JP4774545B2 - Method for obtaining value of relative permittivity and device for obtaining relative permittivity value - Google Patents

Description

  In particular, the present invention relates to a technique capable of obtaining an accurate relative dielectric constant in an LSI product without destroying the product.

  In the LSI manufacturing process, in order to maintain a high product yield, various device inspections and evaluation methods are used to check the stable operation of the device and the quality of the members used for manufacturing the LSI product.

  For this purpose, for quality control in the semiconductor pre-process, quality inspection samples are periodically created, and film thickness and refractive index variations and measurement values are managed using a film thickness inspection device. In case of deviation from the standard, the device or member is corrected. Also, variations in processing dimensions and measurement values are measured by a scanning electron microscope for length measurement, and if the measurement value is not within the standard, the apparatus or member is corrected. That is, at this stage, since the electrical circuit is not yet completely completed, electrical measurement cannot be performed, and therefore, management is performed using the above method. Therefore, after completion of the predetermined manufacturing process of the product with LSI, the wiring resistance value and the inter-wiring capacitance value are measured, and the pass / fail judgment is made by confirming that each value is within the specified range. Made.

  By the way, software called capacitance calculation simulator (device simulator) for predicting device characteristics (electric resistance R, capacitance C, inductance L, current I-voltage V characteristics) of LSI (for example, resistance / capacitance of SYNOPSYS) Analysis software (Raphael)) is commercially available.

  FIG. 12 shows a flow chart of calculation by a conventional device simulator (resistance / capacitance analysis software (Raphael) of SYNOPSYS). Necessary input parameter values are a structure parameter value and a physical property parameter value. The structural parameter value is a characteristic value having dimensions such as the thickness of each laminated film of LSI products, wiring width, wiring spacing, transistor gate length, gate spacing, and the like. These are characteristic values such as the refractive index, relative dielectric constant, and resistivity of the film. When a physical property parameter value or a design value of a bulk material that has been measured in advance is input as an initial value of the input parameter, device characteristics are obtained by calculation by the device simulator as described above.

  However, the calculated value (predicted value) by this device simulator does not necessarily match the actually measured device characteristic value. The reason is that the physical property parameter value and the structural parameter value are affected by various processes during the manufacture of the LSI, and the physical property and structural parameter value change from the initial values. In addition, since the change of the physical property parameter value is almost impossible to capture by appearance observation such as cross-sectional observation, even if there is a variation, the deviation cannot be correctly captured. In other words, a correct value could not be measured without destroying the product.

  As described above, destroying an LSI product in process and obtaining a cross-sectional image of the LSI product using a scanning or transmission electron microscope, from which structural parameter values such as wiring dimensions and insulating film thickness In the conventional method of giving the wiring resistance value and the inter-wiring capacitance value as an LSI product, the physical property value associated with the damage to the insulating film from the management of the wiring height and cross-sectional shape We could not correctly grasp the fluctuation of electrical characteristics as the final product due to the change. This is because there has been no method for analyzing variations in structural parameter values and physical property parameter values received from an LSI manufacturing process using L, C, and R device characteristics as input parameters.

  A method has been proposed (Japanese Patent No. 3208421) in which the cross-sectional structure of an LSI is digitized by image processing to obtain the dimension value and shape of wiring and the film thickness value of an insulating film.

Then, according to the proposed technique, the predicted values of the wiring resistance value and the inter-wiring capacitance value are obtained by measuring the resistivity of the wiring material and the relative dielectric constant of the insulating film material in advance and using them as input variables. Can be calculated. Therefore, there is a temporary effect in that it is possible to quantitatively estimate the amount of displacement and the material property value from the design value of the completed LSI product from the amount of displacement with respect to the predicted value.
Japanese Patent No. 3208421

However, in the technique of Patent Document 1,
(1) To obtain a cross-sectional image, the sample must be destroyed. (2) In the analysis of a small number of cross-sectional images, the dimensional value and shape of the wiring, the average value of the insulating film thickness value, and the variation between samples. It is difficult to grasp.

  Furthermore, considering the application to very fine LSI wiring, the effective resistivity of the wiring material changes due to surface scattering of the current flowing in the wiring, giving an accurate prediction value of the wiring resistance value. Problems that can not be derived.

  As described above, a technique for correctly evaluating and grasping in advance the characteristic value fluctuation due to the influence during the process is also desired for estimating the cause of the decrease in the yield of the LSI product due to the variation or change in the above values.

