阅读说明：本技术 一种介电温度系数修正的深能级瞬态谱测试方法 (Deep energy level transient spectrum testing method for dielectric temperature coefficient correction ) 是由 曾慧中 唐义强 孟奔阳 肖化宇 杨潇 张文旭 张万里 李言荣 于 2021-01-06 设计创作，主要内容包括：本发明属于半导体器件的测试技术领域,具体涉及一种介电温度系数修正的深能级瞬态谱测试方法。本发明通过结合DLTS测试的等效电路,分析了MIS结构中绝缘层电容对DLTS信号的影响,再对DLTS测试数据进行分析,计算出MIS结构中绝缘层电容的影响因子α,然后根据DLTS谱线数据和影响因子对原始的DLTS信号谱经行了修正,得到了准确的DLTS信号谱。解决了介电常数对温度有依赖性的材料制作成MIS结构时,现有DLTS测试不准确的问题,最终本发明准确的得到了该类MIS结构的缺陷能级、陷进浓度和截获面积。(The invention belongs to the technical field of testing of semiconductor devices, and particularly relates to a deep energy level transient spectrum testing method for correcting a dielectric temperature coefficient. The method analyzes the influence of the insulation layer capacitance in the MIS structure on the DLTS signal by combining with an equivalent circuit of the DLTS test, analyzes DLTS test data, calculates the influence factor alpha of the insulation layer capacitance in the MIS structure, and corrects the original DLTS signal spectrum according to the DLTS spectral line data and the influence factor to obtain the accurate DLTS signal spectrum. The method solves the problem that the existing DLTS test is inaccurate when the material with the dielectric constant dependent on the temperature is manufactured into the MIS structure, and finally the defect energy level, the trapping concentration and the trapping area of the MIS structure are accurately obtained.)

1. A deep energy level transient spectrum testing method for correcting a dielectric temperature coefficient is characterized by comprising the following steps:

step 1, performing DLTS test on a target sample, collecting capacitance transient information of the target sample at different temperatures to obtain capacitance transient C-t curves at different temperatures, and then randomly selecting two fixed times t₁，t₂The corresponding capacitances are added to obtain capacitance variation at different temperatures, and the capacitance variation Δ C can be expressed as:

wherein C (t)₁)、C(t₂) Is t₁，t₂Test capacitance of time, n_TIs the concentration of electrons in the trap, C₀Is a defective capacitance per unit area, N_DIs the doping concentration, τ_nA time constant; DLTS Signal usage Δ C/C₀To show that the original DLTS signal spectrum diagram is obtained.

Step 2: according to the original DLTS signal spectral line diagram obtained in the step 1, for the formula 1, regarding tau_nDerivative is carried out to obtain a curve extreme point, and the extreme temperature is obtained; assuming that the target sample has a defect with a defect energy level Delta E, the time constant tau corresponding to the defect_nComprises the following steps:

wherein, delta_nIs the defect's intercept area, T is the temperature, K is the Boltzmann constant, Δ E is the defect's energy level, γ_n＝3.25×10²¹(m_n/m_o)cm^-2·2^-1·K^-1，m_nIs the effective mass of density of electronic states in a semiconductor, m_oIs the effective mass of free electrons; rewritten for equation (2) as:

determining the defect energy level Delta E and the interception area Delta of the sample by the formula (3)_nConcentration of trap N_TAnd (5) waiting for parameters, and finishing DLTS measurement.

And step 3: for the MIS structure, the total capacitance value C is obtained in the capacitance transient C-t curve obtained by testing_mIs an insulating layer capacitor C_iAnd depletion layer capacitance C_dAlso considering the depletion layer capacitance C_iAnd depletion layer capacitance C_dInfluence of, total capacitance value C_mExpressed as:

wherein, C_m(t) is the total capacitance at time t, C_iIs an insulating layer capacitor, C_dIs a depletion layer capacitance, C_d(t) is the depletion layer capacitance at time t; namely, the capacitance transient curve measured in the step 1 is C_mT, change in total capacitance Δ C_mAnd depletion layer capacitance change Δ C_dThe satisfaction between can be expressed as:

by the formula (5), Δ C in the test of DLTS_d/C_d(t₂) < 1, the above formula can be simplified as:

wherein alpha is an influence factor and is at any time t₀Depletion layer capacitance C of_d(t₀) And a stabilized depletion layer capacitance C_dThe difference between (∞) is not large; c in the above formula (5)_d(t₂) By substitution with C_d(∞) has little effect on α, which can be expressed as:

calculating an influence factor alpha through a formula (7) and DLTS test data in the step 1 to obtain a relation between the test temperature and the influence factor;

and 4, step 4: the influence factor alpha is greater than 1, and the formula (7) shows that in the DLTS test of the MIS structure, the DLTS signal is attenuated by a certain proportion due to the existence of the insulation layer capacitance, and the attenuation proportion is equal to the influence factor alpha;

and (3) correcting the DLTS spectral line according to the data obtained by testing the sample in the step 1, the original DLTS signal spectral line diagram and the alpha value calculated in the step 3, namely the corrected DLTS is equal to the original DLTS signal multiplied by alpha, and obtaining the corrected DLTS signal spectral line diagram, the defect energy level delta E and the interception area delta_nAnd trap concentration N_T。

