GaAs Charge-coupled Device Research Articles

Novel combined through-wafer-groove fabrication approach and its application in wafer level packaging of GaAs CCD

This paper presents a novel combined through-wafer-groove fabrication approach, which is applied to the wafer level packaging (WLP) of GaAs charge coupled devices (CCD) for electrical interconnection. The combined methodology includes mechanical dicing of the groove and wet chemical etching for polishing. The parameters of the mechanical dicing are researched, including feed speed, dicing directions of the wafer, and cutting depth, to minimize the chipping. Two kinds of chemical solution are tried, and the results are discussed. Besides, the etch rate is measured, which provides a guideline for the process design. Finally, GaAs-CCD WLP sample is achieved and the electrical properties are tested to validate the feasibility of this fabrication approach. This methodology is featured by low cost, low process temperature, and good process uniformity.

Light absorption in a resistive-gate GaAs charge-coupled device

A 128 pixel uniphase resistive-gate GaAs charge-coupled device was fabricated at the TRI University Meson Facility to study its optical absorption properties under monochromatic illumination in the visible and infrared range. It was found that when the incident light energy exceeded that of the band gap, there was a progressive reduction in the charge collection efficiency. This was attributed to a carrier recombination effect at the cermet–GaAs interface. A simple recombination model was proposed to explain the observation. A comparison between theory and measurement suggested a match between the two would require a minority carrier lifetime on the order of nanoseconds and an interface recombination velocity in the neighborhood of 107−8 cm/s. Such a high recombination velocity suggested a surface state density as high as the density of the surface atoms. Extrinsic absorption measurement indicated the presence of additional carrier generation in the wavelength range between 1.3 and 2.0 μm.

Equivalent circuit for a GaAs CCD

An equivalent circuit model was proposed for the charge transfer mechanisms in a GaAs charge coupled device (CCD). Realistic simulations were carried out to assess the model at different transfer times and for different sizes of the charge packet. A comparison was also made with an analytical model and the results were quite close. Simulations of the input and the output stages of the device were also examined. The key advantages of the equivalent circuit model appeared to be substantial reduction in the computing time in addition to its primary use in a circuit simulator.

Uniphase operation of a GaAs resistive gate charge-coupled device

The operation of a GaAs charge-coupled device (CCD) within the UHF band is a technically challenging problem. The GaAs CCDs that were reported previously in the literature were typically operated with multiphase clocks, with the majority operated using clocks in phase quadrature. To provide the multiple UHF clock wave forms with phase correlation to a GaAs CCD demands very stable picosecond timing accuracy between the adjacent clock phases, otherwise unpredictable variations will occur in the device performance. This requirement is difficult to achieve when attempting to incorporate a GaAs CCD in a high-speed analog application, such as the 500 MHz multichannel transient digitizer application being developed at TRIUMF, because of the nonlinear behaviour of the device as a load impedance. In addition to the difficult timing requirement, there are also the electronic circuit disadvantages associated with providing a multiphase clock to a GaAs CCD. The general circuit approach used to provide a multiphase UHF clock consists of splitting or dividing a master clock signal into a set of clock signals with the desired phase properties. The disadvantages of this approach are (i) the circuits are generally narrowband, restricting the range of clock frequencies that can be used; (ii) the circuits usually consume considerable power, producing substantial heat; and (iii) the circuits are typically complex, making them costly and difficult to incorporate with a GaAs CCD. A novel uniphase clock scheme was recently developed at TRIUMF that overcomes the above technical limitations for the UHF operation of a GaAs resistive gate CCD (RGCCD). The method is unique because the static transverse electric field within the GaAs RGCCD channel required to direct the motion of charge is established using the surface potential control offered by the resistive gates, under dc-biassed conditions. This permits a simple planar device comprising four electrodes per pixel to be used instead of a more elaborate castellated or ion-implanted device. Application of a single clock signal to the GaAs RGCCD provides the required temporal transverse electric field variation to cause charge motion to occur. Charge transfer efficiencies exceeding 0.999 have been achieved with a 128 pixel GaAs resistive gate CCD using (1) a fixed frequency uniphase clock operating below 100 MHz and (2) using a triggered dual frequency uniphase clock operating at 500 MHz during the signal acquisition period and at 15.6 MHz during the signal expenditure period.

