Deflection microwave amplifier with field-emitter arrays

Summary form only given. We have developed a new method for fitting the current-voltage (I-V) characteristics of the field emission arrays (FEAs) that may be operated at high current density. In recent experiments of FEAs, the Fowler-Nordheim (FN) plots show that the experimental results are apart from the conventional FN fits. Based on the past studies, it is taken for granted that the cause of the current saturation by space charge effects can be excluded at these comparatively lower current density levels. In this work, the FN equation is modified with a global effective work function which is an ensemble average of the local effective work function of individual tips. Within the framework of the effective work function approximation, a self-consistent FN equation including the space charge effects of the field emission electrons has been verified by the 2D finite-difference time-domain particle-in-cell simulation and compared with the experiments performed at Stanford Research Institute, giving excellent agreement. The self-consistent model is demonstrated to be a good fit to the I-V characteristics of high current density FEAs. This also indicates that the space charge effects play an essential role in the saturation behavior of FEAs.

Field Emitter Arrays (FEA) are an attractive candidate to replace thermionic and photo-cathodes as a source of a high quality electron beam, reducing cathode power consumption and increasing reliability. For some applications (accelerators, microwave amplifiers) it is necessary to modulate or produce electron pulses with sub-nanosecond time structure. Sine wave grid modulation of an FEA cathode up to few GHz has already been demonstrated [1, 2]. Sub-microsecond pulsed electron emission was also reported by S. Leemann et al. [3]. Our interest was to explore the limitations of short electron pulses generation based on electrically gated FEAs. Since field emission is an instantaneous process the main modulation speed limitation is due to the finite FEA gate capacitance. Using short current pulses to control the gate potential we demonstrated that FEAs are capable of generating single sub-nanosecond electron pulses. The practical implementation of the method and its limitations are discussed.

Sharp, uniform, stable, and environment‐hard field‐emitter arrays (FEAs) have been fabricated by Transfer‐Mold fabrication method. These FEAs have potential for realizing stable and low operation voltage field‐emission displays and field‐emission lamps that are resistant to highly oxidizing atmospheres during the fabrication process. The extremely sharp and uniform Transfer‐Mold FEAs, which have the average tip radius of 2.6‐3.5 nm, have been developed by using the Transfer‐Mold fabrication method. These tip shapes including height are extremely uniform being different from CNTs' nonuniform tip radii, height and pitch. Also, these tip radii are similar to the diameter of the singlewalled CNTs, 0.4–6 nm, or less than the diameter of the multiwalled CNTs, 5–10 nm. The low operation voltage Transfer‐Mold FEAs has turn‐on fields of 17.5–21.8 V/μm at the short distance between anode and emitter, less than 30 μm. However, other conventional FEAs such as Spindt‐type FEAs, Gray‐type FEAs and CNTs FEAs have the approximated turn‐on fields of 50–600 V/μm. Therefore, those values of the turn‐on voltages are lower than those of other conventional FEAs and can be applicable for vacuum nanoelectronic devices such as FEDs and FELs. The current fluctuations of FEAs without resistive layer are the lowest value ever reported, ±1.6%, compared with 5‐several hundred % of the conventional FEAs. Also, the Transfer‐Mold Mo FEAs having the base lengths of 36–1570 nm without resistive layer, exhibited extremely stable current fluctuations of as low as ±1.6‐3.3% due to the extremely uniform geometric factors. Moreover, those of the environment‐hard Transfer‐Mold FEAs even under the oxygen radical or plasma treatments are as low as ±5.0–7.5%. However, for conventional FEAs with and without resistive layers, those are usually 5–100% and more than 100%, respectively. Therefore, the extremely sharp, uniform, stable, and environment‐hard Transfer‐Mold FEAs are useful for the stable and reliable vacuum nanodevices such as FEDs and FELs to a great extent.