Technical Field

The invention belongs to the technical field of testing of semiconductor devices, relates to a DLTS (digital Living transform System) measuring method based on a phase-locked amplification technology, and particularly relates to a deep energy level transient spectrum testing method for correcting a dielectric temperature coefficient.

Background

Deep Level Transient Spectroscopy (DLTS) is an important technical means for researching and detecting semiconductor impurities, defect deep levels, interface states and the like in the field of semiconductors. When the method is applied to the field of semiconductors, a DLTS spectrum representing the distribution of impurities, defect deep energy levels and interface states in a semiconductor forbidden band range along with temperature (namely energy) can be given; DLTS can explain the cause of degradation in electrical characteristics of semiconductor devices in terms of changes in microscopic physical quantities.

The DLTS technique was first applied to samples of asymmetric PN junction or schottky junction structures. The detection sensitivity of the technology is usually one ten thousandth of doping concentration in a semiconductor material or even lower, and a plurality of information such as a majority carrier trap, a minority carrier trap, trap concentration and distribution, a trap energy level, an interception area and the like can be obtained.

With the development of DLTS technology, which is applied to metal-oxide-semiconductor field effect transistors (MOS field effect transistors), which are often substrates of n-type Si or p-type Si, the capacitance of the substrate and the capacitance of the insulating layer in such samples do not change much substantially over the entire temperature range, and the dielectric constant is not strongly dependent on temperature. However, for materials with a temperature-dependent dielectric constant (e.g., haHfO oxide)₂Strontium titanate STO), and is also a metal-insulator-semiconductor structure (MIS structure). Compared with a sample of a Schottky junction structure, the structure has new influence on DLTS test due to the newly added insulating layer. There is a phenomenon on the STO bombarded with Ar +: as the temperature decreases, the conductance and capacitance of the bombarded STO conductive layer increases dramatically, and the dielectric constant changes dramatically with changes in temperature. For such samples, the influence of the dielectric constant change on the DLTS needs to be considered, so that the applicability of the current DLTS technology has certain limitation.

Disclosure of Invention

In view of the above problems or disadvantages, the present invention provides a method for testing a deep energy level transient spectrum with a modified dielectric temperature coefficient for a MIS structure device with a temperature-dependent dielectric constant, so as to improve the applicability of the DLTS technique and solve the problem that the current DLTS technique is inaccurate in dealing with such a situation.

A method for testing a dielectric temperature coefficient modified deep energy level transient spectrum comprises the following steps:

step 1, performing DLTS test on a target sample, collecting capacitance transient information of the target sample at different temperatures to obtain capacitance transient C-t curves at different temperatures, and then randomly selecting two fixed times t₁，t₂The corresponding capacitances are added to obtain capacitance variation at different temperatures, and the capacitance variation Δ C can be expressed as:

wherein C (t)₁)、C(t₂) Is t₁，t₂Test capacitance of time, n_TIs the concentration of electrons in the trap, C₀Is a defective capacitance per unit area, N_DIs the doping concentration, τ_nA time constant. DLTS signals often use Δ C/C₀To indicate that the time t is changed₁，t₂Multiple curves can be obtained to obtain the original DLTS signal spectral line diagram.

Step 2: at a certain temperature, Δ C will have an extreme value, and according to the original DLTS signal spectrum diagram obtained in step 1, with respect to τ in equation 1_nAnd (5) obtaining a derivative to obtain a curve extreme point, namely obtaining the extreme temperature. At different times t₁，t₂Multiple curves can be obtained, each curve having an extreme temperature and a corresponding time constant τ_n。

Assuming that the target sample has a defect with a defect energy level Delta E, the time constant tau corresponding to the defect_nComprises the following steps:

wherein, delta_nIs the defect's intercept area, T is the temperature, K is the Boltzmann constant, Δ E is the defect's energy level, γ_n＝3.25×10²¹(m_n/m_o)cm^-2·s^-1·K^-1，m_nIs the effective mass of density of electronic states in a semiconductor, m_oIs the effective mass of free electrons. Rewritten for equation (2) as:

determining the defect energy level Delta E and the interception area Delta of the sample by the formula (3)_nConcentration of trap N_TAnd (5) waiting for parameters, and finishing DLTS measurement.

