Digest of Technical Papers, Kyoto, Japan, June 12, 2003.
Contact: Eric Vogel, eric.vogel@nist.gov, 301-975-4723
 |
 |
High-resolution microwave power imaging at NIST
As part of the program on high-speed electronics, NIST/EEEL has developed a microwave atomic-force microscope for testing and characterizing microwave probes used for measuring microwave voltages and/or currents in microwave circuits. The research is oriented to the development of high-resolution calorimeters for microwave power imaging. The calorimeters are dielectric, micromachined, bi-material cantilevers. A patterned ferromagnetic thin film deposited at the tip of the cantilever serves as a source of heat and at the same time as a ferromagnetic resonance sensor. As the microscope is moved over the microwave circuit due to the eddy currents or ferromagnetic resonance absorption, the metallic sensor heats that leads to the bending of the cantilever. The bending is proportional to the absorbed power. The image shows the microwave power distribution over a 100 µm x 100 µm area at the edge of a microstrip-line resonator with the dark colors representing the low absorption.
"Metrology for Microwave MEMS Probing," J. Moreland, J. Baker-Jarvis, M. Janezic, P. Kabos, D. Williams, IMAPS Conference, Denver CO, 2002
Contact: Pavel Kabos, kabos@boulder.nist.gov, 303-497-3997
NIST Evanescent Microwave Probe for On-Chip Measurements
The semiconductor industry is in need of methods for measuring dielectric properties on-chip. By the end of the decade, state-of-the-art integrated circuits will contain more than a billion transistors, connected by thousands of meters of ultra-fine copper wiring. The extremely small size of the circuits requires micrometer and nanoscale in situ probing methods to measure the permittivity and loss tangent of the materials and also details of the microcircuits. NIST researchers are in the process of developing a measurement method for characterizing the dielectric properties of chips and thin films at the current frequencies of chip operation. The evanescent microwave probe (EMP), shown in the figure on the left, is a quantitatively determines the electrical properties of materials with high spatial resolution. The EMP consists of a quarter-wavelength coaxial resonator terminated with a small spherical or needle probe suspended over a sample. A resonance is excited in the coaxial resonator by two loops connected to a vector network analyzer. The evanescent electric field emanating from the probe interacts with a material, resulting in a shift of the probe's resonant frequency. The probe can be used for permittivity measurements or scanning of microscale circuit features.
The probe's resonant frequency is a function of the specimen's relative permittivity, conductivity, as well as the probe geometry and position above the sample. In the figure at the left, we show the results of a scan of a ceramic substrate where several copper contacts pads were deposited on the surface. The resonant frequency shifts are sufficient to separate the conductive copper regions from the ceramic regions of the circuit. For this particular image, the probe consisted of a tungsten sphere approximately 0.120 mm in diameter, and was positioned 2 µm above the substrate. The total area scanned by the probe was 0.2 mm x 0.2 mm. To improve the spatial resolution of the probe, we will replace the large spherical probes with smaller needle probes. This type of scan allows, for example, a study of the copper penetration into the substrate.
James Baker-Jarvis, Pavel Kabos, and Christopher L. Holloway, "Nonequilibrium Electromagnetics: Local and Macroscopic Fields and Constitutive Relationships," to be published in Phys Rev E.
Contact: Michael Janezic, janezic@boulder.nist.gov, 303-497-3656
Jim Baker-Jarvis, jjarvis@boulder.nist.gov, 303-497-5621
Pavel Kabos, kabos@boulder.nist.gov, 303-497-3997
 |
 |
NIST Researchers Improve Superconducting Qubit
NIST/EEEL researchers have improved the operation of a superconducting "phase" qubit by incorporating a large area Josephson junction into a superconducting loop. This has decoupled the qubit from its environment and has eliminated quasiparticle heating effects which limited the performance of last year's qubit. A new detection scheme based on an asymmetric d.c. SQUID allows detection of the quantum states with high efficiency and a controllable back action. This new device has revealed previously unknown two-state microwave resonators within the qubit Josephson junction itself. Measurements of Rabi oscillations have shown that these spurious resonators strongly decohere the qubit.
Recent improvements in the fabrication of the Josephson junctions show a reduction in the magnitude of the spurious resonators as well as an improvement in the overall performance of the qubit giving coherence amplitudes as large as 76%, with Rabi oscillation decay times ~100 ns. We are currently fabricating devices with alternative materials and processes and performing detailed experiments in an effort to understand the various decoherence mechanisms so that we may eventually produce a highly coherent solid state qubit for quantum computation.