A fabrication method for field emitter array of carbon nanotubes with improved carbon nanotube rooting

The success of a field emitter array (FEA) as a high frequency or high brightness source is critically dependent on the FEA's maximum total current and current density. To date, damage and shorts typically limit the maximum current obtained from high current density FEAs (those not containing large resistances in series with each tip). Because it is not clear what mechanism or what per-tip current initiates the damage, it is difficult to predict maximum FEA performance or engineer optimal FEA structures and circuits. Joule and Nottingham heating have been suggested as damage-causing mechanisms. We have measured the total potential drop (and thus energy loss) in Si FEAs by measuring the energy distribution of the beam.

The electron source generating density modulated electron beam (a train of electron beam pulses) at high frequency directly from the cathode surface has attracted significant interest since they could be used in a wide variety of vacuum electronics devices, such as compact microwave sources [1-4], ultrafast electron microscopy [5], or electron beam lithography [6]. In particular, bunching the electron beam at the cathode surface can not only dramatically reduce dimensions and weight through the elimination of the pre-modulation circuit, but can also provide extremely high efficiency and improve power in electron beam devices due to coherence effects. The high-frequency modulation of the gate potential of field emitter arrays (FEAs) has been proposed and studied for microwave tube applications [1]. However, challenges resulting from parasitic capacitances have still remained [7]. An alternative is producing electron bunches by optical means. The emission current from p-type semiconductor field emitter arrays, unlike metals, is limited by the supply of minority carriers (electrons) in the depletion region in the high field, and is, therefore, highly sensitive to the additional supply of photoexcited carriers. Several studies have revealed the transient photoresponse of p-type Si-FEAs was limited by diffusion of photogenerated electrons outside the surface depletion region in the tip [8-10]. To generate a high frequency modulated electron beam, we need to design the emitter structure which eliminates photoexcitation of carriers outside the depletion region. In this paper, we descuss the photoresponse of volcano-structured p-type Si-FEAs fabricated by etch-back technique, where the gate metal effectively prevents photoexitation of carriers outside the depletion region, under laser light illumination. Our Si-FEA devices tested here consist of 100 emitter tips, a gate aperture of less than a half micrometer in diameter and insulating layer of 200 nm thickness, as shown in Fig.1. In a proof-of-principle experiment, the FEA device was optically modulated from the front side. Figure 2 shows the typical current-voltage characteristics of our FEA device with and without laser pulse illumination of a wavelength of 405-nm. This result clearly suggests our device is a highly sensitive photocathode. In the dark, the emission current saturation by limitation in supply of electrons at high voltages is clearly displayed and the irradiation enhanced the emission current. Figure 3 shows the response of optically modulated emission current from our FEA device and a reference commercial photodetector. The rise and fall times of each pulse response are 5 ns, respectively. These values are similar to the minimum rise time of the current-voltage conversion amplifier used in this experiment, so the measured rise and fall times are determined by the response time of the current amplifier and the real rise and fall times of photoresponse in the device are at least less than 5 ns. A delayed slow response in the falling phase is not included in the component of the transient signal. Although optimization of the FEA structure is underway, it is experimentally demonstrated that the FEA photocathode is a structure suitable for minimizing the photogeneration of electrons outside the depletion layer, giving a high-speed modulated electron beam [11].

A design of a sheet-beam electron-optical system (EOS) for a microfabricated $W$ -band traveling-wave tube (TWT) is presented. The proposed TWT is based on a rectangular field-emitter array (FEA) and a planar helix slow-wave structure with straight-edge connections (PH-SEC). The PH-SEC is designed to work with a 5-kV and 10-mA sheet electron beam. The particle-in-cell simulation results show that the TWT can give 3.7-W peak power at 110 GHz with a 3-dB bandwidth of 60%. The EOS consists of a novel beam-forming electrode, a novel anode, a magnetic circuit to provide solenoidal focusing magnetic field, and a depressed collector. Except for the magnetic circuit, all components are compatible with microfabrication. Moreover, the anode and the collector can be fabricated together with the PH-SEC using the same process steps. Using an FEA with the dimensions of $1000 \times 500\mu \text{m}^{2}$ , a beam of cross section $160 \times 40\mu \text{m}^{2}$ is generated and under ideal conditions 100% transmission is achieved through a 25-mm-long drift tunnel of cross section $240 \times 100\mu \text{m}^{2}$ . The effects of misalignments of the FEA, anode, and the magnetic circuit are investigated, and the limits on misalignments are determined to achieve satisfactory beam transmission.

