Electroabsorption modulators (EAMs) based on the quantum confined Stark effect in multiple quantum wells (MQWs) have advantages for high-speed, low drive voltage, and high extinction ratio applications. In this paper, a traveling- wave electrode structure is proposed to achieve high bandwidths with long devices and lower drive voltages at 1.55 micrometers wavelength. An InGaAsP/InGaAsP MQW traveling-wave electroabsorption modulator (TWEAM) with a bandwidth above 20 GHz is fabricated. A drive voltage of 1.2 V for an extinction ratio of 20-dB is demonstrated. The effects of microwave transmission on the high-speed performance of TWEAMs are discussed. Successful data transmission experiments at 30 Gbit/s show a promising system performance of these devices. By using an integrated tandem TWEAM, pulse data transmission above 100 Gbit/s is achieved.

Balanced electroabsorption modulators (B-EAM) are an attractive alternative to the cross-coupled Mach-Zehnder modulator. The B-EAM enables bias independent suppression of even-order distortions, relative intensity noise (RIN), and common amplified spontaneous emission noise. By biasing the B-EAM at the 3rd order null, a 5th order distortion limited spurious free dynamic range can be achieved. We report on the experimental demonstration of the simultaneous suppression of laser RIN, 2nd and 3rd order distortions using a 300 micrometers long B-EAM.

A DC-voltage controlled optoelectronic phase shifter (OEPS) integrated with a PZT-controlled active stabilizer is proposed for both the reduction in timing jitter and the tuning in delay-time of optical pulses generated from argon- ion-laser-pumped, passively mode-locked femtosecond Ti:sapphire laser with intracavity saturable Bragg reflector (SBR). The rms timing jitter (100 - 500 Hz) of the Ti:sapphire/SBR laser is significantly suppressed down to 290 fs with uncorrelated single-sided-band phase noise of less than -120 dBc/Hz at offset frequency of 5 KHz. The pulsewidth, the repetition rate, and the single-sided-band phase noise of the laser are found to keep invariant during the delay-time tuning process. The maximum phase-tuning range and tuning gain of the OEPS is approximately equals 11.3 ns and 2.3 ns/volt.

Polymer based electro-optic (EO) modulators and other integrated optic devices have the potential to provide low cost and lightweight alternative for high-speed digital as well as analog RF links. To be truly competitive with existing technologies such as LiNbO₃, EO polymer modulators must also meet the criteria of low loss. There are two major causes of loss in EO modulators: waveguide loss (including material loss, scattering, etc.), and fiber- to-waveguide coupling (butting) loss. Various techniques can be utilized to minimize these coupling losses, however, to maintain low cost of component, we resort to the simplest possible approach which is easy to manufacture. Pigtails using standard single mode fiber produce coupling loss on the order of 3 to 5 dB/connection. In order to improve mode size matching yet maintain low drive voltage we incorporate waveguide and fibers tapers. Waveguide tapers resulted to butting losses as low as 1.5 dB/connection, whereas fiber tapers resulted to 2.5 dB/connection butting losses. Combining both techniques together, it was possible to produce 1.3 dB/connection butting loss, however, tapered waveguide devices were less sensitive to alignment tolerance than tapered fiber devices, and therefore less sensitive to environmental conditions.

The analog-to-digital (A/D) interface is generally considered to be the most critical part of any signal acquisition and processing system. Because of the difficulty in achieving high-resolution and high-speed A/D converters, this A/D interface has been and continues to be a barrier to the realization of high-speed, high-throughput systems. Recently, there has been a renewed interest in new and innovative approaches to A/D conversion, with a significant emphasis on photonic techniques. Interleaving is a common approach applied to high-speed photonic A/D conversion which reduces the wide-bandwidth input signal to one which can be converted using conventional high-speed A/D converters. The high-speed sampled input is interleaved to N individual channels with each channel operating at 1/N of the sampling rate. These channelization techniques are known to suffer from performance degradations due to channel-to-channel mismatch. Within the electronic A/D converter community, temporal oversampling and spectral noise shaping have become common practice in high-fidelity audio applications. Here, a low-resolution quantizer is embedded in a feedback architecture in an effort to reduce the quantization noise through spectral noise shaping. A large error associated with a single sample is diffused over many subsequent samples and then linear filtering techniques are applied to remove the spectrally-shaped noise thereby improving the overall SNR of the converter. The approach to wideband photonic A//D conversion described here leverages the 2-D nature of an optical architecture to extend the concept of spectral noise shaping to include 2-D spatial nosie shaping. The proposed approach uses a mode-locked laser to generate the optical sampling pulses, an interferometer to modulate the electronic analog signal onto the optical pulses, and a 2-D smart pixel hardware implementation of a distributed error diffusion neural network.