Contacts: John Martinis, martinis@boulder.nist.gov , 303-497-3597
Ray Simmonds, simmonds@boulder.nist.gov, 303-497-4403
J.M. Martinis, S. Nam, J. Aumentado, C. Urbina, "Rabi Oscillations in a Large Josephson Junction Qubit," Phys. Rev. Lett. 89, 117901 (2002)
R.W. Simmonds, K.M. Lang, D.A. Hite, D.P. Pappas, J.M. Martinis, "Decoherence in Josephson Qubits from Junction Resonances," submitted to Phys. Rev. Lett. (2003)
 |
 |
NIST Researchers Uncover Clue to Successful Manipulation of Paired Electrons
Nanoscale devices that manipulate single electrons of charge e with metrological accuracy are being used at NIST to develop new fundamental electrical standards, such as a capacitance standard based on counting electrons. NIST researchers are currently developing superconducting analogs of these devices, in which electrons are bound in "Cooper pairs" of charge 2e, in order to provide larger currents with metrological accuracy and for quantum computing applications. Past attempts at manipulation of single Cooper pairs for both metrology and quantum computation have been hampered by the existence of a small number of residual unpaired electrons in the superconducting state. Furthermore, the conditions needed to avoid these unpaired electrons in a given device were not clearly understood because different research groups have reported conflicting results on this question.
NIST researchers have recently uncovered an important clue to this mystery by showing that a previously unappreciated factor has a strong effect on the amount of unpaired electrons in Cooper pair devices. Each device consists of two layers of aluminum, and the strength of the pairing of electrons in each layer can be different. This slight difference has generally been thought to be unimportant. However, a study of more than a dozen devices in which this difference was varied in a controlled way, and independently measured in each device, shows that it directly affects device performance. In every device made the 'right' way, unpaired electrons were very rare, allowing manipulation of single Cooper pairs over long time scales and over a wide range of temperature. In all devices made the 'wrong' way, unpaired electrons dominated the device operation and the desired manipulation of single Cooper pairs was not possible. These results, which may also explain previous reports that seemed to be contradictory, point the way to further experiments that can determine the potential of these devices for both metrology and quantum computation.
J. Aumentado, Mark W. Keller, John M. Martinis, and M.H Devoret, "Nonequilibrium quasiparticles and 2e periodicity in single-Cooper-pair transistors," preprint, arXiv:cond-mat/0308253, 2003; and in press, PRL.
Contacts: Mark Keller, mkeller@boulder.nist.gov, 303-497-5430
Jose Aumentado, aumentad@boulder.nist.gov, 303-497-4137
NIST/EEEL Researchers Are Developing Microcalorimeter Imaging Arrays
NIST researchers are developing arrays of high-performance detectors based on superconducting transition-edge sensors (TES) for application in x-ray materials analysis as well as x-ray and sub-millimeter astronomy. Advances in nanotechnology are achieved through both the application of these detectors and the development of large-scale detector arrays. The microcalorimeter x-ray spectrometer developed by NIST allows improved low-voltage x-ray microanalysis of nanoscale structures and particles. In addition, the fabrication of new microcalorimeter arrays utilizes nanoscale micromachined thermal-isolation structures in order to obtain the desired thermal time constants as well as to provide thermal isolation from adjacent pixels. Until recently, thermal isolation of single–pixel devices was achieved by anisotropic wet etching of the entire Si wafer behind the pixel, leaving the detectors supported by a thin Si3N4 membrane. Limitations of this technique make it undesirable for the fabrication of close-packed arrays. One possible means to achieve thermal isolation is surface micromachining. Here, a TES is fabricated on topic of a Si3N4 membrane that is held above the substrate by a small number of support legs. Because the underlying wafer is not thinned or removed, the resulting detector chip is strong and requires no special handling.
Hilton, G.C., Beall, J.A., Deiker, S., Beyer, J., Vale, L.R., Ullom, J., Irwin, K.D., "Surface Micromachining for Transition-Edge Detectors," to be published in IEEE Transactions on Applied Superconductivity.
