Article

Charge Detection Modeling in Solid-State Image Sensors

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Solid-state image sensors continue to find many applications as fabrication technology improves. Due in part to the relatively small role that image sensors have played in the semiconductor world, there exists very little experience in performance modeling of this class of devices. In this paper we discuss a three-dimensional model of the image sensor responsivity. Responsitivity is simply the amount of charge detected by the image sensor divided by the input photon energy. We discuss the fundamental aspects of charge detection and formulate and solve the appropriate model. We find good agreement between this model and experimental data.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

Article
Nano silicon is emerging as an active element for UV applications due to quantum confinement-induced widening of the Si bandgap, amenability to integration on Si, and less sensitivity to temperature. NanoSi-based UV applications include deep space exploration, high temperature propulsion, solar photovoltaics, and particle detection in high energy accelerators. However, the viability of the technology is limited by a complex nanoSi optical quenching instability. Here, we examined the time dynamics of UV-induced luminescence of sub 3-nm nanoSi. The results show that luminescence initially quenches, but it develops a stability at ∼50% level with a time characteristic of minutes. Upon isolation, partial luminescence recovery/reversibility occurs with a time characteristics of hours. To discern the origin of the instability, we perform first principles atomistic calculations of the molecular/electronic structure in 1-nm Si particles as a function of Si structural bond expansion, using time dependent density functional theory, with structural relaxation applied in both ground and excited states. For certain bond expansion/relaxation, the results show that the low-lying triplet state dips below the singlet ground state, providing a plausible long-lasting optical trap that may account for luminescence quenching as well as bond cleavage and irreversibility. Time dynamics of device-operation that accommodates the quenching/recovery time dynamics is suggested as a means to alleviate the instability and allow control of recovery, which promises to make it an effective alternative to UV-enhanced Si or metal-based wide-bandgap sensing technology.
Article
In this work, we used a commercial charge-coupled device (CCD) camera to detect and monitor terrestrial cosmic rays at ground level. Multi-site characterization has been performed at sea level (Marseille), underground (Modane Underground Laboratory) and at mountain altitude (Aiguille du Midi-Chamonix Mont-Blanc at +3,780 m of altitude) to separate the atmospheric and alpha particle emitter's contributions in the CCD response. An additional experiment at avionics altitude during a long-haul flight has been also conducted. Experiment results demonstrate the importance of the alpha contamination in the CCD response at ground level and its sensitivity to charged particles. Experimental data as a function of CCD orientation also suggests an anisotropy of the particle flux for which the device is sensitive. A complete computational modeling of the CCD imager has been conducted, based on a simplified 3D CCD architecture deduced from a reverse engineering study using electron microscopy and physico-chemical analysis. Monte Carlo simulations evidence the major contribution of low energy (below a few MeV) protons and muons in the CCD response. Comparison between experiments and simulation shows a good agreement at ground level, fully validated at avionics altitudes with a much higher particle flux and a different particle cocktail composition.
Article
This research article proposes a new method that enables silicon to respond to light in the UV wavelength range by doping Platinum (Pt), which is an impurity atom, into silicon. It is widely recognized that silicon semiconductor can respond to visible light. Silicon photodetector can thus detect merely the light with wavelengths of visible light and near IR. The effect of Pt on photocurrent of a metal-semiconductor-metal (MSM) photodetector was investigated. The undoped and Pt-doped silicon photodetectors in this experiment were fabricated to test the current characteristics. It was found that Pt-doped photodetectors can respond to the light wavelength of 365 nm. The energy bandgap (Eg ), a parameter related to the optical response of semiconductor, of Pt-doped silicon was obtained by extracting from Arrhenius plot of generation current of silicon p–n junction. The result was shown that Eg of Pt-doped silicon is increased with the introduction of Pt. Eg of silicon is typically 1.12 eV; however, the Eg in this study is approximated at 1.7 eV.