A photonic oscillator that can generate multi-tone oscillations in both the RF and optical domains is demonstrated. The oscillator consists of a simple closed optoelectronic loop comprising a CW laser source, an optical modulator, a photodetector, a low noise electrical amplifier, an electrical bandpass filter and single or multiple optical delay elements. Multi-tone oscillations with independent control of tone spacing and phase noise is demonstrated using two optical fiber delay lines as optical delay elements. The phase noise of the oscillator is systematically measured as a function of its parameters such as the photodetector current, the amplifier gain and oscillation frequency. Phase noise values better than -125 dBc/Hz at 100 kHz offset frequency are measured at various combinations of the above parameters. Multi-tone photonic oscillators using Mach-Zender or electroabsorption modulators are theoretically analyzed and experimentally demonstrated with the relative merits compared.

We report a simplified version of a mm-wave generator employing the sideband filtering technique which uses a single optical sideband filter. Instead of using a Mach- Zehnder-like fiber network to select a pair of sidebands, we employ a single Fabry-Perot fiber Bragg grating with a pair of passbands separated by the mm-wave frequency. Using a single filter, eliminated the need for polarization control and pathlength matching that was required by the former interferometer-like arrangement. We describe a 30 GHz generator design and present its mm-wave signal spectrum showing an instrument limited linewidth of approximately 20 Hz. The generation of such an extremely narrow signal spectrum from a laser with a 1 MHz linewidth demonstrates the remarkable laser frequency nosie cancellation property of the sideband filtering technique.

Radio systems and, in particular, RF data link systems are evolving toward progressively more bandwidth and higher data rates. For many military RF data link applications the data transfer requirements exceed one Gigabit per second. Airborne collectors need to transfer sensor information and other large data files to ground locations and other airborne terminals, including the rel time transfer of files. It is a challenge to the system designer to provide a system design, which meets the RF link budget requirements for a one Gigabit per second data link; and there is a corresponding challenge in the development of the terminal architecture and hardware. The utilization of photonic circuitry and devices as a part of the terminal design offers the designer some alternatives to the conventional RF hardware design within the radio. Areas of consideration for the implementation of photonic technology include Gigabit per second baseband data interfaces with fiber along with the associated clocking rates and extending these Gigabit data rates into the radio for optical processing technology; optical interconnections within the individual circuit boards in the radio; and optical backplanes to allow the transfer of not only the Gigabit per second data rates and high speed clocks but other RF signals within the radio. True time delay using photonics in phased array antennas has been demonstrated and is an alternative to the conventional phase shifter designs used in phased array antennas, and remoting of phased array antennas from the terminal electronics in the Ku and Ka frequency bands using fiber optics as the carrier to minimize the RF losses, negate the use of the conventional waveguides, and allow the terminal equipment to be located with other electronic equipment in the aircraft suitable for controlled environment, ready access, and maintenance. The various photonics design alternatives will be discussed including specific photonic design approaches. Packaging, performance, and affordability of the various design alternatives will also be discussed.

We propose a novel microwave optical filter design method employing fiber Bragg grating arrays and optimal filter design. In the proposed system, the grating arrays are used to provide different delays. Each of the grating arrays is connected with an adjustable attenuator that sets the weighting of the tap. Both negative and passive coefficients can be obtained by differential detection structure presented in the system. Tuning is realizing by changing the optical carrier wavelength to select the operating gratings. We at the first time adopt linear phase structure in the microwave filter design, which simplifies the design configuration. Optimal filter design method is creatively adopted to optimize weightings of grating arrays. It is superior to window function method and can realize the total control of filter specifications. Furthermore, experiment has been conducted to confirm above conclusions. The results show that our method solves some pivotal problems and is an effective way to design microwave optical filters.

This paper provides a comprehensive understanding of a tunable wavelength-based photonic high-speed analog-to- digital converter (ADC) approach and its anticipated performance. Analytic models, experimental data, and simulation results will provide the fundamental limits on the expected conversion speed and bit resolution.

