Time and Frequency Division

Technical Highlights

• New High-Power Laser. Staff of the Optical Frequency Measurement Group, in a collaboration with Spectra Diode Laboratories Inc., have developed a new high power laser designed for both scientific and commercial applications. The laser generates 20 times more power (as much as 1 W) than previous single-mode tunable laser systems. The laser is designed to replace the argon-ion-laser-pumped Ti:sapphire laser. The new system is very small, does not require a pump laser, and requires less than 10 W of input power.
  
  The laser was developed under a Cooperative Research and Development Agreement (CRDA). It has potential applications in frequency doubling, pumping of solid-state lasers, a wide range of spectroscopy, atom cooling and trapping, analytical chemistry, and trace detection of atoms and molecules. (L. Hollberg)
• New Laser Source at 194 nm. Jim Bergquist, Leo Hollberg, and Linda D'Evelyn have recently collaborated on a major simplification of the 194 nm laser system used for cooling and probing in all of the trapped-mercury-ion experiments. The first source ever operated at this wavelength was developed nearly 10 years ago by Jim Bergquist. The new system typically generates 60 µW to 80 µW of 194 nm power, whereas the old system had an output of between 2 µW and 5 µW. The new system also eliminates two krypton-pumped dye lasers reducing power consumption from 90 kW to 30 kW. The two dye lasers were replaced by a Ti:sapphire laser and a diode laser, both built at NIST.
  
  The next improvement will replace an argon-ion laser with a diode-pumped crystal laser. This will complete the conversion to an all solid-state system simplifying operation and improving reliability. Evolving technology will continue to provide the simplifications required to make the cooled, mercury-ion frequency standard a practical reality. (J.C. Bergquist)
• Calcium Optical Frequency/Wavelength Standard. The Optical Frequency Measurements Group has achieved very narrow linewidths in a new calcium beam-cell system. Their system (shown in Figure 1) exhibits a 65 kHz linewidth and signal-to-noise measurements indicate a stability of 3 × 10^-14 τ^-1/2 where τ is the averaging time. For comparison, the natural linewidth of the iodine transition used to stabilize helium-neon length standards is 400 kHz. The calcium system is still far from optimized, and no cooling of the atoms has been used in the experiments. With laser cooling the system linewidth should be dramatically better, since the natural linewidth of the calcium transition is 400 Hz. (L. Hollberg)



Figure 1: Calcium optical frequency standard. Radiation at 657 nm from the Extended Cavity Diode Laser (ECDL) probes atoms in the calcium beam/cell.

• Optical Frequency Doubling. Carl Weimer, Hugh Robinson, Rich Fox, and Leo Hollberg have developed two different systems for frequency doubling IR light from diode lasers. In the first system KNbO₃ is non-critically phase matched to convert 150 mW of 850 nm IR to produce more than 35 mW of single-frequency, tunable blue light. The second system uses LiIO₃ in an angle-matched configuration to generate about 100 µW of light at 405 nm.
  
  The more powerful blue light is being used to demonstrate detection of Cr in water using laser-enhanced-ionization spectroscopy. This system will soon be tuned to a calcium resonance for laser cooling of a calcium atomic beam which should improve this system for length applications. The 405 nm radiation is now being used to detect trace quantities of lead in a small rf discharge. In general, diode lasers will find more metrology applications as spectral coverage is broadened toward shorter (blue) wavelengths and longer infrared wavelengths. (H. Robinson)
• Cooled Cesium Atoms. John Marquardt, Rich Fox, Hugh Robinson, and Leo Hollberg of the Division along with Sarah Gilbert of EEEL have been using magneto-optic trap technology in a vapor cell to confine and cool cesium atoms to temperatures of a few hundred microkelvin. These atoms are then probed using an independent optical beam in a cascade, two-photon transition. Figure 2 below shows the absorption of 658 nm laser light, whose frequency corresponds to the second step of the two-photon process, as a function of both this laserfrequency and the 852 nm frequency which serves as the first step in the process as well as producing the magneto-optic trapping. These measurements provide good trap/cold-atom diagnostics. The concept is being developed for application to high accuracy multiphoton transitions for frequency/wavelength references. (R. Fox)



Figure 2: Absorption at 658.8 nm in a cesium magneto-optical trap as a function of trapping laser frequency (852 nm) and the frequency of the 658 nm source.

