Open access peer-reviewed chapter

Physical Design Fundamentals of High-Performance Avalanche Heterophotodiodes with Separate Absorption and Multiplication Regions

By Viacheslav Kholodnov and Mikhail Nikitin

Submitted: February 16th 2012Reviewed: June 18th 2012Published: December 19th 2012

DOI: 10.5772/50778

Downloaded: 2187

1. Introduction

Minimal value of dark current in reverse biased pnjunctions at avalanche breakdown is determined by interband tunneling. For example, tunnel component of dark current becomes dominant in reverse biased pnjunctions formed in a number semiconductor materials with relatively wide gap Egalready at room temperature when bias Vbis close to avalanche breakdown voltage VBD(Sze, 1981), (Tsang, 1981). The above statement is applicable, for example, to pnjunctions formed in semiconductor structures based on ternary alloy In0.53Ga0.47Aswhich is one of the most important material for optical communication technology in wavelength range λup to 1.7 μm (Tsang, 1981), (Stillman, 1981), (Filachev et al, 2010), (Kim et al, 1981), (Forrest et al, 1983), (Tarof et al, 1990), (Ito et al, 1981). Significant decreasing of tunnel current can be achieved in avalanche photodiode (APD) formed on multilayer heterostructure (Fig. 1) with built-in pnjunction when metallurgical boundary of pnjunction (x=0)lies in wide-gap layer of heterostructure (Tsang, 1981), (Stillman, 1981), (Filachev et al, 2010), (Kim et al, 1981), (Forrest et al, 1983), (Tarof et al, 1990), (Clark et al, 2007), (Hayat & Ramirez, 2012), (Filachev et al, 2011). Design and specification of heterostructure for creation high performance APD must be such that in operation mode the following two conditions are satisfied. First, space charge region (SCR) penetrates into narrow-gap light absorbing layer (absorber) and second, due to decrease of electric field E(x)into depth from x=0(Fig. 1), process of avalanche multiplication of charge carriers could only develop in wide-gap layer. This concept is known as APD with separate absorption and multiplication regions (SAM-APD). Suppression of tunnel current is caused by the fact that higher value of Ecorresponds to wider gap Eg. Electric field in narrow-gap layer is not high enough to produce high tunnel current in this layer. Dark current component due to thermal generation of charge carriers in SCR (thermal generation current with density JG) is proportional to intrinsic concentration of charge carriers niexp(Eg/2kBT), here kB– Boltzmann constant, T– temperature (Sze, 1981), (Stillman, 1981). Tunnel current density JTgrows considerably stronger with narrowing Egthan niand depends weakly on T(Stillman, 1981), (Burstein & Lundqvist, 1969). Therefore, component JTwill prevail over JGin semiconductor structures with reasonably narrow gap Egeven at room temperature. Another dark current component − diffusion-drift current caused by inflow of minority charge carriers into SCR from quasi-neutral regions of heterostructure is proportional to ni2×N1(Sze, 1981), (Stillman, 1981) (where Nis dopant concentration). To eliminate it one side of pnjunction is doped heavily and narrow-gap layer is grown on wide-gap isotype heavily doped substrate (Tsang, 1981). Thus heterostructure like as pwg+nwgnngnwg+is the most optimal, where subscript ‹wg› means wide-gap and ‹ng› − narrow-gap, properly. To ensure tunnel current’s density not exceeding preset value is important to know exactly allowable variation intervals of dopants concentrations and thicknesses of heterostructure’s layers. Thickness of narrow-gap layer W2is defined mainly by light absorption coefficient γand speed-of-response. But as it will be shown further tunnel current’s density depends strongly on thickness of wide-gap layer W1and dopant concentrations in wide-gap N1and narrow-gap N2layers. Approach to optimize SAM-APD structure was proposed in articles (Kim et al, 1981), (Forrest et al, 1983) (see also (Tsang, 1981)). Authors have developed diagram for physical design of SAM-APD based on heterostructure including In0.53Ga0.47Aslayer. However, diagram is not enough informative, even incorrect significantly, and cannot be reliably used for determining allowable variation intervals of heterostructure’s parameters. The matter is that diagram was developed under assumption that when electric field E(x)(see Fig. 1b) at metallurgical boundary of pwg+nwgjunction E(0)E1is higher than 4.5×105 V/cm then avalanche multiplication of charge carriers occurs in InPlayer where pwg+nwgjunction lies at any dopants concentrations and thicknesses of heterostructure’s layers. However, electric field E1=E1BDat which avalanche breakdown of pnjunction occurs depends on both doping and thicknesses of layers (Sze, 1981), (Tsang, 1981), (Osipov & Kholodnov, 1987), (Kholodnov, 1988), (Kholodnov, 1996-2), (Kholodnov, 1996-3), (Kholodnov, 1998), (Kholodnov & Kurochkin, 1998). As a consequence, avalanche multiplication of charge carriers in considered heterostructure can either does not occur at electric field value E1=4.5×105 V/cm or occurs in narrow-gap layer (Osipov & Kholodnov, 1987), (Osipov &, Kholodnov, 1989). Value of electric field required to initialize avalanche multiplication of charge carriers can even exceed E1BD(Sze, 1981), (Osipov & Kholodnov, 1987), (Kholodnov, 1996-2), (Kholodnov, 1996-3), (Kholodnov, 1998), (Kholodnov & Kurochkin, 1998) that has physical meaning in the case of transient process only (Groves et al, 2005), (Kholodnov, 2009). Further, in development of diagram was assumed that maximal allowable value of electric field in absorber at hetero-interface with multiplication layer E2(see Fig. 1b) is equal to 1.5×105 V/cm. But tunnel current density JTin narrow-gap absorber In0.53Ga0.47As(Osipov & Kholodnov, 1989) is much smaller at that value of electric field than density of thermal generation current JGwhich in the best samples of InPIn0.53Ga0.47AsInPheterostructures (Tsang, 1981), (Tarof et al, 1990), (Braer et al, 1990) can be up to 10-6 A/cm2. However, diagram does not take into account the fact that tunnel current in wide-gap multiplication layer can be much greater than in narrow-gap absorber (Osipov & Kholodnov, 1989). Therefore, total tunnel current can exceed thermal generation current.

In present chapter is done systematic analysis of interband tunnel current in avalanche heterophotodiode (AHPD) and its dependence on dopants concentrations N1in nwgwide-gap and N2in nngnarrow-gap layers of heterostructure and thicknesses W1and W2, respectively (Fig. 1) and fundamental parameters of semiconductor materials also. Performance limits of AHPDs are analyzed (Kholodnov, 1996). Formula for quantum efficiency ηof heterostructure is derived taking into account multiple internal reflections from hetero-interfaces. Concentration-thickness nomograms were developed to determine allowable variation intervals of dopants concentrations and thicknesses of heterostructure layers in order to match preset noise density and avalanche multiplication gain of photocurrent. It was found that maximal possible AHPD’s speed-of-response depends on photocurrent’s gain due to avalanche multiplication, as it is well known and permissible noise density for preset value of photocurrent’s gain also. Detailed calculations for heterostructure InPIn0.53Ga0.47AsInPare performed. The following values of fundamental parameters of InР(I, Fig. 1) and In0.53Ga0.47As(II, Fig. 1) materials (Tsang, 1981), (Stillman, 1981), (Kim et al, 1981), (Forrest et al, 1983), (Tarof et al, 1990), (Ito et al, 1981), (Braer et al, 1990), (Stillman et al, 1983), (Burkhard et al, 1982), (Casey & Panish, 1978) are used in calculations: band-gaps Eg1= 1.35 eV and Eg2= 0.73 eV; intrinsic charge carriers concentrations ni(1)=108 сm-3 and ni(2)=5.4×1011 сm-3; relative dielectric constants ε1= 12.4 and ε2=13.9; light absorption coefficient in In0.53Ga0.47Asγ=104 сm-1; specific effective masses m*=2mc×mv/(mc+mv)of light carriers m1= 0.06m0and m2= 0.045m0, where m0– free electron mass. The chapter material is presented in analytical form. For this purpose simple formulas for avalanche breakdown electric field EBDand voltage VBDof pnjunction are derived taking into account finite thickness of layer. Analytical expression for exponent in well-known Miller’s relation was obtained (Sze, 1981), (Tsang, 1981), (Miller, 1955) which describes dependence of charge carriers’ avalanche multiplication factors on applied bias voltage Vb. It is shown in final section that Geiger mode (Groves et al, 2005) of APD operation can be described by elementary functions (Kholodnov, 2009).

Figure 1.

Energy diagram of heterostructure in operation mode (a) and electric field distribution in it (b). Ec and Ev − energy of conduction band bottom and valence band top. Solid lines −N2=N2(1), dashed −N2>N2(1)


2. Formulation of the problem: Basic relations

Let’s consider pwg+nwgnngnwg+heterostructure at reverse bias Vbsufficient to initialize avalanche multiplication of charge carries. This structure is basic for fabrication of AHPDs. From relations (Sze, 1981), (Tsang, 1981), (Filachev et al, 2011), (Grekhov & Serezhkin, 1980), (Artsis & Kholodnov, 1984)


can be determined, in principal, dependences of multiplication factors Min pnstructures on Vb, where Mnand Mp– multiplication factors of electrons and holes inflow into space charge region (SCR); value of multiplication factor of charge carriers generated in SCR M˜lies between Mnand Mр; specific rate of charge carriers’ generation in SCR g=gd+gphconsists of dark gdand photogenerated gphcomponents; Lpand Ln– thicknesses of SCR in pand nsides of structure; α(E)and β(E)= K(E)×α(E)– impact ionization coefficients of electrons α(E)and holes β(E); Е(х)– electric field. Let’s denote by N1ptdopant concentration N1so that for N1<N1pt“punch-through” (depletion) of nwglayer occurs that means penetration of non-equilibrium SCR into nnglayer (Fig. 1). Optical radiation passing through wide-gap window is absorbed in nnglayer and generates electron-holes pairs in it. When N1<N1ptthen photo-holes appearing near nwg/nngheterojunction (х=W1) are heated in electric field of non-equilibrium SCR and, at moderate discontinuities in valence band top Evat х=W1, photo-holes penetrate into nwglayer (layer I) due to emission and tunneling. If W1is larger than some value W1min(N1,N2,W2)(Osipov & Kholodnov, 1989), which is calculated below, then avalanche multiplication of charge carriers occurs only in nwglayer, i.e. photo-holes fly through whole region of multiplication. In this case photocurrent’s gain (Tsang, 1981), (Artsis & Kholodnov, 1984) Mph=Mр. Let pwg+layer is doped so heavy that avalanche multiplication of charge carriers in it can be neglected (Kholodnov, 1996-2), (Kholodnov & Kurochkin, 1998). Under these conditions thicknesses in relations (1) and (2) can be put Lp=0and Ln=W1, i.e.