Article
We discuss on the development and characterization of a p–n heterojunction of polyaniline–silicon as a photo detector for the ultraviolet (UV) region. The hybrid heterojunction consists of micro and nano-films (<300 nm thick) of polyaniline deposited on the top of a silicon wafer through the spin-coating technique, where in the backside of silicon and on top of polyaniline, aluminum and gold were respectively deposited by evaporation to make the electric contacts. The electrical characteristics of the devices present excellent reproducibility and high rectification ratio. In addition, the spectroscopic and the photon response analysis of the devices strongly indicates that it can be used as a broad band photon detector, with good sensitivity especially in the UV region ≈2.0–3.5 eV, where it presents enhanced sensitivity (≈200%) when compared to commercial all-silicon diodes such as the OPT 301 UV detector.
Article
Amorphous silicon is used, at an industrial level, in at least three different fields of application: photovoltaic cells, flat TV screens and line scan
Article
We constructed ultraviolet (UV) photodetectors using thin films of silicon nanoparticles as active media. The Si nanoparticle films are electrodeposited at room temperature on Si p-type substrates. Uniform silicon nanoparticles of 1-nm diameter are dispersed from Si wafers using electrochemical etching. The nanoparticles are ultrabright under UV excitation, with nanosecond luminescence time characteristics. Current-voltage (I--V) characteristics indicate a photoconductor in series with a diode-like junction with a large enhancement in the forward current under UV illumination. Our results point to a sensitive UV detector with good visible blindness where the particle films effectively constitutes a wide-bandgap material.
Article
We constructed ultraviolet (UV) photodetectors by room-temperature deposition of Si nanoparticle films on Si p-type substrates. Silicon nanoparticles of 1-nm diameter are dispersed from Si wafers using electrochemical etching. The current-voltage characteristics indicate a photoconductor in series with a diode-like junction with a large enhancement in the forward current under UV illumination. With increasing wavelength, the response drops rapidly, dropping to a few percent at 560 nm. These results point to a sensitive UV detector with good visible blindness where the particle films effectively constitutes a wide-bandgap material.
Article
The CMOS active pixel sensors (APSs) can easily be designed for specific applications. Charge collection measurements have been achieved under heavy ion, alpha, and laser exposures on APS test die. The obtained results allow us to determinate the parameters to modify in order to design detectors dedicated to the detection of the space environment ionizing particles. Modeling of alpha exposure on APS complete this work. All these results demonstrate the important role of heavy ion-induced charge diffusion in APSs space applications.
Article
When a camera system is operated in a flux of charged particles, it will generate transient signals. The charged particles will impinge on the focal plane array and generate tracks of ionized electron-hole pairs. The charge tracks will diffuse, and some of the charge will be collected as signal and the rest will recombine. Bremsstrahlung radiation will be generated in the structure around the camera, and the Bremsstrahlung photons will generate secondary electrons in the focal plane array. The radiation hits will look like small stars. The charged particles will also make the optics “glow” due to luminescence and Cerenkov radiation. This paper utilizes a charged particle flux found in the open literature for a future spacecraft called Europa Orbiter to show the effects of charged particles on a camera system. A simple focal plane model is constructed based on typical values, and simulations of the transient noise will be discussed. In this specific simulation, the radiation will generate average values of 3600 photoelectrons for impinging electrons, 6000 photoelectrons for Bremsstrahlung-generated secondary electrons in the focal plane, and 15000 photoelectrons for impinging protons. These transient signals mill make it difficult for the camera system to operate under low light conditions
Article
Spatial fluctuations in absorbed dose became important as device dimensions are scaled downward and the number of elements on a single chip increased. Furthermore, the extreme events in the distribution could determine the satisfactory operation of the array. The large sample size available in a charge injection device (CID) has made it possible, for the first time, to compare experimental results for gamma-induced dose fluctuations with theoretical calculations of extreme value distributions. The agreement between theory and experiment is excellent. In order to predict correctly the variance in the dose distribution from the basic model for solid-state imagers, it is found to be necessary in the case of a CID to include a term for carrier diffusion
Article