Significant difference in temporal and spectral characteristics of THz radiation emitted by large- (1mm) and small- (5micrometers ) aperture dipole antennas fabricated on arsenic-ion-implanted GaAs and undoped semi-insulating GaAs is reported and attributed to the geometry of the antenna.

This paper discusses the principle of operation of an electrically tunable THz detector, working around 2.54 THz, integrated with a bowtie antenna. The detection is based on the idea of photon-assisted tunneling (PAT) in a double quantum well (DQW) device. The bowtie antenna is used to collect the THz radiation and feed it to the detector for processing. The Bowtie antenna geometry is integrated with the DQW device to achieve broadband characteristic, easy design, and compatibility with the detector fabrication process. The principle of operation of the detector is introduced first. Then, results of different bowtie antenna layouts are presented and discussed.

We consider physical principles of a two-phonon light scattering in crystals in a special case, when direct transitions between all the orders of scattering are allowed and, moreover, the probabilities of these transitions can be regulated electronically. The exact and closed analytical model for describing this process is presented. The feasibility of exploiting such a phenomenon for creation of an all-optical logic is substantiated. A few possibilities of arranging the optical logic gates are revealed. Fulfilling a set of logic operations is algorithmically analyzed and estimated.

A fiber grating prism consisting of a chirped grating delay line and three delay lines with thirteen discrete fiber Bragg gratings is fabricated and experimented. The system comprises a chirped grating for short time delays and three discrete Bragg grating arrays for long time delays. The introduction of the chirped grating delay line enables the system to work for beamforming at higher microwave frequencies. The experimental results of the time delay measurements agree well with the theoretical calculations. Based on the experimental data, the radiation patterns of a 4-elements array are calculated and analyzed. This fiber grating prism can provide four-channel time-delayed signals that can be fed to an antenna array to achieve beam scanning at different radiation angle. The beampointing direction of the system is determined by the grating spacing and independent of the microwave frequency. The proposed true- time delay unit with 4 X 13 phased array steerer is suitable for beamforming at frequencies up to 6 GHz.

We present theoretical and experimental results of THz-wave generation via optical rectification in periodically poled lithium niobate (PPLN) crystal. The rectified field in both frequency and time domain is calculated taking into account the divergence of the exciting optical beam. The dependence of the central frequency of the narrow-band THz-wave on of direction of field emission is investigated. The surface- emitting geometry (when the THz-wave is observed in direction perpendicular to the direction of the optical pulse propagation) reduces the damping of the THz-wave in the PPLN crystal considerably. In this geometry the measured central frequency is around 1 THz for our PPLN crystal with a poling period of (Lambda) equals 127 micrometers . The frequency is tuned by rotation of the crystal over a range of 0.65 - 1.1 THz. Typical bandwidths of 50 - 100 GHz were observed, depending on the collection angle and the number of periods involved.

In this reference processes in a high temperature superconducting (HTSC) thin film circuit with nonlinear parametric resistance and inductance are theoretically examined. The fact that the active and reactive components of the impedance of HTSC film are modulated by the law of optical signal is accepted as a basic precondition for this discussion. By the analogy with the Josephson phenomenon we give certain phases for superconducting and normal states and accept that the alternating components of the photocurrent I_Phi and photovoltage U created by the radiation can be presented as a phase difference between superconducting and normal states respectively. Based on this, the equation for the equivalent circuit of the film presents a complicated differential heterogeneous equation of second rank. As a particular state of the homogeneous equation has been investigated and the Mathieu equation has been obtained. In a result the expression for the gain in power is obtained, which depends on the parameters of the optical signal and thin film. The possibility to control the gain in power by the selection of various values of the depth of modulation of the active and reactive components of the surface impedance, the intensity of radiation and other parameters of the HTSC film is shown.

The influence of alloy composition on the noise behavior of heterostructure semiconductor devices is investigated by using a rigorous two-dimensional physical simulator. The model takes into account non-stationary transport properties and quantization effects to allow a better understanding of the carrier transport properties inside the heterostructure devices and consequently to explain the noise performance of these devices by making use of the microscopic nature of the model. As an example, the model is applied to study the effects of alloy composition and the resulting band discontinuity on the 2DEG properties and on the noise performance of Hetero-FETs at millimeter-wave frequencies, and to extract the optimum alloy composition which leads to the minimum noise figure in different frequency ranges.

