CN103593581A - Method for extracting defect time constant through transient current spectrum - Google Patents

Abstract

The invention discloses a method for extracting a defect time constant through a transient current spectrum. The method comprises the step a of measuring transient response of a GaN HEMT drain current, the step b of conducting nonlinear fitting on the measured transient current, and the step c of establishing the relation between the defect time constant and a relative density and determining the defect time constant according to the relative density. The method for extracting the defect time constant through the transient current spectrum is a simplified scheme for solving the problems that an existing method for extracting a defect time constant of a short-grid-length device is high in equipment dependency degree and large in debugging difficulty, and is also used for extracting a defect time constant of a long-grid-length device.

Description

A kind of method of extracting defect time constant by transient current spectrum

Technical field

The present invention relates to technical field of semiconductor device, especially a kind of method of extracting defect time constant by transient current, affects the defect time constant of current degradation computational problem in order to solve in the long device of short grid.

Background technology

The current collapse effect of GaN high electron mobility transistor (GaN HEMT) is the key factor that affects device frequency/power characteristic and reliability.Current collapse is mainly that the defect in material causes, so phenetic analysis defect is the primary content that solves current collapse.

For grid length, at the long device of micron-sized long grid, the defect analysis technology based on electric capacity substantially can Efficient Characterization, such as capacitance-voltage analytical technology (CV technology), electric capacity deep level transient spectroscopy (C-DLTS); But along with grid length continues to be reduced to submicron order, the parasitic component of gate capacitance (edge capacitance) will have a strong impact on sign and the analysis of defect, and so far the defect analysis technology based on electric capacity can not adapt to analysis demand gradually; At present, the deep energy level defect analysis of the long device of corresponding short grid is mainly by low-frequency noise technology (LNA) and electric current deep level transient spectroscopy (I-DLTS), but LNA technology is responsive to electromagnetic environment, need set of equipments to support, I-DLTS also needs expensive specialized equipment to support and very high temperature/low temperature environment, for waveform, detect, and action need rich experiences, so above-mentioned two kinds of methods are difficult to use as conventional method is universal.

Summary of the invention

(1) technical matters that will solve

In view of this, fundamental purpose of the present invention is to provide a kind of method of extracting defect time constant by transient current spectrum, the demand long device defects time constant of short grid being characterized to meet general user.

(2) technical scheme

For achieving the above object, the invention provides a kind of method of extracting defect time constant by transient current spectrum, the method comprises:

Step a: the transient response of measuring GaN HEMT drain current;

Step b: surveyed transient current is carried out to nonlinear fitting;

Step c: set up the relation of defect time constant and relative density, and determine defect time constant according to relative density.

In such scheme, the transient response of drain current described in step a is to be applied in after voltage bias when device, and the response of drain current in time domain, is denoted as I _data.Described in step a, measure the transient response of GaNHEMT drain current, the minimum sampling interval that testing apparatus gathers current signal is less than 50ms.

In such scheme, described step b comprises: step b1: according to the distribution range of defect time constant, determine nonlinear fitting formula; Step b2: the optimum undetermined parameter that obtains nonlinear fitting by iterative computation.

In such scheme, described step b1 comprises: suppose that defect capture and dispose procedure are separate processes, transient current can be expressed as the summation of different index component,

$I_{\mathrm{fitted}}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{a}_i\exp (-t/{\mathrm{\τ}}_i)+I_{\∞}$

Wherein, a _icorresponding to τ _irelative defect concentration, treat fitting parameter; τ _idefect time constant, predefine parameter; N has τ _ithe exponential term item number of parameter, need determine according to fitting precision; I _∞be current constant item, treat fitting parameter; Estimation defect time constant distribution range, chooses suitable τ _isequence makes substantially to cover defect time constant scope.

In such scheme, obtain the optimum undetermined parameter of nonlinear fitting described in step b2 by iterative computation, the condition that iteration stops is | I _data-I _fitted| ²minimum.

In such scheme, the relation of the time constant of defect described in step c and relative density is with τ _ifor horizontal ordinate, a _igraph of a relation for ordinate.Described in step c, according to relative density, determining defect time constant, is the relation based on this defect time constant and relative density, according to a _iamplitude determine the defect time constant affect curent change.

(3) beneficial effect

From technique scheme, can find out, the present invention has following beneficial effect:

1, the method for extracting defect time constant by transient current spectrum provided by the invention, based on current parameters, can avoid the impact of electric capacity, and equipment needed thereby is simple, only require power supply to there is time-sampling function, can effectively meet the demand that general user characterizes the long device defects time constant of short grid.

2, this method of obtaining the long device defects time constant of short grid that the present invention proposes, is the method based on current measurement, to the time constant of the contributive defect of electric current, all can effectively detect.The technology the present invention relates to is without expensive specialized equipment and abundant tester, and do not relate to the interference problem of edge capacitance.

Accompanying drawing explanation

Fig. 1 extracts the method flow diagram of defect time constant according to the embodiment of the present invention by transient current;

Fig. 2 carries out the method flow diagram of nonlinear fitting to surveyed transient current in Fig. 1;

Fig. 3 is drain current time domain response and the matched curve under target biasing according to the device of the embodiment of the present invention;

Fig. 4 is according to the defect time constant of the embodiment of the present invention and the graph of a relation of relative density.

Embodiment

For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with specific embodiment, and with reference to accompanying drawing, the present invention is described in more detail.Embodiment adopts GaN HEMT device as specimen.

Fig. 1 extracts the method flow diagram of defect time constant by transient current according to the embodiment of the present invention, the method comprises the following steps:

Step a: the transient response of measuring GaN HEMT drain current.