• NOAA-NIST Collaboration on Atmospheric Chemistry. In a continuing collaboration between NOAA's Aeronomy Laboratory and NIST's Optical Frequency Measurement Group, red diode lasers are being used to detect NO₃ radicals. This molecule plays an important role in atmospheric photochemistry. In the stratosphere its photochemistry affects ozone depletion, and it acts as a strong oxidizer in the troposphere.

  NO₃ has a strong electronic absorption band at 662 nm that can be reached with commercial red diode lasers. In these experiments a solitary diode laser is used to detect the time-dependent concentration of NO₃ in an excimer-laser photodissociation experiment. Good detection sensitivities are achieved with simple direct absorption measurements. Current sensitivity is about 5 × 10¹⁰ NO₃ molecules/cm³ with a 1 m, single-pass absorption path length. The sensitivity could be improved by subtracting residual low-frequency AM noise, using high frequency modulation, or multi-passing the beam in the absorption cell.

  The experiment illustrates the potential of very simple diode laser systems for detecting reactive molecules that play an important role in the laboratory and in the earth's atmosphere. (R. Fox)
• Optically Pumped Far-Infrared Laser Line. The Laser Spectroscopy Group has just discovered a 124 µm methanol laser line that is more than two times stronger than the previous "strongest" line (a 119 µm line also in methanol). The experiments were performed by guest researchers Che-Chung Chou and Jow-Tsong Shy from Tsing Hua University in Taiwan and Ken Evenson of the Time and Frequency Division. The discovery is of special interest to plasma physicists who use far infrared radiation to measure electron densities in plasmas. It is also of interest to spectroscopists because spectroscopic sensitivity is very high in the infrared.
  
  The new line was found using other newly discovered (at NIST) lines of the 9 µm hot bands in the NIST ribbed-tube CO₂ laser. The radiation from one of these new CO₂ lines was used to pump methanol to produce the 124 µm radiation. Twenty five additional far-infrared lines in methanol were also discovered using the new CO₂ lines. (K.M. Evenson)
• Precision Measurement of Atomic Fine-Structure Frequencies. Ken Evenson of the Laser Spectroscopy Group recently made highly accurate (0.1 parts per million) measurements of the fine-structure frequencies of singly ionized atomic nitrogen (N⁺). In the past 1-1/2 years, fine-structure intervals have also been accurately measured in Al, Fe, S, Si, O-17, and F⁺. These high-accuracy spectral measurements by NIST have been supported by NASA's Office of Laboratory Astrophysics. The objective of such accurate spectral measurements is to provide sufficient information on atomic and molecular species to enable searches in the upper atmosphere and interstellar medium.

  These intervals were previously known by subtraction of optical spectral wavelength data, but their direct measurement has improved accuracy of the intervals by more than 10³.

  The N⁺ measurements provided solid confirmation of tentative radio astronomy observations of this species which is an important constituent of the interstellar medium. Aside from confirmation of this particular measurement, the laboratory measurements yield a second well-defined frequency that can be used in future radio astronomy studies. (K.M. Evenson)
• New Carbon-Dioxide Laser Lines. The Laser Spectroscopy Group has recently observed 40 new continuous-wave laser lines of carbon dioxide. These lines, the 9 µm hot-band lines, some developing over 8 W of output power, will provide a new source of laser radiation for spectroscopy. Frequencies of 36 of the 40 lines have been directly measured relative to other well known CO₂ lines using a heterodyne technique.
  
  The carbon dioxide laser has been so thoroughly studied over many years that it is surprising to find new laser lines in it. The new observations were made possible by the use of a special high-resolution grating (a grating-coupled laser cavity was used), a ribbed laser tube, and a higher than normal discharge current. The end components of this new laser are shown in Figure 3. The lines had been predicted by theory, but never observed until this work. (K.M. Evenson)



Figure 3: End components of the CO₂ laser used for observing new laser lines.

• International Comparison of Phase-Noise Measurements. The first international comparisons of phase-noise measurements between the United States, France and Switzerland have been successfully completed. NIST-developed systems were demonstrated by Fred Walls of the Phase Noise Measurements Group at the 5th European Frequency and Time Forum in Neuchâtel, Switzerland. He then took these systems to laboratories in France and Switzerland for comparisons with their measurement systems. Phase noise is rising in importance in the specification of equipment for telecommunications, radar and aerospace electronics.
  