It is remarkable that responsivity SI(λ)(where λ– is wavelength) of heterostructure increases dramatically once SCR reaches absorber nng(layer II on Fig. 1) and then depends weakly on bias Vbtill avalanche breakdown voltage value VBD(Stillman, 1981). This effect is caused by potential barrier for photo-holes on nwg/nngheterojunction and heating of photo-holes in electric field of non-equilibrium SCR. If losses due to recombination are negligible (Sze, 1981), (Tsang, 1981), (Stillman, 1981), (Forrest et al, 1983), (Stillman et al, 1983), (Ando et al, 1980), (Trommer, 1984), for example, at punch-through of absorber, then SI(λ)in operation mode is determined by well-known expression (Sze, 1981), (Tsang, 1981), (Stillman, 1981), (Filachev et al, 2011):


where λin µm and value of quantum efficiency ηis considered below. Photocurrent gaining and large drift velocity of charge carriers in SCR allow creating high-speed high-performance photo-receivers with APDs as sensitive elements (Sze, 1981), (Tsang, 1981), (Filachev et al, 2010), (Filachev et al, 2011), (Woul, 1980). Reason is high noise density of external electronics circuit at high frequencies or large leakage currents that results in decrease in Noise Equivalent Power (NEP) of photo-receiver with increase of Мphdespite of growth APD’s noise-to-signal ratio (Tsang, 1981), (Filachev et al, 2011), (Woul, 1980), (McIntyre, 1966). Decrease in NEP takes place until Мphbecomes higher then certain value Мphoptabove which noise of APD becomes dominant in photo-receiver (Sze, 1981), (Tsang, 1981), (Filachev et al, 2011), (Woul, 1980). Even at low leakage current and low noise density of external electronics circuit, avalanche multiplication of charge carriers may lead to degradation in NEP of photo-receiver due to decreasing tendency of signal-to-noise ratio dependence on APD’s Мphunder certain conditions (Artsis & Kholodnov, 1984). Moreover, excess factor of avalanche noise (Tsang, 1981), (Filachev et al, 2011), (Woul, 1980), (McIntyre, 1966) may decrease with powering of avalanche process as, for example, in metal-dielectric-semiconductor avalanche structures, due to screening of electric field by free charge carriers (Kurochkin & Kholodnov 1999), (Kurochkin & Kholodnov 1999-2). Using results obtained in (Artsis & Kholodnov, 1984), (McIntyre, 1966), noise spectral density SNof pwg+nwgnngnwg+heterostructure which performance is limited by tunnel current can be written as:


where q– electron charge; АS– cross-section area of APD’s structure; Fef,i(Mph)– effective noise factors (Artsis & Kholodnov, 1984) in wide-gap multiplication layer (i=1) and in absorber (i=2); JT,i(V)– densities of primary tunnel currents in those layers, i.e. tunnel currents which would exist in layers I and II in absence of multiplication of charge carriers due to avalanche impact generation. Comparison of two different APDs in order to determine which one is of better performance is reasonable only at same value of Мph. Expression (5) shows, that for preset gain of photocurrent, noise density is determined by values of primary tunnel currents IT1=JT1×ASand IT2=JT2×AS(total primary tunnel current IT=IT1+IT2). Distribution of electric field Е(х)that should be known to calculate parameters (4) and (5) of AHPD is obtained from Poisson equation and in layers I and II is determined by expressions:




U(x)and U+(x)– asymmetric unit stepwise functions (Korn G. & Korn T., 2000), ε0– dielectric constant of vacuum, ε1and ε2– relative dielectric permittivity of nwgand nnglayers (Fig. 1).


3. Avalanche multiplication factors of charge carriers in p-n structures

3.1. Preliminary remarks: Avalanche breakdown field

For successful development of semiconductor devices using effects of impact ionization and avalanche multiplication of charge carriers is necessary to know dependences of avalanche multiplication factors M(V)of charge carriers in pnstructures on applied bias Vb. We need to know among them dependence of avalanche breakdown voltage VBDon parameters of pnstructure and distribution of electric field E(x)related to VBDdependence. Usual way to compute required dependencies is based on numerical processing of integral relations (1) and (2) in each case. Impact ionization coefficients of electrons α(E)and holes β(E)depend drastically on electric field E. At the same time theoretical expressions for α(E)and β(E)include usually some adjustable parameters. Therefore, to avoid large errors in calculating of multiplication factors, in computation of (1) and (2) are commonly used experimental dependences for α(E)and β(E). Avalanche breakdown voltage VBDis defined as applied bias voltage at which multiplication factor of charge carriers tends to infinity (Sze, 1981), (Tsang, 1981), (Miller, 1955), (Grekhov & Serezhkin, 1980). Therefore, as seen from (2), breakdown condition is reduced to integral equation with m=1where field distribution E(x)is determined by solving Poisson equation. Bias voltage at which breakdown condition V=VBDis satisfied can be calculated by method of successive approximations on computer. Thus, this method of determining VBDand, hence, E(x)at V=VBDrequires time-consuming numerical calculations. The same applies to dependence Mon V. Similar calculations were performed for a number of semiconductor structures for certain thicknesses of diode’s base by which is meant high-resistivity side of p+nhomojunction or narrow-gap region of heterojunction (Kim et al, 1981), (Stillman et al, 1983), (Vanyushin et al, 2007). In addition to great complexity, there are other drawbacks of this method of M(V)and VBDdetermination – difficulties in application and lack of illustrative presentation of working results. Availability of analytical, more or less universal expressions would be very helpful to analyze different characteristics of devices with avalanche multiplication of charge carriers, for example, expression describing E(x), when we estimate tunnel currents in AHPDs. In this section are presented required analytical dependences (Osipov & Kholodnov, 1987), (Kholodnov, 1988), (Kholodnov, 1996-3). For quick estimate of breakdown voltage in abrupt p+nhomojunction or heterojunction is often used well-known Sze-Gibbons approximate expression (Sze, 1981), (Sze & Gibbons, 1966):

VBD=AV×N(s2)/s, V,E10



Gap Egof semiconductor material forming diode’s base and dopant concentration Nin it are measured in eV and cm-3, properly. As follows from Poisson equation, voltage value given by (10) corresponds to value of electric field at metallurgical boundary (x=0,Fig. 2) of p+njunction:


where at s=8


ε0 and ε− dielectric constant of vacuum and relative dielectric permittivity of base material; q− electron charge. Unless otherwise stated, in formulas (12) and (13) and below in sections 3.1-3.3 is used SI system of measurement units.

Figure 2.

Schematic drawing of diode based onp+nn+heterostructure (a) and distribution of electric field in it at avalanche breakdown voltage (b);1N1=N1(0),2N1>N1(0);N1− dopant concentrationnwgin wide-gap layer I

Formulas (10) and (11) cannot be used for reliable estimates of VBDand EBDin semiconductor structures with thin enough base. Indeed, dependence of VBDon Nis due to two factors. First, as follows from Poisson equation, the larger Nthe steeper the field E(x)decreases into the depth from x=0comparing to value E1=E(0)(Fig. 1b). Second, value of electric field E1=E(0)at V=VBDfalls with decreasing of Ndue to decreasing of |E|in SCR. Drop of E(x)becomes more weaker with decreasing of N(Fig. 1b), therefore, at preset base’s thickness W, initiation of avalanche process will require fewer and fewer field intensity E1. At sufficiently low concentration N, the lower the thicker Wwill be, variation of electric field E(x)on the length of base is so insignificant that probability of impact ionization becomes practically the same in any point of base. It means that breakdown voltage VBDand field EBDare independent on Nand at the same time are dependent on W, moreover, the thinner Wthen, evidently, the higher EBD. So using of formulas (10) and (11) at any values of W, that done in many publications, contradicts with above conclusion. In next section 3.2 will be shown that value of breakdown field of stepwise p+njunction in a number of semiconductor structures can be estimated by following formula:




It seen from expression (14) that at N<N˜(W)electric field of avalanche breakdown EBDis practically independent on dopant concentration Nin diode’s base.

3.2. Avalanche breakdown field

Consider pwg+nwgnngnwg+heterostructure (Fig. 2). Symbols nwgand nngindicate to unequal, in general, doping of high-resistivity layers of structure. Denote as W1, W2and N1, N2thicknesses of nwgand nnglayers and dopant concentrations in them, properly. Case W2=0corresponds to diode formed on homogeneous p+nn+structure. Let values N1 and W1such that upon applying avalanche breakdown voltage VBDto structure, SCR penetrates into narrow-gap nnglayer (Fig. 2). When W1and N1, N2are small enough and W2is thick enough then avalanche process develops in nnglayer. In other words, with increasing bias Vbapplied to heterostructure, electric field E=E2in narrow-gap layer on nwg/nngheterojunction (Fig. 2) reaches avalanche breakdown field E2BDin this layer earlier than electric field E1on metallurgical boundary (х=0) of p+n junction becomes equal to breakdown field E1BDin wide-gap nwglayer. This is due to the fact that at small values of W1and N1variation of field E(х)within wide-gap layer is insignificant and probability of impact ionization in narrow-gap layer is much higher than in wide-gap. If, however, W1and N1, N2are large enough and W2thin enough, then avalanche process is developed in wide-gap nwglayer only. For these values of thicknesses and concentrations electric field E1reaches value E1BDearlier than E2– value E2BD. Because of significant decreasing of electric field E(х)in nwglayer with increasing distance from х=0, field E2remains smaller E2BDdespite the fact that band-gap Eg1in nwglayer is wider than band-gap Eg2in nnglayer. Distribution of electric field E(x)in nwgand nnglayers of considered heterostructure is obtained by solving Poisson equation as defined by (6)-(9). When avalanche breakdown voltage VBDis applied to structure, then either E1=E1ВD(N1,W1)or E2=E2ВD(N2,W2). In section 3.1 is noted that at low enough concentrations Niavalanche breakdown fields EiBD(Ni,Wi)should not depend on Niand have definite value depending on Wi, where i=1,2. To account for this effect, formula (12) should be modified so that when N0then breakdown field EBDtends to some non-zero value. It would seem that it is enough to add some independent on Nconstant to right side of (12). It is easy to see that such modification of formula (12) leads to contradiction. To verify that let’s consider situation when avalanche multiplication of charge carriers occurs in nwglayer, i.e. E1is close to E1ВDand multiplication factor of holes Мр(1) is fixed. Then, with increasing concentration N1, field EI(W1)(Fig. 2b) shall be monotonically falling function of N1. Indeed, with increasing N1, field E1ВDand |EI(x)|are increasing also. Increasing |EI|must be such that when xbecame larger some value x¯then value EI(x)has decreased (Fig. 2b). Otherwise, field E(x)would increase throughout SCR that reasonably would lead to growth of Мр. This is evident from (1) and (2). On the other hand, adding constant to right side of expression (12) does not change E1BD/N1and therefore results in, as follows from (6) and (9), non-monotonic dependence EI(W1)on N1. Equation (14) which can be rewritten for each of nwgand nnglayers as:


does not lead to that and other contradictions, From (17) follows that:


To determine dependences EiBD(0,Wi), let’s consider behavior of EI(W1)when parameters of heterostructure N1, N2and W2are varying. From (6)-(9), (17) and (18) we find that when value


then avalanche breakdown is controlled by nwglayer. It means that


If, however, Δ<0then avalanche breakdown is controlled by nwg/nngheterojunction, i.e.