Optimization of thin-film resistive-gate and capacitive-gate GaAs charge-coupled devices

Computer simulation of high-speed gallium arsenide charge coupled devices (CCDs) was performed using a two-dimensional semiconductor device simulation program. The measured performance of fabricated GaAs CCDs is compared with modeled results. The effect of active layer thickness and doping concentration on the charge transfer efficiency (CTE) and the dynamic range is investigated using different layers. Various gate architectures are compared for optimum dynamic range and compatibility with GaAs MESFET technology. The effect of clock shape on CTE is studied. Sensitivity analysis is performed on output structures. Both capacitive-gate CCD (CGCCD) and resistive-gate CCD (RGCCD) techniques are considered.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A 1-GHz charge-packet replicator/subtractor circuit for GaAs CCD signal processing

A novel charge-packet replicator/subtractor circuit based on GaAs charge-coupled device (CCD) technology is described. The circuit exhibits linear gain of 0.989 operating at 1-GHz replication frequency, while consuming only several milliwatts of dynamic power. Experimental results for a prototype circuit operating over the frequency range of 1 MHz to 1 GHz are presented. Due to the low parasitic capacitance of the GaAs semi-insulating substrate, the gain is linear and nearly unity. The chip area required for the circuit consists of several CCD gates and a small MESFET. Based on C-V measurements, the dynamic power consumed for the replication/subtraction process (operating with 5-V clock swings) is estimated to be on the order of several milliwatts at 1 GHz where the operation takes place in 1 ns. Design constraints are outlined and compared to experimental results in an effort to explore the tradeoff between speed and gain. The circuit is shown to be relatively insensitive to the precise voltages used during operation.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

GaAs charge-coupled devices

This paper reports on the first Canadian involvement in the design, process development, fabrication, and evaluation of a gallium arsenide (GaAs) charge-coupled device (CCD). The project is applications driven, and is eventually expected to yield devices capable of performing at microwave frequencies with broad bandwidths. The devices were originally designed to operate in a transient digitizer for high-energy nuclear-event analysis at the the Tri- University Meson Facility (TRIUMF) in B.C.The prototype device consists of a 32-stage, four-phase GaAs CCD intended to function at radio-frequencies (if). The work that has been undertaken will be discussed under the following aspects: device design, layout description, and substrate-material specifications. This discussion will be supported by an in-depth explanation of the functioning and operation of the GaAs CCD, and how the above considerations affect it. A description will be given of the procedures, methodology, and the test-bed utilized to evaluate the devices. Results will be given for CCDs operating close to 1 GHz. The rf waveform will be analyzed in the context of charge-transfer efficiency (CTE) and the presence of clock feedthrough. Because of the need for an eventual integration of the GaAs CCD on-chip with other essential auxiliary electronic circuitry, results will be given for devices working in conjunction with other GaAs digital and analog circuits, such as a broad-band four-phase clock generator and a sample-and-hold circuit.

Inhibition of charge packet broadening in GaAs charge-coupled devices

Computer simulation of charge transport in GaAs resistive-gate charge-coupled devices is reported. The simulation has been performed for 10 μm finger spacings, doping concentrations in the range of 1×1016–1×1017 cm−3, effective channel thicknesses 0.1–1.0 μm, lateral applied fields 3–10 kV/cm, and charge packet sizes 5%–100% of full bucket capacity. Inhibition of charge packet broadening due to transferred electron effects has been observed. Charge transfer time of 99.9% charge transfer efficiency across the intermediate phase finger was investigated and found to decrease monotonically with increasing electric field despite the turnaround in average carrier velocity. This may be attributed to a combination of improved initial charge confinement and inhibition of charge packet broadening.

Radiation Effects on GaAs Charge Coupled Devices with High Resistivity Gate Structures

The results of a study on the effects of 1 MeV electrons and 1 MeV neutrons on the operation of high speed GaAs Charge Coupled Devices (CCDs) are presented. Radiation-induced trapping levels are characterized using a linear array CCD structure and the periodic pulse technique. 1 MeV electron irradiation introduced traps at 0.1 eV and 0.39 eV with bulk trap introduction rates of 1 cm-1 and 0.33 cm-1, respectively. The devices were irradiated to a maximum fluence of 9×1014 electrons/cm2. 1 MeV neutron irradiation introduced an electron trap level at 0.64 eV with a bulk trap introduction rate of 0.5 cm-1. Catastrophic device failure occurred at neutron fluences of 6×1013 neutrons/cm2. Device charge transfer efficiency was characterized pre- and post-irradiation over the temperature range of 80°K to 300°K.