  Therefore, the problem to be solved by the present invention is to solve the above problem. That is, it is to provide a technique capable of correctly evaluating and grasping in advance a characteristic value variation due to an influence in the process, which is a cause of the cause of a decrease in the yield of an LSI product due to variations and changes in the above values. In particular, the present invention provides a technique that is fragile because it has a porous structure, and therefore can obtain a correct value of a relative dielectric constant that easily fluctuates due to the action of some force in the manufacturing process without breaking the product.

The above-mentioned problem is a method for obtaining a value of a dielectric constant of an insulator in a sample in which a conductor is provided in a predetermined pattern in the insulator,
Structural parameter value measuring step for measuring the thickness of the insulator in the sample, the depth of penetration of the conductor into the stopper film provided in the lower layer of the insulator, and / or the structural parameter value of the distance between the conductors When,
A capacitance value measuring step for measuring a capacitance value between the conductors in the sample;
Capacitance for calculating a capacitance value by a predetermined capacitance calculation simulator using a virtual value of the relative dielectric constant of the insulator and a structure parameter value corresponding to the measurement value of the structure parameter value measurement step A value calculation step;
The relationship X between the measurement value obtained in the structural parameter value measurement step and the capacitance value obtained in the capacitance value measurement step, the structure parameter value used in the capacitance value calculation step, and the A comparison step for comparing whether or not the relationship Y with the capacitance value calculated in the capacitance value calculation step matches;
And determining the virtual value of the relative dielectric constant when the relation X and the relation Y are matched in the comparison step as the relative dielectric constant value of the insulator. This is solved by a method for obtaining the value of the relative dielectric constant.

An apparatus for obtaining a value of a dielectric constant of an insulator in a sample in which a conductor is provided in a predetermined pattern in the insulator,
The thickness of the insulator in the sample, the depth of penetration of the conductor into the stopper film provided in the lower layer of the insulator, and / or the structural parameter value of the conductor spacing, and the conductor in the sample A relation X calculation means for obtaining a relation X with a measured capacitance value between;
A capacitance value calculating means for calculating a capacitance value based on a virtual value of a relative dielectric constant of the insulator and a structure parameter value;
A relationship Y calculating means for obtaining a relationship Y between the structural parameter value used for the calculation by the capacitance value calculating means and the calculated capacitance value;
Comparing means for comparing whether or not the relationship X calculated by the relationship X calculating means and the relationship Y calculated by the relation Y calculating means match;
When the relation X and the relation Y match as a result of the comparison by the comparison means, the virtual relative dielectric constant value of the corresponding calculated capacitance value is output as the relative dielectric constant value of the insulator. It is solved by a relative dielectric constant value obtaining apparatus characterized by comprising output means.

  The exact dielectric constant value of the insulating film of the LSI product can be known by nondestructive. Therefore, the obtained information can be fed back to the LSI manufacturing process to help improve the quality of the LSI product.

  The present invention uses a TEG pattern having regularity, and from a capacitance value between wirings measured after reaching a stage where electrical measurement can be performed as a device made on a substrate, as well as wiring dimensions and insulating film films. The relative permittivity value can be obtained nondestructively from the structure value such as thickness. Here, examples of the regular TEG (Test Element Group) pattern include an equal wiring pitch pattern TEG (equal wiring pitch comb pattern TEG) or an equal wiring density pattern TEG (equal wiring density comb pattern TEG).

  That is, the present invention is a method for determining the value of the dielectric constant of an insulator in a sample in which a conductor is provided in a predetermined pattern in the insulator. Then, a structural parameter value measuring step for measuring a structural parameter value of the thickness of the insulator in the sample (TEG), the penetration depth of the conductor into the stopper film provided in the lower layer of the insulator, and / or the distance between the conductors Have Moreover, it has a capacitance value measuring step for measuring a capacitance value between conductors in the TEG. Also, using the virtual value of the relative dielectric constant of the insulator of TEG and the structure parameter value (design value), a predetermined capacitance calculation simulator (for example, resistance / capacitance analysis software (Raphael) of SYNOPSYS): However, the present invention includes a capacitance value calculating step for calculating a capacitance value. In addition, the relationship X between the measurement value obtained in the structural parameter value measurement step and the capacitance value obtained in the capacitance value measurement step, the structure parameter value used in the capacitance value calculation step, and the static value. A comparison step for comparing whether or not the relationship Y with the capacitance value calculated in the capacitance value calculation step matches. Further, there is a determination step for determining that the virtual value of the relative dielectric constant when the relation X and the relation Y match in the comparison step is the value of the dielectric constant of the insulator.