Impedance characteristics of semiconductor barrier-injection transit-time diodes (BARITT) structures made of Silicon Carbide containing quantum wells in the drift region are theoretically examined. It is shown that the magnitude of the negative dynamic resistance can be increased due to trapping and escape effects of injected charge carriers in quantum wells. It is shown that the negative resistance of the BARITT structure made of different polytypes of SiC is one order of magnitude higher in absolute value in comparison with the Si structure, all other factors being equal. In the proposed structure significantly higher operation frequencies can be realized in comparison with usual BARITT'S.

Processes in a high temperature superconducting thin film circuit with nonlinear active resistance and inductances are theoretically examined. A possibility to control values of equivalent parameters of the circuit by optical radiation modulated on the intensity is shown.

The U.S. Air Force Advanced Electro-Optical System (AEOS) which includes a 941 actuator adaptive optics system on a 3.7 m telescope has recently been made available for astronomical programs. Operating at a wavelength of 750 nm, the diffraction-limited angular resolution of the system is 0".04; currently, the magnitude limit is V approximately 7 mag. At the distances of nearby open clusters, diffraction- limited images should resolve companions with separations as small as 4 - 6 AU - comparable to the Sun-Jupiter distance. The ability to study such close separations is critical, since most companions are expected to have separations in the few AU to tens of AU range. With the exceptional angular resolution of the current AEOS setup, but restricted target magnitude range, we are conducting a companion search of a large, well-defined sample of bright early-type stars in nearby open clusters and in the field. Our data set will both characterize this relatively new adaptive optics system and answer questions in binary star formation and stellar X- ray activity. We will discuss our experience using AEOS, the data analysis involved, and our initial results.

The AMOS 1.2-m telescope is being used 18 nights per month to search for Near-Earth Asteroids (NEA). Since telescope time is a very valuable resource, our goal is to use the telescope as efficiently as possible. This includes striving to maximize the utility of each observation. Since the NEAT searches are within the ecliptic, the same part of the sky as geosynchronous satellites, these search fields contain satellite tracks as well as asteroids. We present the results of simulations of the number of satellites that should be found within the field of view based upon the field centers and times for several nights. We have also examined the NEAT images for geosynchronous objects and present these results. During the remaining nights each month, we use the NEAT camera to obtain observations of deep-space satellites. This data will also be presented. We also present the results of simulations for optimizing search strategies for deep-space objects using NEAT and other AMOS sensors.

The Air Force Research Laboratory Directed Energy Directorate has collected and analyzed photometric data using the SILC filters for Space Object Identification (SOI) of geosynchronous (GEO) satellites. This set of filters was designed as part of the Space Battlelab initiative, SOI In Living Color (SILC). The photometric data of geosynchronous satellites were taken using a charge-coupled device (CCD) on the 24-inch Ritchey-Chretien telescope at Capilla Peak Observatory of the University of New Mexico. The objects under discussion are satellites with similar and dissimilar bus structures in a cluster. The data and analysis results to date are discussed.

GLINT (Geo Light Imaging National Testbed) is a program to image geo-synchronous satellites using Fourier telescopy. The design of the GLINT system requires knowledge of the reflectance properties of the satellites in certain specific wavelength ranges. Calibrated measurements of satellite brightness due to solar illumination can be made with a telescope. This report details such measurements and the data processing necessary to yield curves of normalized satellite return versus phase angle in given wavelength ranges. These measurements can be used to check the accuracy of satellite reflectivity models.

The Spica and Kala spectrographs located at the rear blanchard of the 1.6 m telescope and the trunnion port of the AEOS 3.67 m telescope, respectively, have been utilized by several DoD and NASA agencies requiring relatively high resolution spectroscopic observations. The sensors are located at the Air Force Maui Optical Station (AMOS), Haleakala, Maui. Three R&D programs utilizing these instruments will be described. The AFRL propulsion directorate's demonstration called the electric propulsion space experiment (ESEX) utilized Spica to evaluate high powered arc-jet thruster firings from the ARGOS satellite. AFRL Det. 15 and Air Force Battlelab sponsored a project called SILC to explore the advantages of applying spectroscopic analysis to help reduce satellite cross- tagging and augment Satellite Object Identification (SOI). Thirdly, the NASA Johnson Space Center Space Debris Program obtained spectroscopic data on Low Earth Orbit (LEO) targets to help determine albedo and material composition of space debris.