Signal correlation using a one-dimensional electroabsorptive CCD spatial light modulator

Electrically addressed spatial light modulation (SLM) may be carried out in a GaAs charge-coupled device (CCD) by charge-transfer control of the electroabsorption in the individual storage wells. One-dimensional modulator arrays may be utilized in both space-integrating and time-integrating correlators operated as incoherent optical processors. In the space-integrating case, one SLM controlled by the input signal is operated in cascade with a second SLM controlled by a reference waveform. This arrangement yields a single-detector output which is the correlation as a function of time. In the time-integrating correlator, the input signal is carried by the light source while a single SLM with a traveling reference wave controls the output intensity to a detector array. The modulator arrays may be operated with the light at normal incidence to a CCD with transparent gates or in a guided-wave mode, where the light propagates beneath the gate electrodes and parallel to the surface. The calculated dynamic range is greater for the guided-wave case because of the greater depth of modulation, although the fabrication and optical coupling are more critical. Space-integrating correlator performance is limited by bleaching or well-filling by the modulated light and predicted dynamic ranges are 20 dB for normal-incidence operation and 40 dB for the guided-wave mode. In the time-integrating correlator, saturation of the detector array is limiting, but the 100-percent modulation available in the input optical source yields 35-dB dynamic range for normal-incidence and 55 dB for guided-wave operation.

Spatial light modulation using electroabsorption in a GaAs charge-coupled device

High-speed one- and two-dimensional spatial light modulation may be carried out using the electroabsorption (Franz-Keldysh) effect in a GaAs buried-channel charge-coupled device (CCD). For photon energies slightly lower than the energy gap, the transmission through or along the surface of a CCD structure may be controlled by the signal charge in the wells, through the change of electric field with charge. The modulator is thus electrically addressable and it promises optical quality sufficient for coherent applications. Experimental measurements on a specially designed GaAs CCD structure with semitransparent gates have verified the predicted performance.

GaAs and related heterojunction charge-coupled devices

A Schottky-barrier gate GaAs charge-coupled device (CCD) technology is described. Experimental devices fabricated in this technology have operated with charge transfer efficiency >0.999/transfer and-have operated at clock frequency <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">f_{cl} = 500</tex> MHz. Implications of this technology in high-speed signal processing and in visible and near-infrared imaging are discussed.

A 500-MHz GaAs charge-coupled device

The higher electron mobility in GaAs relative to silicon makes it possible to design an ultra-high-Speed charge-coupled device. A buried-channel Schottky barrier gate GaAs charge-coupled device (CCD) is described which has been operated at up to 500 MHz clock frequency. The transit-time-limited upper clock frequency in this device is predicted to exceed 5 GHz.

Charge-coupled devices in gallium arsenide

The electronic properties of GaAs and their implications to device performance have been the subject of discussion for some time, and discrete devices such as diodes and f.e.t, with performance exceeding that which has been achieved in similar silicon devices, are commercially available. Recent advances in GaAs material and processing technologies have made possible the realisation of integrated devices with substantial complexity which have demonstrated greater speed and lower power dissipation than equivalent silicon devices. Digital integrated circuits with sub-100 ps propagation delay and power dissipation compatible with large scale integration have been demonstrated. In this paper, we describe a GaAs charge-coupled device which has demonstrated some of the attributes expected from GaAs devices of this type. This device has been operated at fcl = 500 MHz which is faster than any other ccd. Design considerations and application areas for this device are discussed.

MP-A5 planar GaAs charge-coupled devices with one and two conducting channel layers

MP-A5 planar GaAs charge-coupled devices with one and two conducting channel layers

GaAs charge-coupled devices

A Schottky-barrier-gate buried-channel GaAs CCD has been successfully demonstrated. A 10-cell (30 gates) three-phase device was operated at room temperature. The device employs a natural channel stop formed by the transfer gates extending from an n-type active region onto a semi-insulating GaAs substrate.