Numerical methods are presented to investigate charge transfer in charge-coupled devices (CCDs) when potential barriers or wells occur. A Monte Carlo simulation of electron thermal diffusion and field-aided drift is used to determine the time scale for charge transfer. The Monte Carlo approach is useful for exploring new problems, but it requires considerable amounts of computer time. A quicker technique, that of the mean first passage time, is introduced. This method reduces the solution of the carrier continuity equation for charge transfer to the evaluation of a double integral that yields the characteristic time τ for e<sup>-t</sup>τ/. This provides the leading or dominant time dependence of the carrier continuity equation's solution. Numerical examples are presented to show how τ varies with the size and location of the potential obstacle. The mean first passage time approach permits rapid estimates of the effects of potential obstacles on charge transfer in CCD's. These estimates are in excellent agreement with the results of the Monte Carlo simulations
Article
A two-dimensional analytical model is formulated for calculating the pixel response, modulation transfer function (MTF), and quantum efficiency of front-side illuminated, solid-state image sensors. Included in this unified model are the effects of lateral diffusion of charge carriers within a two-layer substrate and less than full pixel sampling apertures. The results of this model are compared to those of a numerical, three-dimensional Monte Carlo algorithm and to the analytical results reported by Blouke and Robinson. We find good agreement between the quantum efficiency and MTF calculated by the present model and by the three-dimensional Monte Carlo method. However, we find higher quantum efficiency and lower MTF than the previously reported analytical two-layer model. The unified aspect of the present model correctly combines the effects of sampling aperture and lateral diffusion
Article
SiC photodiodes were fabricated using 6 H single-crystal wafers. These devices have excellent UV responsivity characteristics and very low dark current even at elevated temperatures. The reproducibility is excellent and the characteristics agree with theoretical calculations for different device designs. The advantages of these diodes are that they will operate at high temperatures and are responsive between 200 and 400 nm and not responsive to longer wavelengths because of the wide 3-eV bandgap. The responsivity at 270 nm is between 70% and 85%. Dark-current levels have been measured as a function of temperature that are orders of magnitude below those previously reported. Thus, these diodes can be expected to have excellent performance characteristics for detection of low light level UV even at elevated temperatures
Article
Theory of photoemission: Independent particle model.- The calculation of photoionization cross sections: An atomic view.- Many-electron and final-state effects: Beyond the one-electron picture.- Fermi surface excitations in X-ray photoemission line shapes from metals.- Angular dependent photoemission.
Article
The collection process of the photogenerated charge in charge-injection device (CID) image sensors is examined and a method is presented by which the collection efficiency can be calculated. Numerical techniques are utilized in the analysis. The method is applied to a specific epitaxial type structure. The contribution of the depleted and undepleted regions to the total collection efficiency of the sensor is calculated. The relation between output signal charge and incident light intensity is also investigated. It is shown that the collection efficiency decreases with increasing signal charge and this results in non-linear transfer characteristics at high signal levels.
Article
A two-phase silicon charge coupled shift register was operated as a line imaging sensor and experiments were conducted to measure quantum efficiency, spectral response, and output as a function of integration time and incident light power. This CCD imaging device was shown to have a useful quantum response over the visible through near infrared range of 0.4 to 1.0 μm with measured quantum efficiency peaks at 66% at 0.6 μm and 48% at 0.85 μm. A simple model was then developed to predict the illuminated response of basic CCD imaging cell and the results of calculations of quantum efficiency and output photovoltage based on this model were compared with the experimental data. Finally the results show that useful information on device quantum response can be obtained using a simple model.Ein ladungsgekoppeltes Zweiphasen-Silizium-Schieberegister wurde als optischer Liniensensor geprüft und der Quantenwirkungsgrad, die Ausgangsspannung und das Spektral-verhalten als Funktionen der Integrationszeit und der Lichtleistung gemessen. Die Anordnung zeigte eine brauchbare spektrale Übertragung im sichtbaren und infraroten Bereich von 0,4 bis 1,0 μm mit Meßwerten des Quantenwirkungsgrades von 66 und 48% bei 0,6 und 0,85 μm. Ein einfaches Modell wurde entwickelt, um die Lichtkonversion einer solchen Grundzelle zu berechnen. Die ausgeführten Rechnungen des Quantenwirkungsgrades und der Ausgangsspannung wurden mit Experimenten verglichen. Es zeigt sich, daß das entwickelte einfache Modell eine wertvolle Information über den Quantenwirkungsgrad liefert.