From (17)-(21) we obtain that

EI(WI)N1|N10={A1s×N˜1(1/s)1q×W1ε0ε1 , at Δ >00                               , at Δ <0E22

Formulas (15) and (16) follow from expressions (18), (19) and requirement (23)


which means smoothness of field dependence E(x)in real heterostructures, where parameters are varying continuously. Particularly, in semiconductors for which relations (11) and (13) are valid, breakdown field at metallurgical boundary of p+njunction (or at heterojunction boundary, in narrow-gap layer of heterojunction, including isotype) can be described by formula




And values for InPsemiconductor widely used for manufacturing of AHPDs (Tsang, 1981), (Stillman, 1981), (Filachev et al, 2010), (Filachev et al, 2011) are as follows:

EВD(InP)(0,W)=4.3×105×W1/7, V/cm;N˜InP(W)=3.4×1015×W8/7, cm3E26

Xε=12.4/ε, Xg=1.35/Egand gap Egin diode’s base is measured in eV and its thickness W– in μm, respectively.

3.3. Avalanche breakdown voltage

It follows from expressions (6)-(9) and (14)-(16) that breakdown voltage VBDfor p+nn+structure is given by expressions

VBD=εε02qA2×[1+N˜(W)N]2/s×Ns2sAV×[1+N˜(W)N]2/s×Ns2s, ifN˜N<1θE27

i.e. when diode’s base is not punch-through and

VBD(N,W)=VBD(0,W)×{[1+NN˜(W)]1/sN2s×N˜(W)} , ifN˜N>1θE28

i.e. when diode base is punch-through. In expression (28)


Value of parameter θis defined from equation θ=s×(1+θ)1/sand with good degree of accuracy it equals to ss/(s1). Because θ>>1, therefore expression (27) practically coincides with formula (10), i.e. VВDof diode with thick base is independent on its thickness W. For diodes with thin base formed on semiconductors with parameters satisfying relations (11) and (14), namely when


breakdown voltage of diode depends on Wand Nas follows




In expressions (30)-(32) Xε=12.4/ε, Xg=1.35/Egand gap Egin base, dopant concentration in it Nand thickness Wis measured in eV, cm-3 and μm, respectively.

Avalanche breakdown voltage of double heterostructure discussed in Section 4 (Fig. 1) depends on relations between fundamental parameters of materials of nwgand nnglayers, their thicknesses and doping, and is determined, as follows from (6)-(9) and (14)-(16), by different combinations (with slight modification) of expressions (27)-(29) for these layers of heterostructure.

3.4. About correlation between impact ionization coefficients of electrons and holes

One of main goals of many experimental and theoretical studies of impact ionization phenomenon in semiconductors is to determine impact ionization coefficients of electrons α(E)and holes β(E)as functions of electric field E(Sze, 1981), (Tsang, 1985), (Grekhov & Serezhkin, 1980), (Stillman & Wolf, 1977), (Dmitriev et al, 1987). Parameters of some semiconductor devices, for example, APDs (Sze, 1981), (Filachev et al, 2011), (Artsis & Kholodnov, 1984), (Stillman & Wolf, 1977) depend significantly on ratio K(E)=β(E)/α(E). Performance of APD can be calculated on computer if α(E)and β(E)are known (Sze, 1981), (Tsang, 1985), (Filachev et al, 2011), (Grekhov & Serezhkin, 1980), (Stillman & Wolf, 1977), (Dmitriev et al, 1987). Dependences α(E)and β(E)are known, with greater or lesser degree of accuracy, for a number of semiconductors (Sze, 1981), (Tsang, 1985), (Grekhov & Serezhkin, 1980), (Stillman & Wolf, 1977), (Dmitriev et al, 1987). However in works concerned determination of impact ionization coefficients the problem of interrelation between α(E)and β(E)has never been put. Even so, laws of conservation of energy and quasi-momentum in the act of impact ionization are maintained mainly by electron-hole subsystem of semiconductor (Tsang, 1985), (Grekhov & Serezhkin, 1980), (Dmitriev et al, 1987). Therefore, there is a reason to hypothesize some correlation between α(E)and β(E), although perhaps not quite unique, for example, owing to big role of phonons in formation of distribution functions. It is shown in this section that for number of semiconductors the following approximate relation is satisfied (Kholodnov, 1988)


Where: ε– relative dielectric permittivity, and gap Eg, electric field E, αand βare measured in eV, V/cm and 1/cm, properly.

To derive relation (33) let’s consider thin p+nn+structure in which thickness of high-resistivity base layer Wsatisfies to inequality


where ε0– dielectric constant of vacuum; ε– relative dielectric permittivity of base material; q– electron charge; sand A– constants defining dependence of electric field EBDA×N1/sat metallurgical boundary (x=0)of abrupt p+njunction on dopant concentration Nin base for avalanche breakdown in thick p+nn+structure (Sections 3.1-3.3, (Sze, 1981), (Grekhov & Serezhkin, 1980), (Sze & Gibbons, 1966)). When condition (34) is satisfied then avalanche breakdown field can be written as


And, under these conditions, variation of electric field Е(x)along length of base Wis so insignificant that probability of impact ionization is practically the same in any point of base of considered structure. For many semiconductors including Ge,Si,GaAs,InP,GaPrelations given below are valid (Sze, 1981), (Kholodnov, 1988-2), (Kholodnov, 1996), (Sze & Gibbons, 1966)


In this case as it follows from (34) and (35)


And avalanche breakdown electric field for thin p+nn+structure is defined by approximate universal formula


In expressions (37) and (38) and below in this Section 3.4 concentration is measured in cm-3, energy – in eV, length – in µm, electric field – in V/cm. On the other hand condition of avalanche breakdown of p+nn+structure (Sections 2, 3.1 and (Sze, 1981), (Tsang, 1985), (Grekhov & Serezhkin, 1980), (Stillman & Wolf, 1977))


takes the form


That means the same probability of impact ionization in any point of diode’s base. And relation (33) follows from expressions (38) and (40). Let’s estimate applicable electric field interval for this relation. Expression (38) will be valid when inequality (41) is satisfied both for electrons and for holes


where λR,Еion,ЕR– mean free path for charge carriers scattered by optical phonons, threshold ionization energy of electrons or holes and energy of Raman phonon, respectively (Sze, 1981), (Tsang, 1985), (Grekhov & Serezhkin, 1980), (Stillman & Wolf, 1977), (Dmitriev, 1987). Taking into account that for many semiconductors


From (38) and (41) we find desired interval of electric field:


Interval of electric field (43) is most often realized in experimental studies (Sze, 1981), (Tsang, 1985), (Grekhov & Serezhkin, 1980), (Stillman & Wolf, 1977), (Dmitriev, 1987). Ratio Wmin/λRis usually not more than a few units. Therefore, when W<Wminthen EBDEionW×104and hence when E>Emaxinstead of (33) must be valid relation


where Еionto be understood by largest in value threshold ionization energy of electrons and holes. On basis of relations (33) (or its upgraded version, if parameters sand Adiffer from values of (36)) and (44)) can be obtained although approximate but relatively simple and universal analytical dependences of charge carriers multiplication factors and excess noise factors (Tsang, 1985), (Stillman et al, 1983), (Artsis & Kholodnov, 1984), (Woul, 1980), (McIntyre 1966), (Stillman & Wolf, 1977) on voltage as well as analytical expressions for avalanche breakdown voltage at different spatial distributions of dopant concentration in pnstructures.

3.5. Miller’s relation for multiplication factors of charge carriers in p-n structures

Usual way to calculate dependences of avalanche multiplication factors of charge carriers M(Section 2) in pnstructures on applied voltage Vbis based on numerical processing of integral relations (1) and (2) in each case. Distribution of specific rate of charge carriers’ generation g(x)in space charge region (SCR), i.e. when Lp<x<Ln(see inset in Fig. 3), is accepted in this Section 3.5 as exponential (and as special case − uniform). It is valuable for practical applications to have analytical, more or less universal, dependences Mon Vb. In article (Sze & Gibbons, 1966) was proposed analytical expressions for avalanche breakdown voltage VBD, i.e. applied voltage value at which M=, in asymmetric abrupt and linear pnjunctions. Expression for VBD(Sze & Gibbons, 1966) in the case of asymmetric abrupt p+njunction was generalized in (Osipov & Kholodnov, 1987) for the case of thin p+n(p)n+structure (like as pin) as discussed in Section 3.3. Using as model abrupt (stepwise) pnjunction under assumption that K(E)=β/α=const(Kholodnov, 1988-2) has been shown that from (1), (2) and approximate relation (33), which is valid for number of semiconductors including Ge,Si,GaAs,InP,GaP, can be obtained analytical dependences of multiplication factors of charge carriers on voltage.

Figure 3.

Dependences of exponents in Miller’s relation for electronnnand holesnpfor "thick" abruptрnjunction on applied voltageVat different valuesK=β/αequal to 1, 2, 3, and 4

Rewrite (33) in the form


In (45) and below in this Section 3.5 is accepted (unless otherwise specified) the following, convenient for this study, system of symbols and units (Sze, 1981): gap Egand threshold ionization energy Еionin eV; electric field Ein V/cm; bias Vbin V; multiplication factors αand βin cm-1, electron charge qin C; dielectric constant of vacuum ε0in F/m; concentration including shallow donors NDand acceptors NAin cm-3; concentration gradient ain cm-4; width of SCR Lрand Lnin pand nlayers and thicknesses of these layers (inset in Figure 3) in μm, light absorption coefficient γin cm-1. In this section, analytical dependences M(V)in pnstructures have been calculated under no K(E)=constcondition. Such calculations are possible because ratio K(E)1lnK(E)varies, typically, much slighter than E7. In some cases it allows using relation (45) to integrate analytically (in some cases – approximately) expressions (1) and (2) and, thus, get analytical, more or less universal, relatively simple dependences M(V). The most typical cases are considered: abrupt (stepwise) and gradual (linear) pnjunctions like as in model given in (Sze, 1987), (Sze & Gibbons, 1966) and thin p+n(p)n+structure (like as pin) with stepwise doping profile as in model presented in (Osipov & Kholodnov, 1987). For purposes of discussion and comparison of obtained results with numerical calculations and experimental data, multiplication factors will be written in traditional common form


where v=V/VBD. This form was first proposed by Miller in 1955 (Miller, 1955) and then, despite lack of analytical expressions for exponents nn, np, n˜, has been widely used as "Miller’s relation" (Sze, 1981), (Tsang, 1985), (Grekhov & Serezhkin, 1980), (Leguerre & Urgell, 1976), (Bogdanov et al, 1986). It was found that values of these exponents depend on many factors including, in general, voltage as well (Kholodnov, 1988-2), (Grekhov & Serezhkin, 1980), Fig. 3. Form of writing (46) clearly shows that M(V)when VVBD.