The authors have developed a highly sensitive, compact image sensor comprising a field emitter array (FEA) and a high-gain avalanche rushing amorphous photoconductor (HARP) target with the ultimate aim of developing an ultrahigh sensitivity, compact, high-definition television camera. Double-gated field emitters have the advantage of a compact electron beam focusing system; however, image intensities reproduced by a sensor with the double-gated, Spindt-type field emitter array with the focusing electrode stacked 1.5 μm above the gate electrode were nonuniform owing to low electron beam current. The minimum required electron beam current extracted from the double-gated field emitter array is considered for possible use with the sensor. Furthermore, a suitable field emitter array pitch to balance the electron beam current and the electrostatic-focusing lens strength was simulated. For the sensor's design guidelines, a field emitter array pitch of approximately 3 μm would be reasonable in the case of employing the volcano-structured, double-gated, Spindt-type field emitter array with the focusing electrode placed 0.2 μm below the gate electrode hole. Based on simulation results, an image sensor with the volcano-structured, Spindt-type, 3.1-μm-pitch FEA with the focusing electrode placed 0.2 μm below the gate electrode hole was fabricated. It was confirmed that the minimum electron beam current extracted from the volcano-structured FEA was approximately 2 μA/pix, and the sensor could obtain images with improved image intensity uniformity by utilizing the electrostatic-focusing effect. These results indicate that the volcano-structured, double-gated, Spindt-type FEA is potentially suitable for high-definition television FEA-HARP image sensors.

Summary form only given, as follows. Summary form only given. Calabazas Creek Research, Inc is funded by the US Department of Energy (DOE), the National Aeronautics and Space Administration (NASA), and the US Air Force (USAF) to investigate applications for micro-electromechanical systems (MEMS) and field emission array (FEA) cathodes for vacuum electron devices. These include high power klystrons, traveling wave tubes (TWT's), and backward wave oscillators (BWOs). This presentation will describe how new advances in MEMS and FEA technology are enabling development of a new generation of RF devices.

An indirect flat panel imager (FPI) with programmable avalanche gain and field emitter array (FEA) readout is being investigated for low-dose and high resolution x-ray imaging. It is made by optically coupling a structured x-ray scintillator, e.g., thallium (Tl) doped cesium iodide (CsI), to an amorphous selenium (a-Se) avalanche photoconductor called high-gain avalanche rushing amorphous photoconductor (HARP). The charge image created by the scintillator/HARP (SHARP) combination is read out by the electron beams emitted from the FEA. The proposed detector is called scintillator avalanche photoconductor with high resolution emitter readout (SAPHIRE). The programmable avalanche gain of HARP can improve the low dose performance of indirect FPI while the FEA can be made with pixel sizes down to 50 microm. Because of the avalanche gain, a high resolution type of CsI (Tl), which has not been widely used in indirect FPI due to its lower light output, can be used to improve the high spatial frequency performance. The purpose of the present article is to investigate the factors affecting the spatial resolution of SAPHIRE. Since the resolution performance of the SHARP combination has been well studied, the focus of the present work is on the inherent resolution of the FEA readout method. The lateral spread of the electron beam emitted from a 50 microm x 50 microm pixel FEA was investigated with two different electron-optical designs: mesh-electrode-only and electrostatic focusing. Our results showed that electrostatic focusing can limit the lateral spread of electron beams to within the pixel size of down to 50 microm. Since electrostatic focusing is essentially independent of signal intensity, it will provide excellent spatial uniformity.