  In addition, the present invention is an apparatus for obtaining a value of a relative dielectric constant of an insulator in a sample in which a conductor is provided in a predetermined pattern in the insulator. Then, the thickness of the insulator in the sample (TEG), the depth of penetration of the conductor into the stopper film provided in the lower layer of the insulator, and / or the structural parameter value of the conductor interval, and the conductivity in the sample (TEG) It has a relation X calculation means for obtaining a relation X with the capacitance value between the bodies. Also, a capacitance value calculation means (capacitance calculation simulator (for example, a resistance of SYNOPSYS, Inc.) that calculates a capacitance value based on a virtual value of dielectric constant of an insulator and a structural parameter value (design value)・ Capacity analysis software (Raphael): Of course, it is not limited to this)). Further, there is a relationship Y calculating means for obtaining a relationship Y between the structural parameter value used for the calculation by the capacitance value calculating means and the calculated capacitance value. Further, a comparison unit is provided for comparing whether or not the relationship X obtained by the relationship X calculation unit matches the relationship Y obtained by the relationship Y calculation unit. Further, when the relation X and the relation Y match as a result of the comparison by the comparison means, the virtual relative dielectric constant value of the corresponding calculated capacitance value is output as the value of the relative dielectric constant of the insulator. It has an output means.

  An object of the present invention is to obtain changes in desired physical property parameter values and structural parameter values from changes in device characteristics. Then, using a structural parameter value having regularity, here, a TEG having a predetermined wiring interval or wiring width, a change rate of device characteristics or waveform data with respect to a change in the structural parameter value is acquired.

The method of the present invention includes the steps as shown in FIG.
[Step 1]
FIG. 2 is a plan view showing the wiring of the comb pattern TEG used for the evaluation of the present invention. 3 is a cross-sectional view taken along a broken line XX ′ in FIG. Line and Space in FIG. 3 represent a wiring width and a wiring interval. 2 and 3, A is a comb-like conductor pattern, B is a lead wire, C is an electrode terminal, D is a substrate, E is a low-k film, F is an etching stopper film, G is a low-k film, and H is a cap. A film, I is a barrier film, and J is a passivation film.

  In the case of distinguishing the difference in the design value of Sn as the actually measured wiring interval and n as the wiring interval, in the case of the equal wiring pitch comb pattern TEG, the wiring pitch is always constant regardless of the value of Sn. When the equal wiring density comb pattern TEG is used, Line and Space in FIG. 2 are equal values.

  First, comb patterns TEG with equal wiring pitch or equal wiring density are created. Then, in the process of creating the TEG, the following three structural parameter values are measured and acquired.

  The film thickness is measured by spectroscopic ellipsometry, the penetration depth of the wiring into the stopper film is measured by an optical interference type three-dimensional pattern dimension measuring machine, and the wiring interval is measured by a scanning length measuring electron microscope. In this measurement, all are measured nondestructively.

The evaluation method using the optical interference type three-dimensional pattern dimension measuring machine (OCD measuring apparatus) is explained in the following literature.
"Line-profile and critical-dimension correlation between a
normal-incidence optical CD metrology system and SEM ''
Weidong Yang, Roger Lowe-Webb, Rahul Korlahalli, Vera G. Zhuang, Hiroki Sasano,
Wei Liu, David Mui, Proc. SPIE Vol. 4689, p. 966-976, Metrology, Inspection, and Process
Control for Microlithography XVI; Daniel J. Herr; Ed. Publication Date: Jul 2002

[Step 2]
The device characteristic value (capacitance value) of each TEG created in Step 1 is measured.
Then, a function F (Sn) representing the dependency on the wiring interval or the wiring width measured in step 1 is obtained. Sn is the measured wiring interval, and the subscript n is a number that distinguishes the difference in the design value of the wiring interval.

  When using the equal wiring pitch comb pattern TEG, the vertical axis plots the device characteristic value measurement result (capacitance value) corresponding to each TEG, and the horizontal axis plots the wiring interval obtained in Step 1, as shown in FIG. Get an approximate curve. The function of this approximate curve is F (Sn) in FIG. Incidentally, the set of points surrounded by circles in FIG. 4 means that it is a group of patterns having the same design dimension, and corresponds to the difference in the subscript n of Sn.

  When using the uniform wiring density comb pattern TEG, the vertical axis represents the device characteristic value measurement result (capacitance value) corresponding to each TEG, and the horizontal axis represents the reciprocal of the wiring interval obtained in step 1 as shown in FIG. And an approximate straight line is obtained. The slope of this approximate straight line is F (Sn) in FIG.