This is the fourth paper in a continuing study on the standard photometric signatures of geosynchronous earth orbit, GEO, communication satellites. Here we present the results of photopolarimetric measurements taken at the Air Force Research Laboratory Directed Energy Directorate's Starfire Optical Range.. These limited set of measurements were conducted in order to determine if GEO communication satellites have measurable polarization, and is so, if there are differences between the satellites. Measurable polarization was detected. This polarization was seen throughout the night varying smoothly from a minimum at local midnight, approximately 10%, to a maximum at dawn of approximately 40%. This is distinctly different from the radiometric signals which is a maximum at local midnight and decreases toward dawn and dusk. This polarization is found to be distinct for each GEO satellite bus. Also, serendipidously it is found that when measuring through cirrus clouds, the photopolarimetric signal is not lost, although it is changed.

The purpose of this research is to improve the knowledge of the physical properties of orbital debris, specifically the material type. Combining the use of the fast-tracking United States Air Force Research Laboratory (AFRL) telescopes with a common astronomical technique, spectroscopy, and NASA resources was a natural step toward determining the material type of orbiting objects remotely. Currently operating at the AFRL Maui Optical Site (AMOS) is a 1.6-meter telescope designed to track fast moving objects like those found in lower Earth orbit (LEO). Using the spectral range of 0.4 - 0.9 microns (4000 - 9000 angstroms), researchers can separate materials into classification ranges. Within the above range, aluminum, paints, plastics, and other metals have different absorption features as well as slopes in their respective spectra. The spectrograph used on this telescope yields a three-angstrom resolution; large enough to see smaller features mentioned and thus determine the material type of the object. The results of the NASA AMOS Spectral Study (NASS) are presented herein.

The Air Force Research Laboratory (AFRL) has initiated the LARRA (Laser Radar for Recognition and Assessment) program to investigate the use of laser radar signatures for identifying and determining the pose of satellites. The algorithms will use the 1-D range-amplitude information in the returns. To determine the identification and/or pose, the field returns will be correlated over a library of simulated returns. It is critical that the algorithms that produce the simulated returns do so accurately. AFRL has previously sponsored the development of TASAT (Time-Domain Analysis Simulation for Advanced Tracking), which has the capability to simulate the returns from satellites, to include the effects of atmospheric propagation. In this paper, we describe the modification and verification of TASAT using returns from mock satellites placed on the ground at a distance of 1 km from the laser. The laser is a mode-locked CO₂ (10.6(mu) ) and the FWHM of the micropulses is 1.4 ns.

A methodology for analyzing an imaging sensor's ability to assess target properties is developed. By applying Cramer- Rao covariance analysis to a statistical model relating the sensor measurements to the target, a bound on the accuracy with which target properties can be estimated can be calculated. Such calculations are important in understanding how a sensor's design effects its performance for a given assessment task, and in performing feasibility studies or trade studied between sensor designs and sensing modalities. A novel numerical model relating a sensor's measurements to a target's three-dimensional geometry is developed in order to overcome difficulties in accurately performing the required numerical computations. An example use of the approach is presented in which the influence of viewing perspective on orientation accuracy limits is analyzed. The example is also used to examine the potential for improving the accuracy bound by fusing multi-perspective data.

OSSim (Optical System Simulation) is a simulation toolbox of optical and processing components. By using full wave-optics in the time-domain, OSSim simulates diffractive effects and control loop interactions missed by simpler analyses. OSSim also models the atmosphere, with user customizable turbulence strength, wind, and slew. This paper first presents 2 introductory examples: a simple 2-lens imaging system and a simple tilt-control system. Then it presents a simulation of the 3.67-meter AEOS (Advanced Electro-Optics System) telescope on Maui. The OSSim simulation agrees well with the AEOS experimental results.

The uses of all classes of space systems both military and commercial are rapidly increasing. In turn the surveillance of these systems and their activities are becoming of increasing importance to the uses of these systems. Unfortunately, the present surveillance infrastructure is not keeping pace with the demand for information and new infrastructure and capabilities are needed. As a result new technologies are being developed to support the need. These technologies can be used by both the military and commercial use and thus 'dual-use technologies' have come into reality. The background conditions, current research directions, and their dual use applications are briefly examined.