  Comparisons were made at carrier frequencies of 5 MHz, 10 MHz and 100 MHz at Fourier frequencies extending up to 10 % away from the carrier. Good agreement (±1 dB) was found at 5 MHz and 10 MHz, but some recalibration of equipment was required before agreement was achieved at 100 MHz. Because the NIST measurement method provides an internal capability for evaluating errors, it is easy enough to trace problems of this sort. (F.L. Walls)
• Microwave Phase-Noise Standard. Fred Walls of the Phase Noise Measurements Group has recently completed development of an accurate phase-noise standard at 10.6 GHz. The accuracy of ±0.5 dB for Fourier offsets from 100 Hz to 300 MHz from the carrier is at least a factor of 10 better than that achieved in any conventional measurement system. The standard is portable and thus useful in providing on-site verification of calibration accuracy. It can also be used to verify noise-floor measurements in phase-modulation and amplitude-modulation noise measurement systems. The standard facilitates more rapid measurement of added noise in amplifiers, mixers, and multipliers. (F.L. Walls)
• High-Stability Rubidium Frequency Standards. Fred Walls, John Lowe, and Csaba Szekely have developed a concept which promises to greatly improve the stability of rubidium frequency standards. In earlier work with Bob Drullinger of the Atomic Beam Standards Group, the discharge lamp in these standards was replaced by a diode laser providing substantial improvement in performance, but noise on the oscillator which probes the rubidium transition still limits performance.
  
  The concept currently under development is aimed at reducing noise at ±2 times the modulation frequency of the phase lock loop. This is done using a specially fabricated crystal notch filter. Figure 4 shows the reduction in noise for one of the two noise sidebands. The hope is that this will provide rubidium stability performance approaching 1 × 10^-14 τ^-1/2. In preliminary experiments, this concept has been used to reduce deliberately added noise by a factor of 20. Further development of the rubidium standard is required before the concept can be applied to improve performance beyond that of conventional rubidium standards. (J.P. Lowe)



Figure 4: Spectral density of phase noise as a function of frequency for filtered probe-oscillator signal for the rubidium frequency standard. Modulation frequency is 37.5 Hz.

• Building Vibration Studies. Fred Walls of the Phase Noise Measurements Group has provided measurement support to Boulder's Technical Services Division (responsible for buildings and grounds) in the joint development of a 12 channel portable system for measuring the mechanical vibration transfer function of laboratory areas, equipment mounting systems, and isolation tables. The analysis frequency range for the system is 0.1 Hz to 500 Hz. The system can handle inputs from 12 independent vibration sensors allowing for rapid measurement of vibration levels and identification of the various nodes in vibrational modes of laboratory areas, mechanical equipment supports, and isolation tables. Development of the system was motivated by recent positive experience in using vibration measurements to identify and eliminate sources of vibration noise. This latter work was carried out under contract and suggests a number of measures which might be taken to further reduce such noise on the Boulder site. The availability of local equipment should provide more rapid assessment of problems and the level of success of any cures. (F.L. Walls)
• NIST-7 Cited as Significant Achievement. The NIST-7 atomic clock was selected by the editors of Popular Science Magazine for inclusion in the feature section, "The 100 Best of What's New in 1993," of their December 1993 issue. This special 24-page section honors the year's most significant new products and achievements. The award, an embossed medallion, was presented to NIST at an awards luncheon held November 17, 1993 in New York City.
  