3.5.1. Stepwise pnjunction

In this case from relations (1), (2) and (45) and Poisson equation (SI units)

dEdx={qNAεε0,       x<0qNDεε0,     x>0E47

follow that


where K0– value K(х)when Е(х)=Е(0)=Е0, i.e. value of Kat metallurgical boundary of pnjunction (see inset in Fig. 3). Formula (49) for VBDat ND<<NAor NA<<NDbecomes well-known Sze-Gibbons relation (Sze, 1981), (Sze & Gibbons, 1966). If charge carriers are generated uniformly in SCR then computations lead to following expressions:






e(x)– unity function (Zeldovich & Myshkis, 1972), Keff=K0+K01. Expression (50) is obtained by expanding the function Y(x,Lp)as a power series in


and expression (52) was derived by standard method of integrating fast-changing functions (Zeldovich & Myshkis, 1972).

3.5.2. Gradual (linear) pnjunction

In this case Poisson equation can be written as (SI units):


where σ- slope of linear concentration profile

and therefore


In derivation of relations (55) and (56) was used known expression for voltage distribution on linear pnjunction (Sze, 1981) and was also taken into account that (Gradstein & Ryzhyk, 1963)


Formula (56) differs from known formula Sze-Gibbons for avalanche breakdown voltage of linear pnjunction (Sze, 1981), (Sze & Gibbons, 1966) by last multiplicand, which for typical values of ε10(Sze, 1981), (Casey & Panish, 1978) is close to unity.

3.5.3. Thin p+n(p)n+structure (pin)

When thickness of high-resistivity region (base) of considered structure


where N– dopant concentration (for example, donor) in base, and when Vb=VBDthen SCR does not extend to entire thickness of base ((Osipov & Kholodnov, 1987), Sections 3.1-3.3, inset in Fig. 4). In this case, expressions (48)-(53) remain apparently valid. In opposite case, base is depleted by free charge carriers when Vb<VBDthat gives in the result substantially other expressions for avalanche multiplication factors of charge carriers and avalanche breakdown voltage. When W<W˜then from relations (1), (2) and (45) and Poisson equation


we find that




Figure 4.

Dependences of analytical (solid lines) and numerical (dashed lines) (Leguerre & Urgell, 1976) limiting values of exponentnB=limVVBDn(V)in Miller’s relation (46) on concentration of donor dopantNDin "thick" high-resistivity layer of stepwisep+nn+structure, 1 − Si, 2 − Ge, 3 − GaAs, 4 − GaP. ValuesK(E), as in (Leguerre & Urgell, 1976), are taken from (Sze & Gibbons, 1966). In inset − scheme of "thick"p+nn+structure

In deriving expressions (60)-(63), multiplication of charge carriers in p+and n+layers and voltage drop on them is considered negligible. This is justified because of significant decreasing of electric field E(х)deep into high-doped layers of the structure (Sze, 1981), (Kholodnov 1996-1), (Kholodnov 1998), (Leguerre & Urgell, 1976). Admissibility of such neglect is confirmed also by formula (49) when NA<<NDor when ND<<NA. Avalanche breakdown voltage is determined by equation v˜= 1 which has no exact analytical solution. However, till Wsurpasses W˜/8, then value of field at x=Wis much less than value of field at x=0. In this case, using smallness parameter


we find that in zeroth-order approximation with respect to this parameter


In the case of very thin base when


electric field varies so slightly along base that probability of impact ionization is practically the same in any point of it ((Osipov & Kholodnov, 1987), (Kholodnov 1988-1), Sections 3.2 and 3.4). As a result


where γ< 0, if structure is illuminated through p+region (front-side illuminated) and γ> 0 if structure is illuminated through n+region (back-side illuminated).

3.6. Discussion of the results. Comparison with computed and experimental data

3.6.1. To formulas for avalanche breakdown electric field and voltage for abrupt p+ - njunction

In sections 3.1-3.3 were derived approximate universal formulas for avalanche breakdown field EBDand voltage VBDfor abrupt p+njunction taking into account finite thickness of high-resistivity layer W. Comparative values of breakdown field EBD(0,W)for Si, Geand InPmost often used for fabrication of APDs computed by formulas (25) and (26) and found from numerical solution of breakdown integral equation m=1, where mis defined by (2) are shown on Fig. 5 (Sze, 1981), (Tsang, 1985), (Stillman, 1981), (Filachev et al, 2010), (Filachev et al, 2011), (Groves et al, 2005), (Stillman et al, 1983), (Trommer, 1984), (Woul, 1980), (Leguerre & Urgell, 1976), (Bogdanov et al, 1986), (Gasanov et al, 1988), (Brain, 1981), (Tager & Vald-Perlov, 1968). It is seen that in the most practically interesting range W(0.2÷10)μm for all a.m. semiconductors analytical EBD(a)(0,W)and calculated EBD(c)(0,W)values of breakdown field differ by less than 20 %. Relatively drastic fall of ratio EBD(a)(0,W)/EBD(с)(0,W)in comparison to unity with decrease of W(for thin enough W) is due to the fact that, as shown in Sec. 3.4, if

W<Wmin5×103×εЕg×Еion7/6, µm,E70

then formulas (25) and (26) are not true. To estimate breakdown field EBD(0,W)at values Wdefined by (70) can be used the following formula

EBD(0,W)=104×EionW, V/cmE71

Figure 5.

Dependence of ratio between analytical value of breakdown fieldEBD(а)(0,W)obtained by formulas (25) and (26) and numerical valueEBD(c)(0,W). Dashed curve − analytical value of effective avalanche breakdown fieldEBD*(0,W)=Xε3/7×Xg6/7×EBD(0,W)EВD(InP)(0,W). Curves 1 and 1' − Si, 2 − InP, 3 − Ge. Valuesα(E)andβ(E)are taken: for curves 1 and 3 − fromTable 1of monograph (Grekhov & Serezhkin, 1980), for curve 1' − from (Kuzmin et al, 1975), for curve 2 − from (Cook et al, 1982)

If assume that in Sithreshold energy of impact ionization Еionof holes is higher than electrons, and it equals to 5 eV (Sze, 1981), then from (70) we find for SiWmin0.1μm. Estimates based on data from studies (Sze, 1981), (Tsang, 1985), (Stillman et al, 1983), (Grekhov & Serezhkin, 1980), (Stillman & Wolf, 1977) show that for Geand InPvalue Wminis 2-3 times smaller.

Therefore curve 1 in Fig. 5 starts to fall significantly below unity at larger values Wthan curves 2 and 3. Analytical and computed dependences EBDon Nfor InPused in high-performance APDs for wavelength range λ=(1÷1,7)µm as wide-gap layers in double heterostructures (Fig. 1, 2) are shown on Fig. 6 (Tsang, 1985), (Stillman, 1981), (Filachev et al, 2010), (Forrest et al, 1983), (Filachev et al, 2011), (Stillman et al, 1983), (Ando et al, 1980), (Trommer, 1984). It is seen that EBD(a)(N,W)and EBD(c)(N,W)differ from each other by less than 10 %. In Fig. 7 and 8 are shown universal dependences of breakdown voltage VBD(a)on Nand Wcalculated by formulas (11), (27)-(29). It is seen from Fig. 7 that Sze-Gibbons relations (10) and (11) (Sze, 1981), (Sze & Gibbons, 1966) can be used to determine VBDwhen N>Nmin10×N˜(W)only. Value of this minimal concentration, for example, for classic semiconductors Si, Ge, GaAs, GaPand InPat W= (1-2) μm equals to (1÷5)×1016cm-3. As shown on lower inset in Fig. 7, dependence VBDon Nis in the strict sense non-monotonic. Such kind of dependence VBDon Nis due to the fact that for small enough Nbreakdown field EBDis growing faster with increasing Nthan |E|Nin diode’s base. Maximum VBDis reached, as it follows from (28), at


and expressed as


when s=8, value Nmax1.2×N˜, ΔVmax(rel)2.86×102<<1and absolute value ΔVmaxcan reach tens Volts, and even more (see Fig. 7). The analytical dependences VBD(a)(N,W)(Fig. 7 and 8) for a number of semiconductors are in good agreement with VBD(c)(N,W)computed on the basis of integral equations (1) and (2) (Sze, 1981), (Tsang, 1985), (Stillman, 1981), (Stillman et al, 1983), (Grekhov & Serezhkin, 1980), (Leguerre & Urgell, 1976). Note that results of comparison VBD(a)(N,W)with VBD(c)(N,W)and ЕBD(a)(N,W)with ЕBD(c)(N,W)depend on accuracy of determination of impact ionization coefficients of electrons α(Е)and holes β(Е)which are sharp functions of electric field Е. As a rule, different authors obtain different results (Sze, 1981), (Tsang, 1985), (Stillman, 1981), (Stillman et al, 1983), (Grekhov & Serezhkin, 1980), (Sze & Gibbons, 1966), (Stillman & Wolf 1977), (Dmitriev et al, 1987), (Tager & Vald-Perlov, 1968), (McIntyre, 1972), (Cook et al, 1982) (see, for example, curves 1 and 1' in Fig. 5). In addition, deducing of relations (1) and (2) is based on local relation between αand β(Sze, 1981), (Tsang, 1985), (Stillman, 1981), (Filachev et al, 2011), (Stillman et al, 1983), (Grekhov & Serezhkin, 1980), (Sze & Gibbons, 1966), (Stillman & Wolf 1977), (Dmitriev et al, 1987), (Tager & Vald-Perlov, 1968), (McIntyre, 1972), (Cook et al, 1982) which is not always valid (McIntyre, 1972), (Gribnikov et al, 1981), (Okuto & Crowell, 1974), (McIntyre, 1999).

Figure 6.