Conference Paper
A readout technique, allowing high speed readout, while lowering the fixed pattern noise and temporal noise, will be described. Method uses a thin polysilicon (< 1000 Å) upper electrode for high spectral sensitivity, strapped with a 2μ wide aluminum line for sensing.
Conference Paper
The charge injection device (CID) imaging technique utilizes an X-Y addressed array of charge storage capacitors which store photon-generated charge in MOS inversion regions. Readout is effected by injecting sequentially this stored charge into the substrate and detecting the resulting displacement current to create a video signal. The use of a common readout port has a number of basic advantages; simple mechanization, tolerance to processing defects, avoidance of charge transfer losses, and minimized blooming. Performance results obtained from 100 multiplied by 100 CID imagers, together with advances in charge collection and readout techniques that have greatly increased the operating speed and application potential of this approach are discussed.
Article
The Monte Carlo method is used to evaluate the extent of the crosstalk in solid-state imagers. The calculations are performed in three dimensions and are in excellent agreement with experiment. The Monte Carlo method is used because it handles adjacent regions that either collect or reflect minority carriers.
Article
A simplified model for the diffusion and collection of the charge generated along the track of a fast ionizing particle in Si is developed and used to discuss "soft failures" due to α-particles from radioactive elements. Several examples are worked out in detail. A scaling analysis shows that the charge collected by a device struck by an α-particle decreases less rapidly as all dimensions are scaled down than does the critical charge required to cause an error. Thus the charge margins preventing soft failures in most current devices will vanish if the devices are made smaller without significant redesign.
Article
The fabrication and performance of large (16K, 60K, and 78K cells) high density CID self-scanned imager arrays is described. Array fabrication used overlapping electrodes of two levels of polysilicon or a first level of polysilicon and a second level of antimony tin oxide. Yield considerations and the special processing steps required to fabricate these large chips are examined. Quantum efficiency curves of these structures are compared. Copyright © 1978 by The Institute of Electrical and Electronics Engineers, Inc.
Article
The use of metal oxide transparent conductive electrodes in a charge-injection device (CID) imaging array has resulted in a quantum efficiency of approximately 70 percent, uniform throughout the visible, with useful array response out to 3500 Å. The advantages of using highly transparent metal oxide electrodes for the fabrication of frontside illuminated arrays is described. A new high density CID cell is described, which does not require any contact windows or diffused regions. This cell structure is simply composed of two crossed rows and column electrodes and is easily fabricated. Because no contacts are required, it can be reduced in size for large high density arrays. The results for a 32 × 32 self-scanned array are presented.
icon gate charge-coupled optical imaging array Modeling diffusion and collection of charge from ionizing radiation in silicon devices
  • R W Brown
  • S G Chamberlain
R. W. Brown and S. G., Chamberlain, " Quantum efficiency of a sil-vol. 20, p. 675, 1913. icon gate charge-coupled optical imaging array, " Phys. Status &>lidi, S. Kirkpatrick, " Modeling diffusion and collection of charge from ionizing radiation in silicon devices, " IEEE Trans. Electron Derices, R. L. Burden, J. D. Faires, and A. C. Reynolds, Numerical Anaiysis. Boston: Prindle Weber, Schmidt, 1981.
Optical Properties of Thin Solid Films New York: Academic, 1955. S . M. Sze, Physics of Semiconductor Devices
  • Heavens
Heavens, Optical Properties of Thin Solid Films. New York: Academic, 1955. S. M. Sze, Physics of Semiconductor Devices, 2nd ed. New '(ork: Wiley, 1981, ch.