In this step, described leakage current transient response refers to and applies after voltage bias when device, and the response of leakage current in time domain, is denoted as I _data.The minimum sampling interval that requires testing apparatus to gather current signal is less than 50ms.

In Fig. 4, black square is the measured value of the drain current time domain response of device under target biasing.Wherein target biasing is by tester's self-defining.

Step b: surveyed transient current is carried out to nonlinear fitting; Fig. 2 carries out the method flow diagram of nonlinear fitting to surveyed transient current in Fig. 1, as shown in the figure, comprise the following steps:

Step b1: according to the distribution range of defect time constant, determine nonlinear fitting formula;

Suppose that defect capture and dispose procedure are separate processes, transient current can be expressed as the summation of different index component,

$I_{\mathrm{fitted}}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{a}_i\exp (-t/{\mathrm{\τ}}_i)+I_{\∞}---\left(1\right)$

Wherein, a _icorresponding to τ _irelative defect concentration, treat fitting parameter; τ _idefect time constant, predefine parameter; N has τ _ithe exponential term item number of parameter, need determine according to fitting precision; I _∞be current constant item, treat fitting parameter; Estimation defect time constant distribution range, chooses suitable τ _isequence makes substantially to cover defect time constant scope.

The defect time constant that affects device creepage in GaN HEMT device generally in ms level to s level, so our predefined timeconstantτ _ineed to cover as far as possible this scope.Consider that n is got τ _ithe total item of sequence, if the larger iteration time of n is longer, n more hour between fitting of constant precision less, so must take into account operation time and operational precision.N gets 46 in the present embodiment, τ ₁～τ ₄₆the selection of sequence is 0.001,0.002,0.003,0.004,0.005,0.006,0.007,0.008,0.009,0.01,0.02,0.03,0.04,0.05,0.06,0.07,0.08,0.09,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1,2,3,4,5,6,7,8,9,10,20,30,40,50,60,70,80,90,100.

Step b2: the optimized parameter that obtains nonlinear fitting by iterative computation;

Described optimized parameter comprises a _iand I _∞, all by nonlinear fitting, obtain.Utilize origin software to carry out nonlinear fitting, during nonlinear fitting, the condition of iteration termination is | I _data-I _fitted| ²minimum.

Step c: set up the relation of defect time constant and relative density, and determine defect time constant according to relative density.

Relation described in step c is with τ _ifor horizontal ordinate, a _igraph of a relation (Fig. 4) for ordinate.In figure, show and have 4 peak values, the time constant that these peak values are corresponding is the defect time constant that leakage current is played to material impact, is respectively 55ms, 500ms, 3s, 20s.These defect time constants and relative density back substitution, in fitting formula, can be obtained to matched curve (matched curve in Fig. 3), and degree of fitting reaches 0.994, has verified the accuracy of calculating.

Above-described specific embodiment; object of the present invention, technical scheme and beneficial effect are further described; institute is understood that; the foregoing is only specific embodiments of the invention; be not limited to the present invention; within the spirit and principles in the present invention all, any modification of making, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.

Claims (8)

1. by transient current spectrum, extract a method for defect time constant, it is characterized in that, the method comprises:

Step a: the transient response of measuring GaN HEMT drain current;

Step b: surveyed transient current is carried out to nonlinear fitting;

Step c: set up the relation of defect time constant and relative density, and determine defect time constant according to relative density.

2. the method for extracting defect time constant by transient current spectrum according to claim 1, is characterized in that, the transient response of drain current described in step a is to be applied in after voltage bias when device, and the response of drain current in time domain, is denoted as I _data.

3. the method for extracting defect time constant by transient current spectrum according to claim 1, is characterized in that, measures the transient response of GaN HEMT drain current described in step a, and the minimum sampling interval that testing apparatus gathers current signal is less than 50ms.

4. the method for extracting defect time constant by transient current spectrum according to claim 1, is characterized in that, described step b comprises:

Step b1: according to the distribution range of defect time constant, determine nonlinear fitting formula;

Step b2: the optimum undetermined parameter that obtains nonlinear fitting by iterative computation.

5. the method for extracting defect time constant by transient current spectrum according to claim 4, is characterized in that, described step b1 comprises:

Suppose that defect capture and dispose procedure are separate processes, transient current can be expressed as the summation of different index component,

$I_{\mathrm{fitted}}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{a}_i\exp (-t/{\mathrm{\τ}}_i)+I_{\∞}$

Wherein, a _icorresponding to τ _irelative defect concentration, treat fitting parameter; τ _idefect time constant, predefine parameter; N has τ _ithe exponential term item number of parameter, need determine according to fitting precision; I _∞be current constant item, treat fitting parameter; Estimation defect time constant distribution range, chooses suitable τ _isequence makes substantially to cover defect time constant scope.

6. the method for extracting defect time constant by transient current spectrum according to claim 4, is characterized in that, obtains the optimum undetermined parameter of nonlinear fitting described in step b2 by iterative computation, and the condition that iteration stops is | _idata-I _fitted| ²minimum.

7. the method for extracting defect time constant by transient current spectrum according to claim 1, is characterized in that, the relation of the time constant of defect described in step c and relative density is with τ _ifor horizontal ordinate, a _igraph of a relation for ordinate.

8. the method for extracting defect time constant by transient current spectrum according to claim 1, is characterized in that, described in step c, according to relative density, determines defect time constant, is the relation based on this defect time constant and relative density, according to a _iamplitude determine the defect time constant affect curent change.

Images