  NIST-7 went into service as the official U.S. standard of frequency on January 1, 1993. It's developers, Robert Drullinger, David Glaze, and John Lowe of the Atomic Beam Standards Group received Gold Medals from the Department of Commerce on October 27, 1993 for their development of the standard. NIST-7 was initially evaluated at an uncertainty of 4 × 10^-14, but the evaluation has recently been improved to 2 × 10^-14. The uncertainty goal of 1 × 10^-14 should be achieved within the next six months. (R.E. Drullinger)
• International Time and Frequency Interactions. During 1993 NIST prepared and presented training workshops on time and frequency metrology at the Egyptian National Institute of Standards and the Singapore Institute of Standards and Industrial Research. The Egyptian workshop, prepared by Wayne Hanson and Jim Jespersen, was funded by the State Department and offered in early 1993 to participants from the Middle East and North Africa (excluding Iraq, Iran, and Libya). During the visit a GPS common-view receiver was installed in Egypt by NIST. This receiver connects this region of the world to Coordinated Universal Time (UTC) through comparison schedules arranged by the BIPM. With funding from the government of Singapore, Dave Howe went to Singapore later in 1993 to present a time and frequency workshop to members of this country's standards laboratory and some representatives from other government agencies. This workshop coincided with their acquisition of a U.S. system incorporating a set of ensembled atomic clocks and a GPS receiver providing coupling to UTC. (D.W. Hanson)
• Potential Time Service. The Division has entered into a Cooperative Research and Development Agreement (CRADA) with NAVSYS Corporation and COMSAT in a study of timing signals relayed by geostationary satellites operated by INMARSAT. Three satellites in this system, called the INMARSAT Geostationary Overlay, will be operated from the U.S. by the FAA. The primary objective of the system is to provide integrity (status) information on civilian GPS navigation. However, one additional objective is to provide signals from the geostationary satellites to augment GPS navigation.
  
  Under this CRADA, NAVSYS, COMSAT, and NIST are studying the potential for using the system for providing accurate, wide-area timing signals that are free of the intentional degradation of the civilian signals available from GPS. In principle this could result in a higher accuracy timing available in real time. A key advantage is that the signals broadcast from these satellites can be accessed using GPS receiving equipment. The experiments, involving receivers placed at NIST-Boulder and the INMARSAT uplink station in Southbury, CT have just begun. (M.A. Weiss)
• INTERNET Time Service. A new nationally available time service has been established on the INTERNET. The servers for this service were developed by Judah Levine and went into operation in August of 1993. The service responds to a pent up need for modestly accurate time on a wide range of systems connected to the INTERNET. With virtually no advertising, use of the system has grown to more than 5000 requests per day in a 3 month span.

  Time on the server is maintained within 1 ms of UTC. Users can receive time codes in three common formats. The time codes include advance notice of changes to and from daylight saving time and advance notice of insertion of leap seconds. User software can be downloaded directly from the DAYTIME directory on the server which has an INTERNET address of time.nist.gov.

  A special algorithm has been developed for software which can be used to operate servers at any node on the INTERNET. The algorithm incorporates the smart-clock concept, a NIST-patented method for improving the performance of a remotely operated clock. Through repeated comparisons with an external standard the clock in the server is characterized and then regularly corrected. The result is substantially improved clock performance and a gradually diminishing need for comparisons with the external source. This software is available for servers in industrial, government, and academic institutions. (J. Levine)
• Two-Way-Time-Coordination Link to Europe. Christine Hackman of the Time Scale and Coordination Group and Wayne Hanson and Al Clements of the Time and Frequency Services Group have completed construction and preliminary testing of a satellite earth station which will provide for extremely accurate time and frequency comparisons between NIST and key national laboratories in Europe. The method promises an order of magnitude improvement in time-transfer accuracy providing better international time coordination. This is important in an era where atomic clock stability and accuracy is advancing rapidly. Improvement in international timekeeping is needed in international activities such as telecommunications and navigation.
  
  The new NIST station can be used with both domestic and international communication satellites, and is expected to achieve a time comparison accuracy of one nanosecond or better and stability of comparison in the 100 picosecond regime. There is substantial signal delay between the earth station and the satellite, but proper handling of two-way data exchanges broadcast through the satellite link results in a nearly perfect cancellation of path-delay errors. (M.C. Hackman)
• Minimization of Second-Order Doppler Shift. As the aspect ratio of a cloud of ions contained in a Penning trap is varied (by radiation-pressure-induced angular momentum), the second-order Doppler shift goes through a minimum. By operating in this condition, the system frequency becomes, to first order, insensitive to small fluctuations in cloud conditions (shape, etc.). This implies improved stability when the system is operated as a clock. Recent improvements in the Penning trap apparatus have allowed loading of substantially larger ion clouds implying still better signal-to-noise ratio and thus even better stability.
  
