Dale E. Gary

THz imaging and sensing for security applications—explosives, weapons and drugs

Over the past 5 years, there has been a significant interest in employing terahertz (THz) technology, spectroscopy and imaging for security applications. There are three prime motivations for this interest: (a) THz radiation can detect concealed weapons since many non-metallic, non-polar materials are transparent to THz radiation; (b) target compounds such as explosives and illicit drugs have characteristic THz spectra that can be used to identify these compounds and (c) THz radiation poses no health risk for scanning of people. In this paper, stand-off interferometric imaging and sensing for the detection of explosives, weapons and drugs is emphasized. Future prospects of THz technology are discussed.

MOTION OF FLARE FOOTPOINT EMISSION AND INFERRED ELECTRIC FIELD IN RECONNECTING CURRENT SHEETS

A systematic motion of Hα kernels during solar flares can be regarded as the chromospheric signature of progressive magnetic reconnection in the corona, in that the magnetic field lines swept through by the kernel motion are those connected to the diffusion region at the reconnection point. In this paper, we present high-cadence and high-resolution Hα-1.3 A observations of an impulsive flare that exhibits a systematic kernel motion and relate them to the reconnecting current sheet (RCS) in the corona. Through analyses of X-ray and microwave observations, we further examine the role of the macroscopic electric field inside the RCS in accelerating electrons. We measure the velocity of the kernel motion to be 20-100 km s-1. This is used together with the longitudinal magnetic field to infer an electric field as high as 90 V cm-1 at the flare maximum. This event shows a special magnetic field configuration and motion pattern of Hα kernels, in that a light bridge divides a flare kernel into two parts that move in different manners: one moving into the stronger magnetic field and the other moving along the isogauss contour of the longitudinal magnetic field. The temporal variation of the electric field inferred from the former type of kernel motion is found to be correlated with 20-85 keV hard X-ray light curves during the rise of the major impulsive phase. This would support the scenario of magnetic energy release via current dissipation inside the RCS, along with the hypothesis of the DC electric field acceleration of X-ray-emitting electrons below 100 keV. However, there is no good temporal correlation between the hard X-ray emission and the inferred electric field from the other motion pattern. Furthermore, the microwave emission, which supposedly comes from higher energy electrons, shows a time profile and electron spectrum that differs from those of the X-ray bursts. We conclude that either the two-dimensional magnetic reconnection theory related to the Hα kernel motion is applicable to only some part of the flare region due to its special magnetic geometry, or the electron acceleration is dominated by other mechanisms depending on the electron energy.

MAGNETIC RECONNECTION AND MASS ACCELERATION IN FLARE-CORONAL MASS EJECTION EVENTS

An observational relationship has been well established among magnetic reconnection, high-energy flare emissions and the rising motion of erupting flux ropes. In this paper, we verify that the rate of magnetic reconnection in the low corona is temporally correlated with the evolution of flare nonthermal emissions in hard X-rays and microwaves, all reaching their peak values during the rising phase of the soft X-ray emission. In addition, however, our new observations reveal a temporal correlation between the magnetic reconnection rate and the directly observed acceleration of the accompanying coronal mass ejection (CME) and filament in the low corona, thus establishing a correlation with the rising flux rope. These results are obtained by examining two well-observed two-ribbon flare events, for which we have good measurements of the rise motion of filament eruption and CMEs associated with the flares. By measuring the magnetic flux swept through by flare ribbons as they separate in the lower atmosphere, we infer the magnetic reconnection rate in terms of the reconnection electric field Erec inside the reconnecting current sheet (RCS) and the rate of magnetic flux convected into the diffusion region. For the X1.6 flare event, the inferred Erec is ~5.8 V cm-1 and the peak mass acceleration is ~3 km s-2, while for the M1.0 flare event Erec is ~0.5 V cm-1 and the peak mass acceleration is 0.2-0.4 km s-2.

Terahertz study of 1,3,5-trinitro-s-triazine by time-domain and Fourier transform infrared spectroscopy

This letter describes the use of THz time-domain spectroscopy (TDS) applied in transmission to the secondary explosive 1,3,5 trinitro-s-triazine. Samples were also subjected to Fourier transform infrared spectroscopy over the same range for comparison. A detailed spectroscopy study is presented. General agreement between results from both methods confirms the absorption features found. A comparison study with computer molecular simulations shows that THz-TDS is sensitive to collective modes or vibrational modes of material.

Coronal temperature, density, and magnetic field maps of a solar acitve region using the Owens Valley Solar Array

We present the first results of solar active region observations with the recently completed five-element Owens Valley Solar Array. On 1991 October 24, maps of Active Region AR 6891 were obtained at 22 frequencies from 1.2-7.0 GHz to provide brightness temperature spectra at each point. This is the first time that both high spatial and frequency-resolution brightness temperature spectra have been available over such a broad radio-frequency range. We find that over most of the region the spectra fall into one of the two well-defined categories: thermal free-free or thermal gyroresonance. In these cases, we use the spectra to deduce the spatial variation of physical parameters-electron temperature, column emission measure (intergral n(sup 2)(sub e) dl), and the coronal magnetic field strength-in and around the active region. Over a limited area of the region, the spectra resemble neither of the simple types, and alternative interpretations are required. The possibilties include the presence of fine structure that is unresolved at low frequencies; the presence of a small number of nonthermal electrons; or the presence of overlying, cooler 10(exp 6) K material which at low frequencies absorbs the hot (3 x 10(exp 6) K) thermal emission generated below.

