Carbon-nanotube cathodes for miniature analytical instruments

Miniaturization enhances the scientific returns from space exploration by allowing more instruments to fit in the limited payload of each mission. Efficient electron sources are key components of miniature analytical instruments that perform in situ analysis of planetary surfaces,¹ radiation-insensitive vacuum microelectronic devices, and miniature vacuum-tube sources for high-frequency applications.² But traditional thermionic cathodes are ill-suited for miniaturization because of their high-temperature operation, high power consumption, and intrinsic bulkiness. In contrast, cold cathodes that exploit field emission (in which sharp metallic tips emit electrons due to an applied field³) are well suited to these applications.

State-of-the-art cold cathodes, made from micromachined tips^4,5 of low work-function metals, produce stable currents but degrade when operated at pressures above the 10⁻⁸ to 10⁻⁹Torr of ultra-high vacuum. A new class of robust field emitters was identified when carbon nanotubes (CNTs) were discovered.⁶ CNTs offer tips with nanometer-range diameters, each capable of emitting tens of nanoamperes with low threshold fields⁷ in poorer vacuums.

When multiple CNTs are used, however, the current does not scale up proportionally because of the “hot-spot” effect⁸ (see Figure 1). Various CNT arrangements have been reported to avoid this problem,^8–12 but producing tens to hundreds of A/cm² is still an active area of research.

Figure 1. In a sample covered with a mat of carbon nanotubes (CNTs), electron are emitted from hot spots instead of uniformly across the sample.

We found that CNTs arranged as arrays of 1-2μm-diameter bundles spaced 5μm apart (see Figure 2) give very high emission-current densities,¹³ in the range of tens of A/cm². The reason for this is not yet clearly understood, but real-time scanning electron microscope (SEM) observation revealed that free ends and outliers in each bundle rearrange themselves under an applied field,¹⁴ which may be causing efficient field emission. We typically use 10 to 20 μm-tall bundles in our electron sources.

Figure 2. Scanning electron microscope (SEM) image of optimum CNT bundle arrays (1μm diameter, spaced 5μm edge-to-edge). The inset shows a magnified view of one of the bundles, containing hundreds of 20nm-diameter nanotubes.

In our research, we optimized the CNT growth process and the architecture to routinely produce 10 to 25A/cm² at applied fields of 5 to 10 V/μm.¹⁵ These tests were conducted using CNT bundle array samples that occupied a circular area of 100 μm diameter: see Figure 3(a). Repeatability was achieved over multiple samples, as shown in Figure 3(b). All emission tests were conducted in a vacuum of ∼10⁻⁵Torr. These CNT bundles have also been successfully tested for emission when packaged inside hermetically−sealed miniature vacuum cavities.

Figure 3. (a) SEM micrograph of 100-μm-diameter sample with 1-μm diameter bundles. (b) Field-emission data from various samples producing 10 to 15 A/cm² at fields ranging from 4 to 9 V/μm.

Figure 4. SEM micrograph of monolithically gate-integrated CNT-bundle cathode using double-silicon-on-insulator (SOI) structure. The two insets show different magnifications of a cell.

Our double-silicon-on-insulator (double-SOI) process monolithically integrates multiple electrodes with CNT bundle arrays.¹⁶ This can be extended to more than two electrodes by stacking. Figure 4 shows SEM micrographs of such gate-integrated bundles. We are working on both multi- as well as single-bundle designs targeted for different applications.

CNT bundles have great promise as cold cathodes that deliver high current densities even in modest vacuums. Lifetime and reliability are currently being studied. Encouraged by preliminary results, we are developing high-performance miniature X-ray tubes for future mineralogy instruments. We are also integrating these cathodes into miniature mass spectrometers for in situ gas analysis and astronaut health monitoring.

This research was carried out at the Jet Propulsion Laboratory (JPL), California Institute of Technology, under a contract with National Aeronautics and Space Administration (NASA). This work was funded by JPL's Research and Technology Development Fund and by Defense Advanced Research Project Agency seedling fund (Task Order NMO#715839 under NAS7-03001).