And step 3: for the MIS structure, the total capacitance value C is obtained in the capacitance transient C-t curve obtained by testing_mCapacitor C of insulating layer_iAnd depletion layer capacitance C_dAlso considering the depletion layer capacitance C_iAnd depletion layer capacitance C_dInfluence of, total capacitance value C_mCan be expressed as:

wherein, C_m(t) is the total capacitance at time t, C_iIs an insulating layer capacitor, C_dIs a depletion layer capacitance, C_d(t) is the depletion layer capacitance at time t. Namely, the capacitance transient curve measured in the step 1 is C_mT, change in total capacitance Δ C_mAnd depletion layer capacitance change Δ C_dThe satisfaction between can be expressed as:

by the formula (5), Δ C in the test of DLTS_d/C_d(t₂) < 1, the above formula can be simplified as:

wherein alpha is an influence factor and is at any time t₀Depletion layer capacitance C of_d(t₀) And a stabilized depletion layer capacitance C_dThe difference between (∞) is not large; c in the above formula (5)_d(t₂) By substitution with C_d(∞) has little effect on α. α can be expressed as:

and (3) calculating an influence factor alpha through a formula (7) and the DLTS test data in the step 1 to obtain a relation between the test temperature and the influence factor.

And 4, step 4: it is obvious that the influence factor alpha>As can be seen from equation (7), in the DLTS test for the MIS structure, the DLTS signal is attenuated by a certain proportion due to the presence of the insulation layer capacitance, and the attenuation proportion is equal to the influence factor α. And (3) correcting the DLTS spectral line according to the data obtained by testing the sample in the step 1, the original DLTS signal spectral line diagram and the alpha value calculated in the step 3, namely the corrected DLTS is equal to the original DLTS signal multiplied by alpha, and obtaining the corrected DLTS signal spectral line diagram, the defect energy level delta E and the interception area delta_nAnd trap concentration N_T。

The method analyzes the influence of the insulation layer capacitance in the MIS structure on the DLTS signal by combining with an equivalent circuit of the DLTS test, analyzes DLTS test data, calculates the influence factor alpha of the insulation layer capacitance in the MIS structure, and corrects the original DLTS signal spectrum according to the DLTS spectral line data and the influence factor to obtain the accurate DLTS signal spectrum. The method solves the problem that the existing DLTS test is inaccurate when the material with the dielectric constant dependent on the temperature is manufactured into the MIS structure, and finally the defect energy level, the trapping concentration and the trapping area of the MIS structure are accurately obtained.

Drawings

FIG. 1 is a flow chart of DLTS test modification of the present invention;

FIG. 2 is a graph of capacitance Cbias U of a target sample obtained in accordance with an example;

FIG. 3 is a C-t curve of a target sample at different temperatures according to the example

FIG. 4 is a spectrum diagram of the original DLTS signal obtained in step 1 of the example;

FIG. 5 is a graph of impact factor versus temperature for target samples of examples;

FIG. 6 is a plot of a DLTS signal spectrum after modification by an embodiment;

FIG. 7 is a graph showing the emissivity of a sample as a function of temperature before and after modification in accordance with the example.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples.

The specific implementation mode is as follows: a deep energy level transient spectrum testing method for correcting dielectric temperature coefficient is disclosed, wherein a target sample is an MOS device with a metal-oxide layer-semiconductor layer structure.

Step 1, the target sample was subjected to C-V testing (see FIG. 2) and it was determined that the sample bias was-5V, the pulse height was 3V and the pulse width was 10 ms. And ensuring that the test parameters are within the range of the test system. And performing DLTS test on the sample, and collecting capacitance transient information of the sample at different temperatures for analysis. Obtaining capacitance transient C-t curves at different temperatures, and selecting two fixed times t₁，t₂(see FIG. 3) and its corresponding change in capacitance can be expressed as:

wherein C (t)₁)、C(t₂) Is t₁，t₂Test capacitance of time, n_TIs the concentration of electrons in the trap, C₀Is a defective capacitance per unit area, N_DIs the doping concentration, τ_nA time constant. DLTS signals often use Δ C/C₀To express, can calculateAnd obtaining a DLTS signal spectrum diagram of a single window rate. Changing the time t₁，t₂Multiple curves can be obtained, resulting in the original DLTS signal profile, as shown in fig. 4. It can be reflected that the peak height of the DLTS curve peak presents a large gradient difference under different rate windows.

