Absolute noise characterisation of avalanche photodiodes

Excess noise in four types of commercially obtained avalanche photodiodes (a.p.d.s) has been measured absolutely, by comparing avalanche noise from the a.p.d. with shot noise from an illuminated p-i-n diode. The method used yields directly the noise-current spectral density, simplifies the deduction of the quantum efficiency keff and hence the true value of the multiplication factor, and ultimately yields a measured value of the noise parameter x.     

Multiplication noise in uniform avalanche diodes

A general expression is derived from which the spectral density of the noise generated in a uniformly multiplying p-n junction can be calculated for any distribution of injected carriers. The analysis is limited to the white noise part of the noise spectrum only, and to diodes having large potential drops across the multiplying region of the depletion layer. It is shown for the special case in whichbeta = kalpha, wherekis a constant and α and β are the ionization coefficients of electrons and holes, respectively, that the noise spectral density is given by2eI_{in}M^{3}[1 + (frac{1 - k}{k})(frac{M - 1}{M})^{2}]where M is the current multiplication factor and Iinthe injected current, if the only carriers injected into the depletion layer are holes, and by2eI_{in}M^{3}[1 - (1 - k)(frac{M - 1}{M})^{2}]if the only injected carriers are electrons. An expression is also derived for the noise power which will be delivered to an external load for the limitM rightarrow infin.     

Improved figure of merit for noise in avalanche photodiodes

The authors propose a new noise figure for avalanche photodiodes (APDs). This new noise figure overcomes the difficulty of estimating the internal multiplication and quantum efficiency in complex APD structures, such as III-V SAM APDs. Measurements of the new noise figure are presented for two commercial SAM APDs and the authors show theoretically that it represents a more complete figure of merit for comparing the performance of one APD with another, and with the ideal     

Temperature dependence of multiplication noise in silicon avalanche photodiodes

The temperature dependence of multiplication noise in silicon avalanche photodiodes with a low-high-low impurity density profile is calculated. The variation of multiplication noise by temperature change can be neglected in practical use at a constant multiplication factor, which is in agreement with experimental results.     

Excess noise in GaAs avalanche photodiodes with thin multiplicationregions

It is well known that the gain-bandwidth product of an avalanche photodiode can be increased by utilizing a thin multiplication region. Previously, measurements of the excess noise factor of InP-InGaAsP-InGaAs avalanche photodiodes with separate absorption and multiplication regions indicated that this approach could also be employed to reduce the multiplication noise. This paper presents a systematic study of the noise characteristics of GaAs homojunction avalanche photodiodes with different multiplication layer thicknesses. It is demonstrated that there is a definite “size effect” for multiplication regions less than approximately 0.5 μm. A good fit to the experimental data has been achieved using a discrete, nonlocalized model for the impact ionization process     

Wavelength dependence of multiplication noise in silicon avalanche photodiodes

Theoretical and experimental results on wavelength dependence of multiplication noise in silicon avalanche photodiodes are described. When the photodiode has a p-n⁺-junction and is illuminated from the n⁺-side, multiplication noise increases by decreasing optical wavelength. Effective ionization coefficient ratio keffis equal tokexp (2Kw_{a}) for a uniform junction electric field, wherekis the ratio of ionization coefficients of electrons α and holes β. The multiplication noise depends on the product of optical absorption coefficientKand the avalanche-region width wa. Calculations show that there exists an optimum wafor minimizing multiplication noise at a given wavelength. Theoretical results are shown to agree with results of experiments on diodes with a low-high-low impurity profile. Measured ionization coefficient ratiokvalues are 0.04 and 0.08 at 0.811- and 0.633-µm wavelength, respectively.     

Effect of stochastic dead space on noise in avalanche photodiodes

A stochastic dead-space model for impact ionization is developed and used to study the effect of the soft nature of the ionization capability of carriers on the excess noise factor of avalanche photodiodes. The proposed model is based on the rationale that the gradual, or soft, transition in the probability density function (PDF) for the distance from birth to impact ionization can be viewed as that resulting from uncertainty in the dead space itself. The resulting soft PDF, which is parameterized by a tunable softness parameter, is used to establish the limitations of the existing hard-threshold ionization models in ultrathin multiplication layers. Calculations show that for a fixed operational gain and fixed average dead space, the excess noise factor tends to increase as a result of the softness in the PDF in very thin multiplication layers (viz, <70 nm), or equivalently, under high applied electric fields (viz., >800 kV/cm). A method is proposed for extracting the softness parameter from noise versus multiplication measurements.     