Dependence of fieldEBDonNforInP: 1 −W=0.5μm, 2 −W=2μm, 3 −W=8μm. Solid lines – formulas (24)-(26), dashed curves − numerical calculation. Valuesα(E)andβ(E)are taken from (Cook et al, 1982). In inset is shown dependence of effective concentrationN˜*=Xε4/7×Xg6/7×N˜N˜InPonW. Concentration is measured in cm-3, field − in V/cm and thicknessW− in μm.

Figure 7.

Dependence of avalanche breakdown voltageVBDof homogeneousp+nn+structure on dopant concentrationNin base: solid line − (31) and (32), dotted line − expressions (10) and (11). In lower inset: dependence of relative voltageΔV(rel)=[VBD/VBD(0,W)]1normalized to concentrationN˜(W)atN4×N˜(W). In upper inset: dependence of effectiveΔVmax*=Xε3/7×Xg6/7×[VBDVBD(0,W)]maxΔVmax(InP)on base thicknessW. Voltage is measured in V, thicknessW− in μm.

Figure 8.

Dependence of effective avalanche breakdown voltageVBD*=Xε3/7×Xg6/7×VBDVBD(InP)of homogeneousp+nn+structure on thickness of its baseWfor three values of effective concentrationN*=Xε4/7×Xg6/7×NNInP: 1 −N*=3×1016cm-3, 2 −N*=3×1014cm-3, 3 −N*=3×1012cm-3. In inset is shown dependenceW˜onN*. Concentration is measured in cm-3, voltage − in V, thicknessW− in μm

3.6.2. To сorrelation between values of impact ionization coefficients of electrons and holes

In Section 3.4 is shown that there is reason to suppose existence of some correlation between values of impact ionization coefficients of electrons α(Е)and holes β(Е), and form of required relation (expression (33) and (45)) is proposed. It is obvious from Fig. 9 that values Z0ε3/Eg6may differ by many orders of magnitude in different semiconductors. At the same time, for presented in Fig. 9 Ge,Siand GaP, function c(Е)(see relations (33) and (45)) in range of fields where α(Е)and β(Е)vary in several orders of magnitude (Okuto & Crowell, 1975), remains, as it follows from (33) and (45), of the order of unity. Calculations based on experimental dependences α(Е)and β(Е)(Cook et al, 1982) show that in InPvalue c(Е)is some more closely to 1. It is evident from Fig. 10 that for GaAs, regardless of orientation of crystal with respect to electric field, function c(Е)depends weakly on Еin comparison with impact ionization coefficients of charge carriers (which values are taken from (Lee & Sze, 1980)), and differs from unity by no more than 2-3 times. A similar situation takes place in Ge(Fig. 11, according to (Mikawa et al, 1980)). As shown in (Kobajashi et al, 1969) dependences α(Е)and β(Е)measured in (Miller, 1955), (McKay & McAfee, 1953) in the range of fields Е=(1.5÷2.7)×105V/cm can be described in Geby formulas α(Е)=7.81×1034×Е7, β(Е)2α(Е). This result agrees well with expression (33). Note that, c(Е)differs from unity approximately by the same factor, as values α(Е)and β(Е)for the same material obtained by different authors differ, respectively, from each other (Sze, 1981), (Tsang, 1985), (Forrest et al, 1983), (Grekhov & Serezhkin, 1980), (Sze & Gibbons, 1966), (Stillman & Wolf 1977), (Dmitriev et al, 1987), (Tager & Vald-Perlov, 1968), (Cook et al, 1982), (Okuto & Crowell, 1974), (Okuto & Crowell, 1975), (Lee & Sze, 1980), (Mikawa et al, 1980), (Kuzmin et al, 1975). Using procedure described in Section 3.4, we can also determine relation between α(Е)and β(E)=K(E)×α(E)in the case when relations (11) and (13) are not satisfied (Grekhov & Serezhkin, 1980). It seems, relation required for such case, i.e. under assumption of power dependence αon Еand K(E)=const, was obtained for the first time in (Shotov, 1958).

Figure 9.

DependenceZ(E)[relation (33)] in Ge, Si, andGaPforα(E)andβ(E)from (Okuto & Crowell, 1975)

3.6.3. To Miller’s relation

From (48), (55) and (67) follow that, exponents in Miller’s relation (46) for multiplication factors of electrons and holes are given by


where ξ= 4, 5 and 7 for stepwise pnjunction, linear pnjunction and very thin (66) p+nn+structure (situation 1, 2 and 3, respectively). If thickness of base in p+nn+structure is not very small, i.e., W0<W<W˜(situation 4) then as it follows from formula (60), exponents nnand nрare also expressed by (74) and (75) but in right side of those expressions v˜substitutes vand ξ=8. Value of exponent n˜lies between values nnand nр. From (1) and (2) apparent that when α=βthen factors Mn, Mpand M˜coincide with each other, i.e., nn=nр=n˜=n, and, as it follows from expressions (74) and (75), regardless of bias voltage applied, n= 4, 5 and 7 for situations 1, 2 and 3, respectively. Exponents in Miller’s relation have the same values when V<<VBD, more exactly, when |lnK0/lnv|<<ξ, regardless of ratio K0=β(E0)/α(E0). When VVBDor more exactly, if


Then for these situations


Graphs in Fig. 4 allow comparing numerical values of exponents nnBand nрBcalculated in (Leguerre & Urgell, 1976) nB(c)and analytical nB(a)computed by formulas (76) for asymmetrical stepwise pnjunction. Like as in (Leguerre & Urgell, 1976), experimentally determined functional dependencies α(E0)and β(E0)(Sze & Gibbons, 1966) were used in calculations of dependences nB(a). As follows from (46), when M>>1, then ratio of analytical value of multiplication factor M(a)to calculated M(c)equals to ratio nB(c)to nB(a)(Fig. 11-13). It obviously from Fig. 11-13 that for all considered semiconductors (with curves α(Е)and β(Е)taken from (Sze & Gibbons, 1966)), dependences M(a)(V)and M(c)(V)do not differ by more than 50 %. Dependences of exponents nn(a)and np(a)on voltage and nnB(a)and npB(a)on ratio K=β/αare illustrated in Fig. 3 and 14, respectively. It should be noted that numerical values of exponent in Miller’s relation, as well as, value VBDdepend, obviously, on what functions α(Е)and β(Е)are used in (1) and (2) in calculations. Let’s take the simplest case when α(E)=β(E)and pnjunction is stepwise. Varying expressions (1) and (2), we find that under considered conditions


where EBD=E(0)at V=VBDis determined from condition


Figure 10.

DependenceС(E)at different orientations ofGaAscrystal with respect to electric field for valuesα(E)andβ(E)from (Lee & Sze, 1980)

Figure 11.

Dependence of ratio between analytical values of avalanche multiplication factorsМ(а)of electrons and holes and numerical valuesМ(с)(Leguerre & Urgell, 1976) in stepwise asymmetricGepnjunction on value of multiplication factorМ=М(а)of charge carriers. Solid lines – electrons, dashed – holes. Dopant concentration in high-resistivity part ofpnjunctionN, cm-3: 1 −1015, 2 −3×1015, 3 −1016, 4 −3×1016, 5 −6×1016. ValuesK(E), as in (Leguerre & Urgell, 1976), are taken from (Sze & Gibbons, 1966)

Figure 12.

Dependence of ratio between analytical values of avalanche multiplication factorsМ(а)of electrons and holes and numerical valuesМ(с)(Leguerre & Urgell, 1976) in stepwise asymmetricSipnjunction on value of multiplication factorМ=М(а)of charge carriers. Solid lines – electrons, dashed – holes. Dopant concentration in high-resistivity part ofpnjunctionN, cm-3: 1 −1015, 2 −3×1015, 3 −1016, 4 −3×1016, 5 −6×1016. ValuesK(E), as in (Leguerre & Urgell, 1976), are taken from (Sze & Gibbons, 1966)

Figure 13.

Dependence of ratio between analytical values of avalanche multiplication factorsМ(а)of electrons and holes and numerical valuesМ(с)(Leguerre & Urgell, 1976) in stepwise asymmetricGaAs(solid lines) andGaP(dashed lines)pnjunctions on value of multiplication factorМ=М(а)of charge carriers. Solid lines – electrons, dashed – holes. Dopant concentration in high-resistivity part ofpnjunctionN, cm-3: 1 −1015, 2 −3×1015, 3 −1016, 4 −3×1016, 5 −6×1016. ValuesK(E), as in (Leguerre & Urgell, 1976), are taken from (Sze & Gibbons, 1966)

Figure 14.

Dependence of limiting valuesnB=limVVBDn(V)of exponents in Miller’s relation for electronnnand holesnpfor "thick" abruptрnjunction onK=β/α

In Fig. 15a are shown dependences nB(Neff)calculated from relations (77) and (78) for four values α(E)=β(E)obtained for GaAsby different authors (Grekhov & Serezhkin, 1980), (Okuto & Crowell, 1975), (Kressel & Kupsky, 1966), (Nuttall & Nield, 1974). It is seen that analytical value nnB=npB=4calculated by formulas (76) approximately equals to mean value with respect to curves 1-4 in Fig. 15a. According to obtained above results expressions (48)-(53) are not valid when concentration


which for many semiconductors is of the order of 1017 cm-3. At such high concentrations, as it follows from Section 3.4 and (Kholodnov, 1988-1) and relations (1) and (2), for stepwise pnjunction




Figure 15.