Contact: Kent Irwin, irwin@boulder.nist.gov, 303-497-5911
Gene Hilton, hiltong@boulder.nist.gov, 303-497-5679
NIST/EEEL Introduces Nano-Stacks of Josephson Junctions for New Voltage Standards
NIST has led the way in the development of programmable DC and AC voltage standards for use in calibration laboratories. The standards have been constructed of superconductor-normal-metal-superconductor Josephson junctions because of their reproducibility, stable voltage steps and their immunity from noise. Now, to improve the quality of the standards, NIST is packing the junctions as densely as possible in three dimensions with junctions separated by a distance of only 20 nanometers. NIST researchers have succeeded in making the first 5-junction stacks using molybdenum di-silicide (MoSi2) as the normal metal and niobium as the superconductor for superconductor-normal-metal-superconductor (SNS) Josephson junctions. Precise three-dimensional control of the junctions during fabrication is critical for achieving uniformity of the electrical characteristics for the junction stacks and large high-density arrays. These new MoSi2 circuits have demonstrated sufficient uniformity for thousands of junctions to display large quantized-voltage steps at frequencies up to 20 GHz. For the past few years the project has been searching for a practical barrier material to allow them to vertically stack junctions in order to make three-dimensional arrays. Higher junction density is required to increase the output voltage as well as the operating bandwidth of both programmable and ac Josephson array circuits. Nanometer control of the barrier thickness, typically 20-30 nanometers, is essential because junction electrical characteristics depend exponentially on barrier thickness. Reproducibility and uniformity of the fabrication process make molybdenum di-silicide the leading candidate for future lumped-array Josephson voltage standard circuits and systems. The best result NIST researchers have demonstrated is a 1 V step with only 8200 stacks (3 junctions in each stack).
Y. Chong, P.D. Dresselhaus, and S.P. Benz, "Thermal transport in stacked superconductor-normal metal-superconductor Josephson junctions," Appl. Phys. Lett., Vol. 8 No. 9, pp. 1794-1796, Sept. 2003.
Contacts: Sam Benz, benz@boulder.nist.gov, 303-497-5258
Paul Dresselhaus, haus@boulder.nist.gov, 303-497-5211
NIST/EEEL Researchers Probe Individual Semiconductor Quantum Dots
Semiconductor self-assembled quantum dots (QD), nanoscale islands of InGaAs grown in a multilayer GaAs structure, are being developed at NIST to create a unique source of single photons for radiometry and quantum cryptography applications. Before this single photon "turnstile" can be made, more information is needed about the electronic properties of the quantum dots. While large ensembles of QDs can be measured by traditional capacitance spectroscopy, the natural distribution of QD sizes allows only the average properties to be determined. To measure the tunneling of individual electrons onto single QDs requires the use of a nanoscale electronic probe. Single Electron Tunneling (SET) devices can be used to make electrometers sensitive enough for this purpose. We have fabricated SET electrometers on top of a wafer containing QDs to perform electronics spectroscopy of the electron states in individual QDs. As shown schematically in cross section, the SET electrometer sits above a field of QDs. For sufficiently low QD density, the properties of single (or a few) QD can be determined. In our first measurements of this type, the SET measured electron tunneling into 3 different dots. We have extracted the Coulomb blockade energy for the dots, and determined the first 2 energy levels in one of the dots. Future work will allow more levels to be determined. This information is critical to designing an electronically triggered single photon source using these QD structures.
K.D. Osborn, M.W. Keller, and R.P. Mirin, "Single-Electron Transistor Spectroscopy of InGaAs Self-Assembled Quantum Dots," in press, Physica E.
Contact: Mark Keller, 303-497-5430, mkeller@boulder.nist.gov
NIST/EEEL Researchers Create New Quantum Capacitance Standard
The unit of capacitance, the Farad, is realized through a complex instrument known as the calculable capacitor, which determines capacitance through a measurement of length. This approach is very precise, but the instrument is difficult enough to create and operate that only a few exist in the world. NIST researchers have developed a prototype new capacitance standard based on the definition of capacitance C=Q/V, where Q is the charge placed on the capacitor and V is the resulting voltage. The enabling technology for this new approach is nanoscale electronic devices that can manipulate and detect individual electrons. These single electron tunneling (SET) devices are based on extremely small tunnel junctions (typically 40 nm on a side) fabricated through electron-beam lithography. A chain of 7 SET devices in series (see figure) is used to pump approximately 100 million electrons (±1 electron) one at a time onto a capacitor held at very low temperature (50 mK). The voltage across the capacitor is measured, calibrating the capacitor. Since the heart of this system is the SET electronic chip, multiple copies of the new standard can be made each with identical intrinsic accuracy. Work is now underway to compare the capacitance determined with the Electron Counting Capacitance Standard to that of the calculable capacitor.
M.W. Keller, A.L. Eichenberger, J.M. Martinis, and N.M. Zimmerman, "A Capacitance Standard Based on Counting Electrons," Science 285, 1706 (1999).
M.W. Keller, "Standards of Current and Capacitance Based on Single-Electron Tunneling Devices," in Proceedings of the International School of Physics "Enrico Fermi" Course CXLVI: "Recent Advances in Metrology and Fundamental Constants," Amsterdam, IOS Press (2001).
Contact: Mark Keller, 303-497-5430; mkeller@boulder.nist.gov