Step 2: at a certain temperature, Δ C will have an extreme value, and according to the original DLTS signal spectrum diagram obtained in step 1, with respect to τ in equation 1_nAnd obtaining the extreme point of the curve by derivation. At different times t₁，t₂Multiple curves can be obtained, each curve having an extreme temperature and a corresponding time constant τ_n。

Assuming that the target sample has a defect with a defect energy level Delta E, the time constant tau corresponding to the defect_nComprises the following steps:

wherein, delta_nIs the defect's intercept area, T is the temperature, K is the Boltzmann constant, Δ E is the defect's energy level, γ_n＝3.25×10²¹(m_n/m_o)cm^-2·s^-1·K^-1，m_nIs the effective mass of density of electronic states in a semiconductor, m_oIs the effective mass of free electrons. Rewritten for equation (2) as:

the defect energy level Delta E of the sample is determined to be 0.2 +/-0.001 eV and the interception area Delta_nIs 1.1X 10^-15cm²And trap concentration N_TIs 3.0X 10¹²cm^-3。

And step 3: for the MIS structure, the total capacitance value C is obtained in the capacitance transient C-t curve obtained by testing_mCapacitor C of insulating layer_iAnd depletion layer capacitance C_dAlso considering the depletion layer capacitance C_iAnd depletion layer capacitance C_dShadow ofSound, total capacitance C_mCan be expressed as:

wherein, C_m(t) is the total capacitance at time t, C_iIs an insulating layer capacitor, C_dIs a depletion layer capacitance, C_d(t) is the depletion layer capacitance at time t. Namely, the capacitance transient curve measured in the step 1 is C_mT, change in total capacitance Δ C_mAnd depletion layer capacitance change Δ C_dThe satisfaction between can be expressed as:

by equation 5, Δ C in the test of DLTS_d/C_d(t₂)<<1, the above formula can be simplified as:

wherein alpha is an influence factor and is at any time t₀Depletion layer capacitance C of_d(t₀) And a stabilized depletion layer capacitance C_dThe difference between (∞) is not large. C in the above formula 5_d(t₂) By substitution with C_d(∞) has little effect on α. α can be expressed as:

the influence factor α is calculated by formula (7) and the DLTS test data in step 1, and a relationship diagram of the influence factor and the temperature can be obtained, as shown in fig. 5. The dependence of the sample influence factor on the temperature can be seen, which causes a large error of the DLTS signal at low temperature, and at the same time, causes the peak height of the DLTS curve peak to show a large gradient difference under different frequency windows. And the defect energy level, the interception area and the trap concentration are calculated to have errors.

And 4, step 4: and (3) correcting the DLTS test spectral line, and correcting the obtained data and DLTS spectral line of the sample according to the step 1 and the original DLTS signal spectral line by alpha calculated in the step 2 (as shown in figure 4), so as to obtain a DLTS signal spectral line graph (as shown in figure 6) after the sample is corrected. The defect energy level delta E is 0.17 +/-0.001 eV and the interception area delta can be obtained by calculation_nIs 8.4X 10^-16cm²And trap concentration N_TIs 6.8 multiplied by 10¹¹cm^-3. By analyzing the data before and after correction, it can be obtained that the defect energy level, the trapping area and the trap concentration before and after correction have obvious changes (as shown in fig. 7). As can be seen from fig. 7, at low temperatures, the emissivity e of the two samples before and after correction is 1/τ, and the temperature T, the two curves before and after correction have large deviations in slope and intercept, and the slope also changes to some extent, so that the curves before and after correction have large separations. It is stated that the variation of the dielectric constant with the temperature variation causes the deviation of the calculation result, and the data must be corrected to obtain the accurate result.

According to the embodiment, the influence of the capacitance of the insulating layer in the MIS structure on the DLTS signal is analyzed by combining with the equivalent circuit of the DLTS test, the DLTS test data is analyzed, the influence factor alpha of the capacitance of the insulating layer in the MIS structure is calculated, and then the original DLTS signal spectrum is corrected, so that the accurate DLTS signal spectrum is obtained. The invention solves the problem that the existing DLTS test is inaccurate when the material with the dielectric constant dependent on the temperature is manufactured into the MIS structure, and the invention accurately obtains the defect energy level, the trapping concentration and the trapping area of the MIS structure, thereby being practical and effective.

While the invention has been described with reference to specific embodiments, any feature disclosed in this specification may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise; all of the disclosed features, or all of the method or process steps, may be combined in any combination, except mutually exclusive features and/or steps.