DependencesnB(Neff)inGaAscalculated on the base of different dependencesα(E)=β(E), taken from: 1 − (Shabde & Yeh, 1970), 2 − (Grekhov & Serezhkin, 1980), 3 − (Okuto & Crowell, 1975), 4 − (Kressel & Kupsky, 1966), 5 − (Sze & Gibbons, 1966). Dashed lines − analytical values

For comparison, in Fig. 15b are presented dependences of nB(c)(Neff)and nB(a)(Neff)=1 for the case α=β, when nnВ=nрВ=nВ. It is seen that value nB(a)(Neff)= 1 is approximately equal to mean value with reference to curves 2, 3 and 5 in Fig. 15b plotted on the base of numerical data. Note that starting from Neff(Neff)maxbreakdown voltage VBDdependence on Neffbecomes, with growth Neff, more and more weaker than that described by equation (49), and in limit tends to value VBD=Eion/q. This conclusion accords with results of studies (Grekhov & Serezhkin, 1980), (Nuttall & Nield, 1974). Obtained results agree well with experimental results for a number of pnstructures, including based on Ge,Si,GaAs,GaP(Sze, 1981), (Tsang, 1985), (Stillman et al, 1983), (Miller, 1955), (Grekhov & Serezhkin, 1980), (Sze & Gibbons, 1966), (Stillman & Wolf, 1977), (Bogdanov et al, 1986), (Cook et al, 1982), (Shotov, 1958). We present here three cases of studies. In experimental study (Miller, 1955) of breakdown in Gestepwise pnjunction was found that measured values of exponents in Miller’s relation were lying in range from 3 to 6.6. The same values of exponents are obtained from expressions (74) and (75) with ξ=4if we take into account that in Gewith doping levels used in (Miller, 1955) К02÷3(Sze, 1981), (Tsang, 1985), (Miller, 1955), (Grekhov & Serezhkin, 1980), (Stillman & Wolf, 1977), (Shotov, 1958). In experimental study (Bogdanov et al, 1986) of APD based on MIS structure (metal-insulator-semiconductor APD) multiplication of charge carriers occurs in thick pSisubstrate. From point of view of avalanche process this structure is similar to asymmetric stepwise n+pjunction. Therefore, avalanche process in MIS APD can be described by expressions (74)-(76) with ξ=4. Concentration of shallow acceptors in substrate of investigated structure was 1015 cm-3. At this doping avalanche breakdown in Sioccurs when electric field near insulator-semiconductor interface reaches value EBD3×105V/cm (Sections 3.1 and 3.2, (Sze, 1981), (Osipov & Kholodnov, 1987), (Sze & Gibbons, 1966)), and therefore К0102(Sze, 1981), (Tsang, 1985), (Grekhov & Serezhkin, 1980), (Sze & Gibbons, 1966), (Stillman & Wolf, 1977), (Kuzmin et al, 1975). Measured in (Bogdanov et al, 1986) value nnat VBDV<<VBDwas found equal to 0.2. From formulas (76) with К0102follows that nnВ=0.186. In Tables 1 and 2 are presented experimental (Shotov, 1958) and calculated by formulas (48) and (55) values of multiplication factors of electrons Mn(V)and holes Mp(V)in Gestepwise and linear pnjunctions. Obviously, for these pnjunctions, experimental and analytical values of multiplication factors differ from each other by less than 20 % in whole voltage Vrange used in measurements.

Experiment(Shotov, 1958)Theory

Table 1.

Experimental (Shotov, 1958) and computed [from Equation (48)] hole avalanche multiplication factor Mpin step-wise pnjunction in pGefor different ratios of applied voltage to avalanche breakdown voltage V/VBD. It is assumed that K0=2(Shotov, 1958)

MpMnK0 (*)
Experiment (*)TheoryExperiment (*)Theory

Table 2.

Experimental (*) (Shotov, 1958) and computed [from Equation (55)] avalanche multiplication factors Mpand Mnfor holes and electrons in Gelinear pnjunction for different ratios of applied voltage to avalanche breakdown voltage V/VBD(Shotov, 1958)

Finally, it is interesting to analyze application of expressions (45) and (76) to describe avalanche process in InSb. The fact is that dependence α(Е)in InSbwas quite well known already in 1967 (Baertsch, 1967), but no one could obtain information about dependence β(Е)(Dmitriev et al, 1987), (Dmitriev et al, 1983), (Dmitriev et al, 1982), (Gavrjushko et al, 1968). Substituting in (45) dependence α(Е)for InSb(Baertsch, 1967), (Dmitriev et al, 1983), (Dmitriev et al, 1982), (Gavrjushko et al, 1968), we find that ratio K=β(E)/α(E)is vanishingly small up to electric field E4×104V/cm resulting in extremely high value nрВwhen at the same time value nnВis extremely small. It means that Mn(V)becomes much larger than unity, even at voltages Vbnoticeably lower avalanche breakdown voltage VBD, and value Mp(V)remains equal to unity up to values Vbvery close to VBD. Effect obtained from application of relations (45) and (76) accords very well with experimental data (Baertsch, 1967), (Dmitriev et al, 1983) and explains why multiplication of holes in InSbis extremely hard to observe (Dmitriev et al, 1987), (Baertsch, 1967), (Dmitriev et al, 1983), (Dmitriev et al, 1982), (Gavrjushko et al, 1968).


4. Tunnel currents in avalanche heterophotodiodes

4.1. Calculation of tunnel currents in approximation of quasi-uniform electric field and conditions of its applicability

In act of interband tunneling electron from valence band overcomes potential barrier ABC (Fig. 16a). The length of tunneling lT, i.e. length on which energy of bottom of conduction band Ес(x)changes by value equal to Еgis found by solving integral equation


If variation of electric field within length of tunneling ΔЕ<<Е, i.e. specific length of variation of field lE>>lT, then expanding function Е(x)in Taylor series around point x=x, we find that in the first order of parameter of smallness lT/lEequation (82) takes the form


When N(x)=constthen equation (83) is exact. As can be seen from Fig. 16a, if


then true ABC barrier coincides to high degree of accuracy with triangle ABC′ to which corresponds uniform field Е(х)(Fig. 16b).

It follows from (83) and Poisson equation that inequalities (84) are satisfied if


at that


Figure 16.

Physical meaning of quasi-uniform field approximation: a − band diagram, b − field distribution on length of tunneling. ABC − true potential barrier, ABC' − potential barrier used de facto. Dashed lines −Е(х)=const

As shown below, due to large values of field Еat avalanche breakdown of pnstructures, inequality (85) is valid for almost all materials up to concentration N=1017cm-3 and even high.

Under these conditions specific rates of charge carriers’ tunnel generation gTi(x)in layers I and II of structure can be described by expression


obtained in (Kane, 1960) (see also (Burstein & Lundqvist, 1969)) for Е(х)=const, in which


Here , Еgiand mi*=2mc×mv/(mc+mv)– crossed Plank constant, gaps and specific effective masses of light charge carriers in proper layers. Approximation of quasi-uniform field (87) and expressions (6)-(9) result in convenient formula for analysis of primary interband tunnel current density


where characteristic dimensions of areas of charge carriers’ tunnel generation in layers I and II


Equation (89) is valid under conditions


These conditions mean the following. If inequalities (91) for gTi(E)are satisfied then expression (87) is valid, at least in the neighborhood of field value E=Ei. When right side of inequalities (91) is satisfied then tunnel generation drops sharply with decreasing E, and therefore ITiat WTi<Wiis mainly determined by tunneling in areas 0xWT1and W1xW1+WT2.

Fulfillment of conditions (92) is necessary at punch-through of proper layers of structure for neglecting tunneling through its hetero-interfaces which is not accounted for by formula (89). We show further, that at avalanche breakdown, inequalities (91) and (92) are valid for almost all real values of material parameters, concentrations Niand layers’ thicknesses Wiof heterostructure. Avalanche breakdown occurs when one of fields Eibecomes close to breakdown field EiBDof proper layer of structure ((Sze, 1981), (Tsang, 1985), (Grekhov & Serezhkin, 1980), Sections 3.1-3.3).

Breakdown fields EiBDcan be obtained by formula (14) (Osipov & Kholodnov, 1987), (Osipov &, Kholodnov, 1989), i.e.,




(sand Ai– some constants).

For many semiconductors including InxGa1xASyP1yalloy which is one of the main materials for avalanche heterophotodiodes fabrication (Tsang, 1981), (Stillman, 1981), (Filachev et al, 2010), (Kim et al, 1981), (Forrest et al, 1983), (Tarof et al, 1990), (Ito et al, 1981), (Clark et al, 2007), (Hayat & Ramirez, 2012), (Filachev et al, 2011), (Stillman et al, 1983), (Ando et al, 1980), (Trommer, 1984), (Woul, 1980)


From expressions (93) and (94) when relations (95) are satisfied we find the following.

  1. When

NiNi(1)=8.9×1019Xmi4×Xεi4×Xgi6, cm-3,WiWi(1)=Χmi3.5×Χεi3×Χgi6×1.4×104, µm,E96

then ratio Еito aiis less than 0.1, where Χmi=0.06/mi0*,Χεi=12.4/εi,Χgi=1.35/Egi(for InPwhich is often used for growing of wide-gap layers of heterostructure (Tsang, 1981), (Stillman, 1981), (Filachev et al, 2010), (Kim et al, 1981), (Forrest et al, 1983), (Tarof et al, 1990), (Ito et al, 1981), (Clark et al, 2007), (Hayat & Ramirez, 2012), (Filachev et al, 2011), (Stillman et al, 1983), (Ando et al, 1980), (Trommer, 1984), (Woul, 1980)), Xmi=Xεi=Xgi=1, mi0*=mi*/m0(m0– free-electron mass)

  1. When


then under avalanche breakdown of proper layer of structure ratio δito ЕiBD/aiis not exceed unity, moreover, even when Ni=Ni(2)

  1. When

Wi>>1.8×102Xεi×Xgi6, µm,E99

then length of tunneling lTiat Ei=EiBDis much shorter than thickness Wiof this layer.

In expressions (96)-(99) Еgiis measured in eV. Analysis shows that under avalanche breakdown of heterostructure inequities (91) and (92) are satisfied for real values of Niand Wiand Ei<EiBD, i.e. in layer which does not control avalanche breakdown also. As can be seen from Fig. 17, when punch-through of layer nwgstops then, obviously, conditions (91) and (92) become no longer valid. Note that calculations of tunnel currents in approximation of quasi-uniform field lead to some overestimation of actually available. In fact, due to high doping of рwg+layer, tunnel current in it can be ignored; this is situation similar to MIS structures (Anderson, 1977). In ntype layers electric field decreases with increasing distance from metallurgical boundary of p+njunction (Fig. 1b), and because gradient of potential is expressed as dφ/dx=Ethen slope of zones Еc(x)and Еv(x)decreases with increasing x. It is shown from Figure 16a that use of quasi-uniform field approximation means underestimating of thickness of actual barrier ABC. As expected, numerical calculations in WKB approximation (Anderson, 1977) give a somewhat smaller value of tunnel currents than formula (89). Since tunnel currents are strongly dependent on parameters of material, which in real samples, usually, more or less different from those used in calculations (moreover, exact dopant’s distribution profile Ni(x)and hence shape of barrier ABC are usually unknown), then slight overestimation of tunnel currents values provides some technological margin that is needed for development of devices with required specifications.

Figure 17.

Dependence of generalized parameters of smallnessδ2*andlT2*in quasi-uniform field approximation on concentrationN1, atМph=100, in case, when charge carriers multiplication occurs innwg:InPlayer. Solid lines −δ2*, dashed −lT2*. ValuesW1, μm: 1 − 0.5, 2 – 2, 3 − 8.N1pt− maximal concentrationN1at which punch-through ofnwglayer is possible;δ2=(N2/1016)×(ε2/ε1)×Eg2×δ2*;lT2=(ε2/ε1)×Eg2×lT2*;Eg2- eV, concentration − cm-3.