How to obtain sharp images when viewing through a turbid medium is a problem that arises in a number of applications, including optical biomedical imaging and optical surveillance in the presence of clouds. The main problem with this type of imagery is that it is difficult to accurately characterize the turbid medium sufficiently well to generate a point spread function that can be used to deconvolve the blurred data (and thus increase the resolution). We discuss the use of blind deconvolution as a means of estimating both the blur-free target and the system point spread function. We compare restorations obtained using a non-linear blind deconvolution algorithm with those obtained using a linear backpropagation algorithm. Preliminary results indicate that the blind deconvolution algorithm produces the more visually pleasing restorations. Moreover, it does so without requiring any prior knowledge of the characteristics of the turbid medium, or of what the blur-free target should look like: an important advance over the backpropagation algorithm.

Space surveillance systems have recently been developed that exploit high order adaptive optics systems to take diffraction limited images in visible light on 4 meter class telescopes. Most astronomical targets are faint, thus driving astronomical AO systems towards larger subapertures, and thus longer observing wavelengths for diffraction limited imaging at moderate Strehl ratio. There is, however, a particular niche that can be exploited by turning these visible light space surveillance systems to astronomical use at infrared wavelengths. At the longer wavelengths, the Strehl ratio rises dramatically, thus placing more light into the diffracted Airy pattern compared to the atmospheric halo. A Lyot coronagraph can be used to suppress the diffracted light from an on axis star, and observe faint companions and debris disks around nearby, bright stars. These very high contrast objects can only be observed with much higher order adaptive optics systems than are presently available to the astronomical community. We describe simulations of high order adaptive optics coronagraphs, and outline a project to deploy an astronomical coronagraph at the Air Force AEOS facility at the Maui Space Surveillance System.

The Air Force Research Laboratory/Directed Energy Directorate (AFRL/DE) via the ALVA (Applications of Lidars for Vehicles with Analysis) program installed in late 2000 a wideband, 12 J 15 Hz CO₂ laser radar (ladar) on the 3.67 meter aperture AEOS (Advanced Electro-Optics System) telescope. This system is part of the Maui Space Surveillance System (MSSS), on the summit of Haleakala, Maui, HI. This ladar adopts the technology successfully demonstrated by the first generation HI-CLASS (High Performance CO₂) Ladar Surveillance Sensor) operating on the nearby 0.6 meter aperture Laser Beam Director (LBD) and developed under the Field Ladar Demonstration program, jointly sponsored by AFRL/DE and the Army's Space and Missile Defense Command. The moderate power (approximately 180 watts) HI-CLASS/AEOS system generates multiple, coherent waveforms for precision satellite tracking and characterization of space objects for 1 m² targets at ranges out to 10,000 km. This system also will be used to track space objects smaller than30 cm at ranges to 2,000 km. A third application of this system is to provide data for developing satellite identification, characterization, health and status techniques. This paper will discuss the operating characteristics and innovative features of the new system. The paper will also review recent results in support of AF needs, demonstrations, experiments, as well as planned activities that directly support applications in the DoD, scientific, and commercial arenas.

Two optical observation campaigns during 1999 and 2000 were conducted to monitor the optical characteristics of the Proton 4 ullage motor population located in mainly geostationary transfer orbits about the earth. This study attempted to physically characterize the intact ullage motors and those that have exploded in space using conventional large aperture binocular optics and 3rd generation image intensified video (commonly employed for non-military astronomical observation) in order to determine if optical signatures could reveal evidence as to which motors might be candidates for explosion. We find that some motors do not spin down through normally expected damping; explosions do not consistently result in the catastrophic destruction of the parent body involved; accelerations in the rotation rates of some intact ullage motors have been discovered; some fragmentations do not result in an increase in rotation of the surviving parent as might be expected. No conclusive evidence was obtained that could lead to positively identifying precursors for future explosion. Yet, through statistical methodology we determined that there is a strong probability of additional explosions until the majority of motors launched between 1982 and 1996 decay from orbit. We estimate the probability of at least one of 46 ullage motors currently in orbit exploding before decay is greater than 0.9999. The threat of explosions is not expected to be mitigated by natural decay of all candidate exploders for at least 50 years.