Silicon avalanche photodiodes with low multiplication noise and high-speed response

Low-noise and high-speed silicon avalanche photodiodes with low breakdown voltage are reported. The diode structure with a low-high-low impurity density profile is proposed to have low-noise characteristics. Multiplication noise and depletion layer width of several structures are compared theoretically, and effects of impurity density profile of the avalanche region are discussed. Built-in field is also provided to realize high-speed response without increasing operating voltage. Silicon avalanche photodiodes with the above mentioned structure have been fabricated with long time substrate annealing, ion implantation, and epitaxial growth. Attained performances are as follows: noise parameter k = 0.027 - 0.040, output pulse half width τ = 260 ps for a mode-locked Nd:YAG laser pulse, gain-bandwidth product up to 300 GHz at M = 400, quantum efficiency 0.55 - 0.66 at the 0.81- to 0.83-µm wavelength, and breakdown voltage about 100 V.     

Signal and noise response of high speed germanium avalanche photodiodes

Germanium avalanche photodiodes, providing gain at microwave frequencies, have been fabricated and tested. The diodes employ a guard ring structure to achieve a uniform, microplasma-free, multiplying region with an active diameter of 40 microns. Low-frequency chopped light current gains of greater than 200, and small-signal 6 GHz current gains of greater than 10 have been obtained at room temperature for a carrier wavelength of 1.15 microns. In the normal operating range, the signal output power is found to vary as the square of the multiplication, and the noise is found to vary as the cube of the multiplication. This limits the maximum useful multiplication of the diode to that level which gives a diode noise equal to the receiver noise. A small-signal equivalent circuit with lumped elements corresponding to the physical processes occurring within the diode, is introduced to describe the small signal behavior. The model is valid over the entire multiplication range, up to frequencies of about 10 GHz.     

低过剩噪声级联倍增雪崩探测器设计

基于弛豫空间倍增理论数值模型和修正的弛豫空间倍增理论模型,分析了不同倍增级数和不同载流子初始能量时级联倍增雪崩探测器的过剩噪声.研究了不同碰撞离化倍增层厚度、不同电子预加热层厚度、不同电场控制层掺杂浓度对过剩噪声因子的影响.同时,比较了DSMT模型、Van Vilet模型和McIntyre模型得到的结果.通过调整碰撞离化倍增层厚度、电子预加热层厚度和电场控制层掺杂浓度,DSMT数值模拟获得了一个相对优化的结构,其过剩噪声与Van Vliet模型k_s=0. 057时相当.     

Excess noise analysis of separate absorption multiplication regionsuperlattice avalanche photodiodes

The operation of a separate absorption multiplication region avalanche photodiode (SAM-APD) introduces noise as results of randomness in the number and in the position at which dark carrier pairs are generated, randomness in the photon arrival number, randomness in the carrier multiplication, and the number and the position of the photogenerated carriers in the bulk of the diode. The dark current results in a smaller mean multiplication gain in excess noise factor versus mean multiplication plot due to the partial multiplication process of these generated carriers compared to the usual values associated with carriers injected at one edge of the diode. Previous analyses of mean multiplication and excess noise factor for an arbitrary superposition of injected carriers are extended to allow the presence of dark carriers in the multiplication region under the model, which admits variation (with position) of the band-gap, dark generated rate, and ionization coefficients with each stage for the superlattice APD, and the presence of impact ionization in the absorption region. The calculations reveal the presence of impact ionization carriers in the absorption region which results in a larger excess noise factor than the usual values associated with carriers injected at one edge of the device, and fits well with experimental results     

Buried-mesa avalanche photodiodes

We have developed a low-cost buried-mesa avalanche photodiode (APD) primarily targeted for 2.5-Gb/s lightwave applications. These APDs are made by a simple batch process that produces a robust and reliable device with potentially high yield and thus low cost. The entire base structure of our InGaAs-InP APD is grown in one epitaxial step and the remaining process consists of four simple steps including a mesa etch, one epitaxial overgrowth, isolation, and metallization. Buried-mesa APDs fabricated in this way show high uniform gain that rises smoothly to breakdown with increasing reverse bias. When biased to operate at a gain of 10, these unoptimized devices show dark current less than 20 nA, excess noise factor less than 5, and a 3-dB bandwidth of about 4 GHz. With a 1550-nm laser modulated at 2488 Mb/s, a maximum sensitivity of -327 dBm was obtained with an optical receiver using one such APD, without antireflection coatings. These APD's not only demonstrate excellent device characteristics but also high reliability under rigorous stress testing. No degradation was observed even after being biased near breakdown for over 2000 h at 200°C     

Effect of dead space on gain and noise ofdouble-carrier-multiplication avalanche photodiodes

The effect of dead space on the statistics of the gain in a double-carrier-multiplication avalanche photodiode (APD) is determined using a recurrence method. The dead space is the minimum distance that a newly generated carrier must travel in order to acquire sufficient energy to become capable of causing an impact ionization. Recurrence equations are derived for the first moment, the second moment, and the probability distribution function of two random variables that are related, in a deterministic way, to the random gain of the APD. These equations are solved numerically to produce the mean gain and the excess noise factor. The presence of dead space reduces both the mean gain and the excess noise factor of the device. This may have a beneficial effect on the performance of the detector when used in optical receivers with photon noise and circuit noise     