4.2. Features of interband tunnel currents in p+nheterostructures under avalanche breakdown

Analysis of expression (89) under avalanche breakdown of p+nheterostructure, i.e., when either E1=E1BDor E2=E2BD, shows that in contrast to homogeneous pnjunction (Stillman, 1981), (Ando et al, 1980) density of initial tunnel current JT, as a rule, is not a monotonic function N1. An increase in N2cause, for some values of N1and Wi, the rise of tunnel current and vice versa – decrease of tunnel current when N1and Wihave different values. Depending on gap Egiof heterostructure’s layers and their thicknesses Withe following situations are possible.

4.2.1. Independent doping levels of wide-gap and narrow-gap ntype layers



In this case, at any concentration N1, field E1=E1BD(N1,W1), and E2<E2BD, i.e., avalanche breakdown is controlled by nwglayer.

As follows from (6)-(9), (89) and (93), if


which is fulfilled with large margin at a2ε2<a1ε1due to large ratio of a1to E1BD(0,W1) (1-2 orders of magnitude) while


then tunnel current is almost independent on N1.

If ssufficiently large ((Sze, 1981), (Osipov & Kholodnov, 1987), (Sze & Gibbons, 1966), Sections 3.1-3.3), then with further increase of N1tunnel current is monotonically falling. However, in most real cases, for example, when relations (95) is valid, tunnel current at N1>N˜1(T)first decreases and then increases.

One can see that at minimum of tunnel current, as a rule, the following inequality is valid




When (103) is fulfilled then WT1<W1.

Therefore, as it follows from (6)-(9), (89), (90) and (93), concentration N1=N1min(T)at which JTreaches minimum is defined by equation





Expression (105) is valid when inequality WT2<W2is fulfilled. This inequality and inequality (103) also are fulfilled at minimum of tunnel current in the most practically interesting cases. Below is explained difference between situations WT2>W2and WT2<W2at N1=N1min(T). Equation (104) can be solved by successive approximations using parameters of smallness ξand 1/s.

As a result we find




It is shown from (105) and (106) that N1min(T)is decreased with growth W1and, also, although weakly, with increase N2.

When N1=N1min(T)then density of tunnel current





From (94), (105) and (108) follow that JTmindecreases sharply with increasing W1. Value JTmindecreases also, although weakly, with increasing N2. Ratio




drops sharply, same as JTmin, with increase W1, but it increases with increasing N2. Value of this ratio is usually several orders of magnitude less than unity. For example, for combination of layers nwg:InP/nng:In0.53Ga0.47As, differential of currents, as can be shown, does not exceed values (N2/1018)0.9×2×104, where N2is measured in cm-3.

When concentrations


then in minimum of JT(N1)takes place punch-through of narrow-gap layer, i.e. non-equilibrium SCR reaches nwg+layer. When N1>N1min(T)then tunnel current increases with increasing N1, and at the same time, non-equilibrium SCR will penetrate into narrow-gap layer until concentration N1reaches value


Nature of above dependence JTon N1is competition between tunnel currents in wide-gap and narrow-gap layers of heterostructure (Fig. 1a). When N1N˜1(T)then field E=EI(W1)in nwglayer at its heterojunction (Fig. 1b) coincide with very high accuracy with E1BD. Due to relatively large field E2=(ε1/ε2)×E1BD, current density JTis determined by tunneling of charge carriers in narrow-gap layer, i.e. JTJT2(Fig. 1a). With increasing N1, field E2and therefore current JT2decrease due to fall EI(W1)(Fig. 18). Decrease EI(W1)with increase N1is caused by requirement (1) of constancy of photocurrent gain Mph=Mр. Indeed, increase N1for given Mphshould lead to growth E1. Otherwise, due to growth |Е(х)|with increasing N1, field would be reduced everywhere in SCR, which in turn would lead to a decrease Mph. However, increase E1should not be too large, and it should be such that Е(х)at хgreater than some value in interval 0<x<W1is decreased. In other words, Е(х)anywhere in SCR would increase, that, evidently, would increase Mph. It can be seen directly from (1) and (2). Note that for sufficiently large values of multiplication factors Mph, field E1is practically independent on Mphand very close to breakdown field E1BD(N1,W1)when value of integral m(2) is equal to unity. This allows to use value E1=E1BD(N1,W1)(93) instead of true value Е1(N1,W1,Mph). When N1>N˜1(T), then variation of field Е(х)at distance W1in nwglayer is still very insignificant, but it is enough to affect value JT2. Due to decrease E2with growth N1(especially when N1>N˜1), current is more and more determined by tunneling of charge carriers in nwglayer, therefore when N1>N1min(T), current density JTJT1increases with increase N1because E1BDgrows with increase N1. Initial plateau (Fig. 18a) on the graph JT(N1)is caused by extremely weak dependences E1BDon N1(93) and Еon хin nwglayer when N1<N˜1(T). Reducing of value JTmin(108) with growth N2is due to increasing length of tunneling in narrow-gap nnglayer (Fig. 1). Indeed, in this layer Е~N2<0, and E2under these conditions does not depend on N2. It means, that Е(х)everywhere in nnglayer, except of point x=W1, falls with increase N2(1b). Since dEcdx=dEvdx=dφdx=E<0, then slopes of Ec(x)and Ev(x)everywhere in nnglayer, except of point x=W1, decrease also with increasing N2, that leads to increase length of tunneling. Reducing of JTis more significant with growth N2when N1<N1min(T)(Fig.18b), because current density JT2increases with decrease N1while JT1decreases. When N1<N˜1(T)then current density JT1JT2, and if N1=N1min(T)it exceeds JT2. Therefore, ratio of JTminto JT|N1<N˜1(T)(109) increases with increasing N2. Because at N1=N1min(T)value JT1>JT2, then, naturally, concentration N1min(T)(106) slightly decreases with increasing N2(Fig. 18b). For small values N2, when WT2>W2, Е(х)in nnglayer coincides with E2with high accuracy. Therefore, length of tunneling in this layer, and hence JTalso, do not depend on N2. Reducing of values N1min(T)(106) and JTmin(108) with increasing W1(Figure 18a) is due to the fact that the more is W1then the less is E1BDand the greater is fall of field Е(х)in depth of nwglayer.

Figure 18.

Dependence of tunnel current density on concentrationN1in case of independent doping levels ofnwg:InPandnng:In0.53Ga0.47Aslayers atW2=2μm.aN2=1014cm-3;W1, μm: 1 − 0.1, 2 − 0.2, 3 − 0.5, 4 − 1.b− neighborhood of valueN1=N1min(T);W2=2μm;N2, cm-3: 1 −1014, 2 −1015, 3 −1016, 4 −1017


Condition (100) is not satisfied. For example, for combination of layers nwg:InP/nng:In0.53Ga0.47Assuch situation takes place when


where N2and Wiare measured in cm-3 and μm, respectively. Under this condition, when N1<N¯1, where N¯1satisfies equation


avalanche breakdown is controlled by nnglayer, i.e. E2=E2BD(N2,W2), and E1<E1BDand it increases linearly with N1. Therefore, strictly speaking, when N1<N¯1then tunnel current increases with increasing N1. At the same time, JT2does not depend on N1under following conditions.

  1. If


then at N1<N¯1,JT2>>JT1with margin of several orders of magnitude, and therefore with very high accuracy JT(N1)=const. If N1>N¯1then due to decrease E2and hence JT2also, density of tunnel current JT(N1)begins drop sharply and, reaching minimum value (108) at concentration (106), then starts to grow again due to growth JT1(N1).

  1. If


then after initial plateau JT(N1)grows monotonically. It is due to monotonic increase in component of tunnel current density JT(N1), which at N1N¯1is considerably superior to JT2.

  1. If


then for small enough thicknesses W1of layer nwgdependence JT(N1)has distinct maximum at N1=N¯1, however, at least in this case minimum is not deep. This is due to the fact that components of tunnel current density JT1and JT2are equal to each other in order of magnitude at small enough W1. Characteristics of tunnel currents in heterostructure with independent doping of nwgand nnglayers are illustrated in Fig. 18. Note that if in case Iincrease N2leads to decrease JTat all values N1, then in case II, increase N2, when N1is small enough, leads to increase of tunnel current, but at sufficiently large N1tunnel current decreases, particularly, in the vicinity of concentration N1=N1min(T).

4.2.2. Equal doping levels of wide-gap and narrow-gap ntype layers

Under this condition density of tunnel current is given by expression (89), where N1=N2=N



At this relation of parameters avalanche breakdown is controlled by nwglayer, i.e. E1=E1BD(N1,W1), and E2<E2BD(N2,W2)regardless of doping. Dependence JTon Nhas identical character with JT(N1)|N2=constin the case of 4.2.1. I, and is caused by the same physical grounds. The only difference is that when N<N2, then curves JT(N)lie higher on plotting area, and when N>N2– lower, than curves JT(N1)|N2=constin the case of 4.2.1. I.

This occurs because at given value E2length of tunneling in narrow-gap layer is the greater the higher is level of doping of this layer.

ii. Condition (117) is not satisfied.

Then, till N<N¯, (where N¯is determined by equation (113), where N¯1=N2=N¯) avalanche breakdown is controlled by nnglayer, i.e. E2=E2BD(N,W2), and E1<E1BD(N,W1)and increases linearly with N. Dependence JT(N)has, in contrast to situation 4.2.1, not only deep minimum, but high maximum also (Fig. 19a). This is due to the fact that when N<N¯then E1grows and E2grows also reaching at N=N¯maximal value (Fig. 19b). As a result, when N<N¯then JT1grows with increase Nand JT2grows also. Note that when doping of nwgand nnglayers are equal then concentration N=Nmin(T), at which tunnel current density JThas minimal value, is determined by formula (106) with accuracy up to small corrections of order ξ=E1BD(0,W1)/a1<<1, as in the case of independent doping of nwgand nnglayers. Formula for JTminmay be obtained from expression (108), if we replace N2by Nmin(T)in it.

Figure 19.

Dependences of tunnel current density (a) and fieldsЕ(b) on dopant concentrationNin case of equal doping levels ofnwg:InPandnng:In0.53Ga0.47Aslayers, atW2=2μm.W1, μm: 1 – 10, 2 – 1, 3 – 0.1, Curves 1', 2', 3' –Е2(N); curve 4 −Е1(N), weakly dependent onW1


5. Basic performance of avalanche heterophotodiode

5.1. Responsivity

In punch-through conditions of absorber nng, current responsivity SI(λ)of heterostructure under study can be described by relation (4). In calculating quantum efficiency ηof heterostructure, we take into account that optical radiation is not absorbed in its wide-gap layers. Let’s assume that light beam falls perpendicularly to front surface of heterostructure (Fig. 1), and absorption coefficient in narrow-gap layer γ(λ)does not depend on electric field. Quantum efficiency is ratio of number of electron-hole pairs generated in sample by absorbed photons per unit time to incident flux of photons.