[Step 3]
A desired physical property parameter (relative permittivity) or a predicted value of a structural parameter for which the amount of change is desired is selected. Here, a set of all kinds of physical property parameters and structural parameters is denoted as {P}, and a desired parameter for which the amount of change is to be known is denoted as Pk. k is a subscript that distinguishes the types of elements in the set {P}.

[Step 4]
The predicted value of Pk, the structural parameter obtained in Step 1, and the set {P} of bulk physical property parameters of each insulating film measured in advance are used as a device simulator (for example, resistance / capacitance analysis software of SYNOPSYS) (Raphael)), the same calculation is performed by the device simulator, and the calculated value D of the device characteristic is obtained.
Since D depends on Sn, it is expressed as D (Sn).

[Step 5]
The dependency of D (Sn) obtained in step 4 on the wiring interval or wiring width is expressed as a function.
Here, the function is expressed as f (D (Sn)).

[Step 6]
It is determined whether F (Sn) and f (D (Sn)) match or do not match. It should be noted that a conventional method such as a least square method or a variational method can be used for the determination of match / mismatch.
And when both correspond, it goes to step 7.
If they do not match, the process returns to step 3. Then, another Pk value is selected, the processing in steps 4 and 5 is advanced, and in step 6 a match or mismatch is determined again. Then, a new Pk value is selected and continued until F (Sn) and f (D (Sn)) match.

[Step 7]
The Pk value when F (Sn) and f (D (Sn)) match is output.

Hereinafter, further specific examples will be described.
That is, a phenomenon in which the dielectric constant of the insulating film increases in the LSI manufacturing process, for example, in the case of a wiring structure using a porous low-k material, the dielectric constant is caused by the penetration of a processing solution into the insulating film during the process. The phenomenon of rising is seen.
A method for quantitatively evaluating the change in the dielectric constant will be described below.
FIG. 2 is a plan view showing the wiring of the equal wiring pitch comb pattern TEG used for the evaluation of the present invention, and FIG. 3 is a sectional view. In the wiring formation process of LSI products, a low dielectric constant insulating film is used to lower the dielectric constant between the wirings and reduce the signal delay of the wirings. However, if a porous low-k material is used for the inter-wiring insulating film G in FIG. 3 in order to lower the dielectric constant, the low dielectric constant insulating film may be chemically damaged due to physical damage during processing or a cleaning process after processing. There is a concern that the film is easily damaged and the dielectric constant of the film increases. Therefore, it is very important to correctly obtain the relative dielectric constant of the inter-wiring insulating film G.
Therefore, a specific method for obtaining the relative dielectric constant of the inter-wiring insulating film G will be described.

[Step 1]
First, when the thickness of each film E, F, G, H, I, J in the comb pattern of FIG. 3 was measured by spectroscopic ellipsometry, the film thickness was 150 nm, 30 nm, 150 nm, 37 nm, respectively. 30 nm and 150 nm.
Next, when the penetration depth of the wiring (Cu film) K into the etching stopper film F was measured with an optical interference type three-dimensional pattern dimension measuring machine, it was 8 nm.
For each TEG, the dimension between adjacent wirings (Cu films) K was measured.
Since it is the relative permittivity of the low-k film G that is to be obtained in this embodiment, the necessary structural parameters are only the film thickness, the penetration depth of the wiring into the etching stopper film F, and the wiring spacing. These can be obtained from the patterns shown in FIGS. 2 and 3, and the sample need not be destroyed.

[Step 2]
Next, the interwiring capacitance in each TEG in Step 1 was measured.
Then, the measurement result (measured inter-wiring capacitance value) is plotted on the vertical axis of FIG. 6 and the wiring interval obtained by the measurement in Step 1 is plotted on the horizontal axis, thereby obtaining a predetermined function curve. This function curve is F (Sn) in FIG. Note that the collection of points surrounded by circles in FIG. 6 corresponds to patterns having the same design dimensions. That is, the design values of the wiring intervals in order from the left of the graph correspond to 120 nm, 125 nm, 130 nm, 135 nm, and 140 nm, respectively.

[Step 3]
What is to be obtained is the relative dielectric constant of the Low-k film G in FIG.
Therefore, it is assumed that the relative dielectric constant is 2.4, 2.5, 2.6, 2.7, and 2.8, and is 2.4 for the time being.

[Step 4]
Therefore, first, the assumed value (2.4) in step 3, the structure parameter value (design value), and the physical property parameter value in the bulk of each insulating film measured with a mercury probe in advance are used. Input into the simulator. Here, among the input structural parameter values, the wiring interval values are 120 nm, 125 nm, 130 nm, 135 nm, and 140 nm, which are design values.