  Simulations of these conditions for both pure plasmas and sympathetically cooled ions are underway, and preparations are being made for experimentally testing them in the new apparatus. (D.J. Wineland)
• Modes in Trapped Electron Plasmas. Carl Weimer of the Ion Storage Group has conducted studies of modes in electron plasmas stored in a cryogenic Penning trap. The electrons are detected by their image currents in one of the trap electrodes. For a plasma of approximately 40,000 electrons, a spectrum analysis of the image currents in the neighborhood of the trap axial frequency is shown in Figure 5. The broad resonance is due to the resonant circuit connected to the electrodes. The narrow spectral features superimposed on the broad resonance are caused by collective modes of the plasma.

  

  Figure 5: Plasma modes for an electron plasma contained in a Penning trap. The sharp features are caused by collective modes of the plasma detected by their image currents in one of the trap electrodes.

  Under conditions of the experiment where the Debye length is much smaller than the plasma dimensions which are much smaller than the trap dimensions, the modes of a nonneutral Penning trap plasma are exactly calculable and, for a given trap axial and cyclotron frequency, depend only on the rotation frequency or, equivalently, the density of the plasma. With this diagnostic, the shape and density of the cloud can be determined (nondestructively), an important consideration where direct imaging techniques are not possible (e.g., for positron and antiproton or ion plasmas which do not have a convenient transition for laser scattering). (J.J. Bollinger)
• Time Scale Reliability. Judah Levine and Jim Gray have substantially enhanced the reliability of the NIST time scale through development of completely redundant clock measurement systems. Two identical measurement systems, controlled by independent PCs, observe the same physical clocks and both run independent copies of the AT1 time scale. Both drive independent micro-steppers to provide redundant physical realizations of UTC(NIST) in real time. One system is used as the official output, but the other can be switched into service should there be any failure of the primary system.

  In addition to providing increased reliability and a continuous check on each other, the dual systems are being used to study the performance of the AT1 algorithm in general and especially its robustness in the presence of measurement noise. Studies of particular value include (1) the effects of noise on the reset algorithm and (2) the averaging time constant for variance of each clock, both of which are implemented in a semi-empirical manner.

  This work provides the basis for long-term changes which are of importance. First, these systems will facilitate replacing all of the large computers with arrays of relatively cheap PCs. In addition to reducing maintenance costs, such a network (combined with GPS receivers) could facilitate wide-spread sharing of data and network-based dissemination of UTC(NIST) at very high accuracy. The development of dual measurement systems also provides the basis for moving the time scale when building and remodelling begins on the site. The longer-term objective is to separate (at opposite ends of the site) two groups of clocks with their independent measurement systems. These two clock systems would be interconnected, but the physical separation would guard against time scale disruption by a catastrophe such as a major fire. (J. Levine)
• Electromagnetic Distance Measurement. Judah Levine, Alan Brewer, and William Waite have further refined their three-color electromagnetic Distance measurement (EDM) system and have performed measurements indicating uncertainty at the 1 mm level (see Figure 6). Their first comparison has been with GPS measurements. In these measurements, they compared EDM and GPS measurements of a 24 m baseline near Boulder. This three month test involved substantial post processing of the GPS measurements. The instrument has been transported to Holloman Air Force Base in New Mexico where it is being compared with well established baselines of 10 km to 40 km.

  

  Figure 6: Comparison of electromagnetic distance measurement and GPS distance measurement for a 24 km baseline. The solid curve and histogram give the EDM results and the lower curbe with points gives the GPS result.

  The EDM uses measurements at multiple wavelengths to develop information needed to correct for dispersion in the atmosphere. This yields an accuracy improvement of better than a factor of 100 over single-color instruments. This new instrument has better reproducibility and smaller scatter than is obtained with GPS, and the measurements are available in real time. No long-term averaging or post processing of data is required.
• Tilt Measurements. Judah Levine and Mary Kohl have completed measurements using tilt meters buried in Southern California. The three instruments were installed near each other but at depths of 24 m, 36 m, and 122 m. The instrument sensitivity is 2 nrad and the long-term stability is 1 µrad per year.
  
  The instruments show different earth-tide signals. The difference was modelled as due to a change in elastic properties above and below the water table which is at about 30 m -- just between the upper two of the instruments. The work demonstrates the importance of local effects on tilt measurements and the ability of finite-element models to explain the observations in a quantitative manner. This is the first quantitative analysis of tilt data at the nanoradian level. The work, which served as a Ph.D. dissertation for Mary Kohl, was awarded the best student paper in Geodesy at the annual American Geophysical Union meeting in San Francisco. (J. Levine)