Therefore, (Fig. 20a)


where reflection coefficient of light from illuminated surface R1=(εexε1)2/(εex+ε1)2and from interfaces of heterostructure R2=(ε2ε1)2/(ε2+ε1)2; εex– relative dielectric constant of environment; and quantum efficiency η1with respect to light ray which has penetrated into narrow-gap layer is written


quantum efficiency η2with respect to light ray which has reached to second interface of heterostructure,


ζ=exp(γW2), R3– reflection coefficient of light from not illuminated (backside) surface. From expressions (118)-(120) follow, that






Dependence ηon W2for heterostructure InP/In0.53Ga0.47As/InPis shown in Fig. 20b. It should be noted that since in operation, electric field is high even in absorption layer, then, due to Franz-Keldysh effect, quantum efficiency is slightly higher than given in Fig. 20b. This is especially true when absorbing layer W2is very thin.

Figure 20.

Layout view of multiple internal reflections and absorptions of light beam in heterostructure (a) and dependence of quantum efficiencyηof structureInP/In0.53Ga0.47As/InPon absorption layer thicknessW2, μm (b): 1 −R3=R12 −R3=1. It is assumed that relative dielectric permittivity of environmentεex=1

5.2. Noise

It was noted above that in order to achieve the best performance of SAM-APD special doping profile is formed in heterostructure which facilitates penetration of photogenerated charge carriers with higher impact ionization coefficient into multiplication layer. In this case, at given voltage bias on heterostructure, current responsivity SI(λ)is maximal, and effective noise factor Fef,i(Mph)is minimal (Tsang, 1985), (Filachev et al, 2011), (Artsis & Kholodnov, 1984), (McIntyre 1966), and hence, as it is evident from expression (5), noise spectral density SNis also minimal. If α=β, then (Tsang, 1985), (Filachev et al, 2011), (Artsis & Kholodnov, 1984), (McIntyre 1966) Fef(Mph)=Mph, and therefore


In InPratio K(E)=β/αin interval of fields of interest E=(3.3÷7.7)×105V / cm varies from 2.3 to 1.4 (Tsang, 1985), (Filachev et al, 2011), (Cook et al, 1982). Therefore, noise spectral density of heterostructure with InPmultiplication layer and optimal doping is slightly less than value given by formula (125). When N1>N¯1, (where N¯1satisfies equation (113) (see Fig. 21), in which N˜i(Wi)is defined by formula (94) for i=1,2) then avalanche multiplication of charge carriers in narrow-gap layer does not occur. Under these conditions, field value at metallurgical boundary of p+njunction (х=0, Fig. 1) equals to E1=E1B(N1,W1)(see (93) and (94)). For many semiconductors (see Sections 3.1-3.2) including InхGa1хAsyP1y, values sand Aiare defined by relations (95). In the case of heterostructure InP/In0.53Ga0.47As/InP, in first approximation in parameters of smallness


we find that value of concentration N1=N1min(T)at which function JT(N1)reaches its minimum

JTmin(W1,N2)=2.19×108×W10,49N20,07×exp(27.88×W11/7), A/cm2,E127

is given by

N1min(T)(W1,N2)=2.33×1016W18/7×[12.52×102W11/7×(lnN2×W18/73.69×10151.41)], cm-3E128

Formulas (127) and (128) are valid when WT2W2, i.e., as follows from Section 4.2.1, when


where concentration and thicknesses, as in (127) and (128), are measured in cm-3 and µm, respectively.

If inequality (129) is not satisfied, then values N1min(T)and JTminwill be again determined by (127) and (128), in which N2is replaced by Q(W1)/W2. It is shown from (127) and (128) that N1min(T)and JTminare decreasing, moreover JTminsharply, with increase W1(see Fig. 21, 22), and, also, although weakly, with increase N2. Decrease of values N1min(T)and JTminwith increase W1is caused by situation when the thicker W1the less E1BDand the greater fall of field E(x)on nwglayer thickness. Slight decrease N1min(T)and JTminwith growth N2is due to increasing of length of interband tunneling lTngin narrow-gap nnglayer with increase N2and the fact that at minimum JT1>JT2. For small values either N2or W2, field E(x)is so weakly dependent on xin nnglayer, that value lTngin it is almost constant. Therefore, when N2W2<Q(W1)then values N1min(T)and JTmindo no longer depend on N2and slightly decrease with increase W2due to reducing the length of tunneling generation region in narrow-gap material. In high performance diode, absorber should be punched-through when voltage bias Vbon heterostructure is less than voltage of avalanche breakdown VBD. This eliminates dark diffusion current from narrow-gap layer and increases operational speed. Condition of punch-through of absorber, as follows from 4.1 and 4.2 is given by:


Allowable intervals of concentrations and thicknesses of heterostructure layers are shown in Fig. 21. As can be seen from Fig. 20b, even, when R1=R3quantum efficiency reaches almost its maximal value when W2=2μm. Therefore, for development of concentration – thickness nomogram in Fig. 21, namely this value W2was selected. Note that decrease in dispersion in N2results in increase in dispersion N1and W1, while increase gives the opposite result. Value of noise current density IN1012A/Hz1/2 corresponds to JT1.8×105А/сm2, and value IN1013А/ Hz1/2 corresponds to JT1.8×107А/сm2.

Figure 21.

Concentration-thickness nomogram for avalancheInP/In0.53Ga0.47As/InPheterophotodiode whenN2=(1÷5)×1015cm-3,W2=2μm,Мph=15, cross-section areaA=5×103μm2. When noise currentIN=SN1013A/Hz1/2, then allowable set of points in space(N1,W1)lies inside figure a-b-c-d; whenIN=SN1012A/Hz1/2 − inside figure a-e-f-g. Dashed and dash-dot curves − dependencesN1min(T)(W1)andN¯1(W1), respectively: 1 −N2=1015cm-3, 2 −N2=5×1015cm-3.N1is measured in units of1016cm-3,W1− in μm

Figure 22.

Dependences of minimal tunnel currentJTmin, A/cm2 of avalanche heterophotodiodeInP/In0.53Ga0.47As/InPon multiplication layer thicknessW1, μm: 1 −N2=1013cm-3, 2 −N2=1015cm-3, 3 −N2=1017cm-3

5.3. Operational speed

Minimal possible time-of-response of this class of devices


is determined by time-of-flight of charge carriers through multiplication layer τtr1and absorber τtr2, and also by value of function f(Mph), which is close to 1 when K>>1, and is equal to Mphwhen K=1(Tsang, 1985), (Filachev et al, 2011), (Emmons, 1967), (Kurochkin & Kholodnov, 1996). It was noted above that in InP1<K2.3. Therefore, in InP/InxGa1xAsyP1y/InPSAM-APD


As is evident from Fig. 20b, in InP/In0.53Ga0.47As/InPheterostructure quantum efficiency value ηlies in interval 0.5η0.686when R3=1and W20.5μm. It means that, because of not so much loss in quantum efficiency ηcompared to maximal possible (only 27 % less), time-of-response value τtr2=5ps can be achieved by forming absorber with thickness W2=0.5μm and fully reflecting backside surface. Minimal value τtr1is determined by maximum allowable minimal value W1min. When JT106A/cm2, then as follows from Fig. 22, W1min2μm, and therefore τmin(4Mph+1)×102ns.


6. Analytical model of avalanche photodiodes operation in Geiger mode

We consider possibility to describe transient phenomena in pinAPDs by elementary functions, first of all, when initially applied voltage V0is greater than avalanche breakdown voltage VBD. Formulation of the problem is caused by need to know specific conditions of APDs operation in Geiger mode. Simple expression describing dynamics of avalanche Geiger process is derived. Formula for total time of Geiger process is obtained. Explicit analytical expression for realization of Geiger mode is presented. Applicability of obtained results is defined. APDs in Geiger mode (pulsed photoelectric signals) make possible detection of single photons (Groves et al, 2005), (Spinelli & Lacaita, 1997), (Zheleznykh et al, 2011), (Stoppa et al, 2005), (Gulakov et al, 2007). It is worked at reverse bias voltages Vb>VBD. Different types of devices are realized on APDs in Geiger mode (Groves et al, 2005), (Spinelli & Lacaita, 1997), (Zheleznykh et al, 2011), (Stoppa et al, 2005), (Gulakov et al, 2007). At the same time, review of publications shows that theoretical studies have tendency to carry out increasingly sophisticated numerical simulations. In (Vanyushin et al, 2007) was proposed discrete model of Geiger avalanche process in pinstructure. Obtained iterative relations allow to determine, although fairly easy, but only by numerical method, options for realization of Geiger mode when ratio Kβ/αdiffers very much from unity, where α(EiGE)and β(EiGE)– impact ionization coefficients of electrons and holes and EiGE– electric field in i– layer (base 0<x<WiGE, Fig. 23). "Continuous" model (Kholodnov, 2009) developed in this section admits value К=1. Considered below approach allows also to describe conditions of realization of Geiger mode and its characteristics by mathematically simple, graphically illustrative relations. It is adopted that photogeneration (PhG) is uniform over sample cross-section area Stransverse to axis x(Fig. 23). Then, in the most important single-photon process, area S, according to uncertainty principle, shall not exceed in the order of magnitude, square of wavelength of light λ. Under these conditions, it is allowably to consider problem as one-dimensional (axis x, Fig. 23). There are grounds to suppose that go beyond one-dimensional model at local illumination make no sense. Single-photon case arises itself when S>>S1π×λ2. The matter is that charge, during Geiger avalanche process, as show estimates below, has no time to spread significantly over cross section area. Consider serial circuit: pindiode – load resistance R– power supply source providing bias Vb>VBD. Let pand nregions are heavily doped, so that prevailing share of bias falls across base i. Then after charging process voltage on it can be considered equals to V0=Vb. When electron-hole pairs appear in the base then occurs their multiplication that results in decrease Vidue to screening of field EiGEin base by major charge carriers inflowing into pand nregions (Fig. 23) in quantity Nnand Ppand voltage drop across load resistor VRand, hence, current in external circuit arise


In present structure charge is mainly concentrated in thin near border niand pilayers (let's call them plates, Fig. 23). Therefore, as in (Vanyushin et al, 2007), field EiGEwill be assumed uniform. Numerical value EiGE=EBDwhen Vi=VBDfor a number of materials can be quickly determined by formulas given in Section 3. As in (Vanyushin et al, 2007), we restrict consideration by PhG in base only, we neglect recombination in it, and we assume that currents of electrons INand holes IPare determined by their drift in electric field with velocity of saturation vs, i.e.,