FLARE-RELATED MAGNETIC ANOMALY WITH A SIGN REVERSAL

In this paper we report a significant magnetic anomaly, specifically an apparent sign reversal of magnetic polarities in small areas of Michelson Doppler Imager (MDI) magnetograms during the impulsive phase of an X5.6 flare on 2001 April 6. Three flare kernels were observed to emit ?50 keV hard X-rays, which are located in strong magnetic fields of order ?1000-1500 G. We find that the apparent sign reversal began and persisted for a few minutes in all three kernels, in precise temporal and spatial correspondence with the hard X-ray sources. We search for a combination of instrumental and flare-induced line profile effects that can account for this behavior. Our studies provide a viable scenario that the observed transient sign reversal is likely to be produced by distorted measurements when the Ni I 6768 ? line comes into emission or strong central reversal as a result of nonthermal beam impact on the atmosphere in regions of strong magnetic fields.

Statistical Study of Two Years of Solar Flare Radio Spectra Obtained with the Owens Valley Solar Array

We present results of analysis of 412 flares during 2001-2002 as detected by the Owens Valley Solar Array (OVSA). This is an in-depth study to investigate some results suggested by a previous study of solar bursts (Nita et al. 2002), which was limited to the peak time of the bursts at a few frequency bands. The new study includes the temporal dependence, at 4 s time resolution, of parameters measured at 40 frequencies in the range 1-18 GHz. We investigate distributions of burst parameters such as maximum flux density in the spectra, peak frequency, spectral slopes below and above the peak frequency (optically thick and thin slopes, respectively), and burst durations. We classify the microwave bursts according to their spectral properties and provide tables of averaged spectral parameters for each spectral type and for different frequency and intensity ranges.

Correlation of Microwave and Hard X-Ray Spectral Parameters

We present the analysis of 27 solar flares with multiple peaks that were observed at hard X-ray and microwave wavelengths. A total of 57 simultaneous peaks were observed by BATSE (hard X-rays) and Owens Valley Radio Observatory (microwaves). Throughout the duration of a flare, its spectra at both wavelengths are fitted independently at all times. The hard X-ray spectra were fitted by a single power law in most cases, whereas the microwave spectra were fitted as gyrosynchrotron emission. For each individual peak, the parameters at both wavelengths (peak flux, turnover frequency, spectral indices, and delays between hard X-ray and microwave peak emission) were then compared and correlated. We have also studied impulsive and nonimpulsive bursts individually. The main results obtained were as follows. (1) In 75% of the bursts, the inferred index of the electron energy distribution of the microwave-emitting electrons, δr, is harder than that of the lower energy hard X-ray-emitting electrons, δX, on average by 0.5-2.0. This implies that there is a breakup in the energy spectra of the electrons, as is sometimes observed in the hard X-ray spectra of giant flares. (2) A soft-hard-harder spectral index temporal evolution is more commonly seen in the microwave spectra (47%) than in the hard X-ray observations (32%) and in nonimpulsive flares than in impulsive ones. (3) Delays larger than 2 s were observed between the radio and hard X-ray peaks in 65% of the bursts, with the delays decreasing as the hard X-ray energy increased. (4) Nonimpulsive flares are more microwave rich, have higher delays between their radio emission and the hard X-ray peaks, and display harder spectral indices than impulsive bursts.

Terahertz imaging using an interferometric array

Most methods of imaging in the terahertz (THz) spectral region utilize either pulsed-laser sources or require the THz generation and detection sources to be phase coherent. The application of interferometric imaging to the THz range is described. Interferometric imaging offers considerable advantages in this regard due to its ability to image with only a handful of detector elements, image many sources of THz radiation at once, image incoherent as well as coherent sources, and provide spectral information as well as spatial imaging information. The THz interferometric imaging method is potentially useful for remote detection of explosives.

The Peak Flux Distribution of Solar Radio Bursts

We have investigated the peak flux distribution of 40 years of solar radio burst data as a function of frequency and time over a wide range of frequencies. The bursts were reported by observing stations around the world during 1960-1999, as compiled by the National Geophysical Data Center (NGDC) of the National Oceanic and Atmospheric Administration (NOAA). This period covers three full and two partial solar cycles. We have analyzed the data set to find correction factors for missed events, and find evidence that nearly half of the events were missed by the worldwide network. We obtain power-law fits to the differential (density) (dN/dS in events sfu-1) and cumulative [N(> S) in events] distributions as a function of frequency, time, and phase of the solar cycle. The typical power-law index, ~-1.8, is similar to that found in many hard X-ray studies. The average waiting time between bursts with flux density exceeding 1000 sfu was found to be 6 days at solar maximum, and 33 days at solar minimum. Taking account of missed events, the expected waiting time decreases to 3.5 and 18.5 days, respectively. Bursts of this flux level can cause problems with wireless communication systems. We present tables of fit parameters that can be used to find burst occurrence rates in a number of frequency ranges. We find no significant variation of power-law index from one solar cycle to the next, or with phase of the solar cycle, but we do find significant changes of power-law index with frequency.