Excess noise design of InP/GaInAsP/GaInAs avalanche photodiodes

An InP/GaInAsP/GaInAs avalanche photodiode (APD) with separate absorption and multiplication (SAM) regions has been designed taking into account the excess noise generated in GaInAsP and GaInAs. The multiplication factor dependence of the excess noise factorFhas been calculated using realistic electron and hole ionization rates in InP, GaInAsP, and GaInAs, assuming that the avalanche multiplication occurs not only in InP but in GaInAsP and GaInAs. The calculatedFvalues have been compared to the experimental ones measured on a planar-type InP/GaInAsP/GaInAs APD for illumination at a wavelength of 1.3 μm. It has been found the the calculated excess noise agrees very well with the experimental measurements. The limited ranges of device parameters in which the conditions of minimal excess noise, tunneling current, and charge pile-up are satisfied have been obtained. We conclude that the excess noise generated in GaInAsP and GaInAs should be considered in a practical device design.     

Ultra-low noise avalanche photodiodes with a "centered-well" multiplication region

We report avalanche photodiodes with a "centered-well" multiplication region that have achieved high gain, low noise, and low dark current. The multiplication region consists of an /spl sim/80 nm-thick Al/sub 0.2/Ga/sub 0.8/As layer sandwiched between two thin (10/spl sim/20 nm) layers of Al/sub 0.6/Ga/sub 0.4/As. Monte Carlo simulation shows the beneficial effect of spatial modulation of the ionization rates in this structure compared to homojunctions.     

Improved germanium avalanche photodiodes

New kinds of germanium avalanche photodiodes with n+-n-p and p+-n structures were devised for improved excess noise and high quantum efficiency performance. Multiplication noise, quantum efficiency, and pulse response were studied and compared with those of the conventional n+-p structure diode. Multiplication noise of the new type of diodes were measured in the wavelength range between 0.63 and 1.52 μm. The effective ionization coefficient ratio of the p+-n diode was lower than unity at a wavelength longer than 1.1 μm and 0.6-0.7 at 1.52 μm, and that of the n+-n-p diode was 0.6-0.7 in the whole sensitive wavelength region. Response times were evaluated by using a mode-locked Nd:YAG laser beam and a frequency bandwidth wider than 1 GHz was estimated. Receiving optical power levels were compared with each other using parameters measured in this study.     

HgCdTe electron avalanche photodiodes

Exponential-gain values well in excess of 1,000 have been obtained in HgCdTe high-density, vertically integrated photodiode (HDVIP) avalanche photodiodes (APDs) with essentially zero excess noise. This phenomenon has been observed at temperatures in the range of 77–260 K for a variety of cutoff wavelengths in the mid-wavelength infrared (MWIR) band, with evidence of similar behavior in other IR bands. A theory for electron avalanche multiplication has been developed using density of states and electron-interaction matrix elements associated with the unique band structure of HgCdTe, with allowances being made for the relevant scattering mechanisms of both electrons and holes at these temperatures. This theory is used to develop an empirical model to fit the experimental data obtained at DRS Infrared Technologies. The functional dependence of gain on applied bias voltage is obtained by the use of one adjustable parameter relating electron energy to applied voltage. A more quantitative physical theory requires the use of Monte Carlo techniques incorporating the preceding scattering rates and ionization probabilities. This has been performed at the University of Texas at Austin, and preliminary data indicate good agreement with DRS models for both avalanche gain and excess noise as a function of applied bias. These data are discussed with a view to applications at a variety of wavelengths.     

Multi-element reachthrough avalanche photodiodes

Two approaches to making multi-element arrays of p⁺-π-p-n⁺reachthrough avalanche photodiodes are reported. In the first approach a single common avalanche region (p-layer) for all elements is used, with the segmentation between elements being on the p⁺layer. This approach has the advantage of having zero dead space between adjacent elements, but is difficult to fabricate, and has a very narrow range of operation in which it is neither noisy due to injection nor suffers from poor element-to-element isolation. In a second approach, the p⁺contact is common and separate avalanche regions are used. The problem for this case is the width of the dead space between adjacent elements which, because of field-fringing effects, is considerably wider than the actual physical distance between elements. A self-aligning technique is described for fabricating arrays by the second approach and the technique demonstrated with a 25-element linear array on 300-µm centers. The measured dead space is in the 60-80 µm range, depending on the gain. The array can be used at an average gain of 100 or more, has excellent element-to-element isolation, and NEP's below 2 × 10¹⁵W/Hz^1/2at 800-900 nm and below 10^-14W/ Hz^1/2over the whole spectral range from 400 to 1060 nm.