Portable Performance

Newly developed handheld gamma-ray spectrometers based on cadmium zinc telluride (CdZnTe) detectors are currently being used to detect and identify radioisotopes in a variety of security and defense applications. Compound semiconductors such as CdZnTe have been developed as room-temperature alternatives to more commonly used Ge,¹ which must be cooled to liquid nitrogen temperatures. In general, though, such compound materials suffer from reduced spectral performance as a result of poor charge transport characteristics. By developing a coplanar grid (CPG) electrode structure, researchers have overcome these limitations and produced a detector suitable for handheld devices.²

Like most other compound semiconductors, CdZnTe exhibits relatively high electron-collection efficiency, but low hole-collection efficiency. This difference is caused by intrinsically low hole-transport characteristics and a high concentration of hole-trapping defects within the crystal. Low hole-collection efficiency causes incomplete charge collection in conventional planar-electrode detectors, which in turn degrades spectral performance. Poor hole-transport results in a partial signal derived primarily from electron movement, and the signal amplitude therefore becomes depth dependent. High-energy gamma-rays interact throughout the detector volume, so position-dependent signal variations of this kind are the main source of energy-resolution degradation in bipolar gamma-ray detectors.

In contrast, the CPG electrode configuration provides a unipolar sensing design that is solely dependent on electron transport, correcting the problem of position-dependent signal variation without requiring complex or power-hungry electronics. A CPG consists of narrow strip electrodes deposited on the detector surface that are alternately connected to give two independent sets of grid electrodes. An electric field is created within the detector by maintaining both sets of grid electrodes at about the same potential and applying a high voltage potential to the planar electrode.

Using a weighting potential, V_w, the signal induced on an electrode by an electron in motion can be determined by the Ramo formula Δq = e ΔV_w, where Δq is the incremental charge induced at the selected electrode, e is the electron charge, and ΔV_w is the change in the weighting potential over the path of the electron. The weighting potential represents the potential that would exist if the chosen electrode were held at unit potential, and all other electrodes were held at zero. The greatest change in the weighting potential of a coplanar grid takes place very near the grid structure. As a result, the induced charge on the electrode increases rapidly in this region.

An electron originating near the planar cathode and ultimately collected at electrode A will induce an equal charge at A and B during most of its journey. As the electron approaches its final destination, however, the induced charge at A (q_A) will rise and eventually reach the value e (once the charge is collected), while the charge at B (q_B) will fall to zero. The difference signal q_A - q_B is solely dependent on electron collection and shows no response to charge-carrier motion across most of the detector; as a result, charge carriers generated anywhere within the detector will generate a full-amplitude signal so long as all the electrons are collected—regardless of whether or not any holes are collected. Over 99% of the difference signal is generated within a distance from the coplanar grids equal to twice the electrode strip pitch (twice the center-to-center distance between adjacent strip electrodes). To achieve unipolar charge sensing over the largest possible portion of the detector volume, then, requires the smallest possible strip pitch compared to the thickness of the detector.

If we maintain the coplanar grids at the same potential, two problems arise. First, the difference signal can have either polarity. Second, electrons from a single radiation-absorption event may be collected at both grids, thereby reducing signal amplitude. To avoid these problems, the coplanar electrodes are always maintained at slightly different potentials during operation. The size of the potential difference depends on the applied potential across the detector and the ratio of the strip pitch to detector thickness.

In the handheld unit, charge-sensing preamplifiers and signal-subtraction circuitry provide analog output from the coplanar CdZnTe detector. An analog-to-digital converter digitizes the output and generates a histogram of the energy spectrum. Taking a second derivative of the histogram reveals the peaks of the spectrum, and a fuzzy-logic algorithm subsequently performs peak-shape analysis. Through software stored in the handheld unit, these peaks are compared to a library of known isotopes and a confidence number is applied according to the possibility that an isotope is present.

The coplanar grid electrode structure eliminates signal dependence on hole transport, producing a high-resolution, room-temperature gamma-ray device. This technique vastly improves the resolution of CdZnTe, and its simple low-power electronics make it ideal for use in portable, battery-powered units. oe

1. C. Szeles, et al., "Development of the High-pressure Electro-Dynamic Gradient Crystal Growth Technology for Semi-insulating CdZnTe Growth for Radiation Detector Applications," J. Electron. Mater., accepted for publication.

2. P. Luke, IEEE Trans. Nucl. Sci. 42, 207 (1995).

In today's world, countries require surveillance systems that keep pace with expanding trade while protecting them from the threats of terrorist attack, illegal immigration, illegal drugs, and other contraband. Official ports of entry are sites of continuous, 24/7 activity; in the United States, alone, approximately 7,500 foreign-flag ships make 51,000 calls in ports annually.

The agencies charged with port security face no small task in addressing the many variables present in highly dynamic port settings. Video surveillance technologies are a useful first line of defense for monitoring port activity. Traditional technologies lack the flexibility to accommodate the fluctuations in lighting due to weather conditions and time of day that are inherent to these environments, however.

New imaging technologies can mitigate many of these issues, providing clear surveillance images in a digital format compatible with powerful digital signal processing applications. Smart surveillance solutions require networked surveillance cameras and digital video recorders, combined with advanced recognition and tracking applications. The resultant automated monitoring solutions free personnel from continually monitoring the cameras, allowing them to spend their time on physical inspections.

A key enabler of effective surveillance automation is the deployment of wide dynamic range (WDR) security cameras. Port lighting varies widely, with the position of the sun and shadows changing throughout the day. WDR cameras provide the ability to clearly see activity in both the bright regions and dark regions of a scene, detecting movement in the shadows, for example. One new WDR technology offers the ability to adjust for lighting conditions by controlling exposure at the pixel level, providing extremely high-quality images even in scenes with both very bright and dark areas. This solution is based on a true digital pixel that allows nondestructive sampling of each pixel many times in each video capture field, which then gets reconstructed into images; the method exposes each image for the optimal time. Although the initial units are designed for visible-light wavelengths, they can also be used effectively in an IR environment.

Technological advances in surveillance and security will have a tremendous effect on ports and border security. Deploying advanced surveillance technologies is a necessary first step in implementing many next-generation security solutions that are now becoming available.

Lee Hirsh is vice president of product management and strategy at Pixim Inc., Mountain View, CA.