A silicon field emitter array (FEA) was fabricated on n‐type, low‐resistance (5Ω·cm) Si(100) wafer by anisotropic etching (dipping in 40 wt.% KOH solutions). To increase the thermochemical durability of the tip, the long‐term stability and the current density, Ti (30 nm) film was deposited with evaporation and the silicide was formed by annealing at 650°C. The C 54 –TiSi 2 phase of the sample was observed by x‐ray diffraction analysis. The emission characteristics of the titanium silicide FEA were compared with those of an Si FEA under high vacuum conditions of ∼10 −7 Torr. The turn‐on voltage of the titanium silicide FEA is ∼1500 V whereas that of the Si FEA is found to be 1950 V. In addition, the emission stability of the titanium silicide FEA proved to be much better than that of the Si FEA. Copyright © 2004 John Wiley & Sons, Ltd.

Of all applications for which field emitter arrays (FEAs) are being designed, RF vacuum microelectronics is the most technically challenging application. The current density is typically three orders of magnitude larger than that required for displays (which require < 0.1 A/cm2). Due to their high current density capabilities and instant turn-on, FEAs may be a promising alternative to thermionic emitters for use in Inductive Output Amplifiers (IOAs). An analytical model of a field emitter is used to estimate Fowler Nordheim A and B parameters, effective resistance and capacitance of the array under several GHz modulation, signal propogation lengths, total current and current density, and effects of emitter non-uniformity on the basis of array geometry and materials. Estimates of inductance, resistance, and capacitance are made to estimate the drive power required to produce a bunched electron beam for Inductive Output Amplifier applications. An electronic efficiency of 32% with 15 dB gain may be possible from an array producing 260 mA peak, 71 mA average, current at 10 GHz using a TWT helix 1.51 cm long.

High field emission (FE) current density from carbon nanotube (CNT) arrays grown on lithographically patterned silicon substrates is reported. A typical patterned field emitter array consists of bundles of nanotubes separated by a fixed gap and spread over the entire emission area. Emission performance from such an array having randomly oriented nanotube growth within each bundle is reported for different bundle sizes and separations. One typical sample with aligned CNTs within the bundle is also examined for comparison. It is seen that the current density from an array having random nanotube growth within the bundles is appreciably higher as compared to its aligned counterpart. The influence of structure on FE current densities as revealed by Raman spectroscopy is also seen. It is also observed that current density depends on edge length and increases with the same for all samples under study. Highest current density of -100 mA cm(-2) at an applied field of 5 V/μm is achieved from the random growth patterned sample with a bundle size of 2 μm and spacing of 4 μm between the bundles.

The emission barrier height of p-type semiconducting diamond has been estimated relatively by comparing the field emission property of a diamond field emitter array (FEA) and a molybdenum (Mo) FEA having the same geometry. Diamond FEA showed lower turn-on voltage than Mo FEA in the emission current-anode voltage characteristics. The slope ratio of diamond FEA to Mo FEA in Fowler-Nordheim (F-N) plots yielded the emission barrier height ratio of the diamond to Mo as 0.44 ±0.06. From the obtained ratio, the emission barrier height of the diamond can be estimated to be 1.87 ±0.26 eV when the work function of Mo (4.24 eV) is assumed.

The realization problem of low-cost, reliable and long-lived field emitter arrays (FEAs) is a most complicated one in the development of the element base for vacuum microelectronics. Technological processes for the fabrication of the FPE matrix FEAs use an electrodeless gas microwave (MW) discharge with electron-cyclotron resonance (ECR). The sets where a highly ionized (5-7%) plasma of ECR MW gas discharge, employed as suitable both for low-energy (20-50 eV) cleaning and precision dry etching with controlled selectivity and anisotropy and low-temperature deposition of high-quality defect-free thin dielectric, semiconducting and conducting films from various materials with specified electrophysical properties. At present, certain progress has been achieved toward the development of small-aperture ECR sources of MW plasma for a processed wafer area of 5 to 10 cm/sup 2/. However, the generation of large MW plasma volumes and large diameter (300-500 cm/sup 2/ and more) ion beams with high current density and processing uniformity which are necessary to fabricate high-dimension flat screens, is still an urgent problem.