[Step 5]
After the input in step 4, the device simulator is activated, and the wiring capacitance values corresponding to the wiring intervals of 120 nm, 125 nm, 130 nm, 135 nm, and 140 nm are calculated.
Note that the value of the inter-wiring capacitance depends on the wiring interval Sn and can be expressed as D (Sn). A predetermined function curve is obtained by plotting the calculation result (calculated inter-wiring capacitance value) on the vertical axis in FIG. 6 and the design wiring interval on the horizontal axis. This function curve is f (D (Sn)) in FIG.

[Step 6]
It is compared whether F (Sn) obtained in step 2 and f (D (Sn)) obtained in step 5 match.
If they match, the relative dielectric constant of the low-k film G is output as it is, assuming that it is the input value (2.4) of the relative dielectric constant used in the calculation of the device simulator in step 5.
If they do not match, the process returns to step 3 to select 2.5 as the next hypothetical value, and thereafter proceed in the same manner. That is, the process is repeated until F (Sn) and f (D (Sn)) match.
In the case of FIG. 6, the measured value of the interwiring capacitance and the calculation result of the device simulator coincided with each other when the relative dielectric constant value input to the device simulator is 2.6. Therefore, the dielectric constant of the inter-wiring insulating film that has undergone the manufacturing process in the present embodiment is 2.6. Since the design value of the relative dielectric constant of this film was 2.3, the relative dielectric constant deteriorated by about 13% due to the influence of the manufacturing process.

  By the way, the relative permittivity in the above embodiment using the equal wiring pitch comb pattern TEG is a result of the curve f (D (Sn)) in FIG. 6 being calculated under the condition of the equal wiring pitch. Whatever the value of the wiring interval of the TEG, the measured value is always on the curve f (D (Sn)) of any Pk value, and does not depend on the value of the wiring interval of the comb pattern TEG. . Therefore, when the TEG pattern is created, if the wiring pitch of the exposure mask is equal, each interval has regularity even if the interval is different, and does not depend on variations such as etching during manufacturing. Such a property cannot be obtained except for an exposure mask having an equal wiring pitch.

Conventionally, measuring a capacitance using a measuring instrument such as an LCR meter, and calculating the dielectric constant of the insulating film from the measurement result of the capacitance is simple, such as a measurement with a mercury probe for a bulk film. If it was not for the structure, it could not be calculated analytically with high accuracy. Therefore, the dielectric constant of a specific insulating film of a structure having complicated physical property parameters such as a multilayer wiring structure cannot be calculated.
However, as described above, if the method of the present invention is employed, the relative dielectric constant of a specific insulating film can be obtained.

In the above embodiment, the equal wiring pitch comb pattern TEG is used. Hereinafter, a specific method for obtaining the relative dielectric constant of the inter-wiring insulating film G when the equal wiring density comb pattern TEG is used will be described.
[Step 1]
First, an equal wiring density comb pattern TEG is produced.
In order to obtain the input parameters in the following step 4, in the middle of the TEG fabrication process, the film thickness is measured using spectroscopic ellipsometry, and the Cu wiring to the etching stopper film is measured using an optical interference type three-dimensional pattern dimension measuring machine. The penetration depth is measured with the structural parameter of the wiring interval using a scanning length measuring electron microscope. This measurement can be performed without breaking the sample.

[Step 2]
Next, the interwiring capacitance in each TEG in Step 1 was measured.
Then, the measurement result (measured inter-wiring capacitance value) is plotted on the vertical axis in FIG. 7 and the inverse value of the wiring interval obtained in the measurement in Step 1 is plotted on the horizontal axis, and a predetermined function curve (straight line) is plotted. ) The slope value (0.604) of this straight line is F (Sn) in FIG.

[Step 3]
What is to be obtained is the relative dielectric constant of the Low-k film G in FIG.
Therefore, assuming that the value of the relative permittivity is a value of 0.1 interval from 2.0 to 3.0 as candidate values, 2.0 is selected for the time being.

[Step 4]
Therefore, first, the assumed selection value (2.0), the structural parameter value (design value) in step 3, and the physical property parameter values of the bulk of each insulating film measured with a mercury probe in advance are obtained. Input into the device simulator. The design values of the wiring pitch are 180 nm, 220 nm, 260 nm, and 360 nm, respectively, and the wiring interval is half of the wiring pitch. Accordingly, 90 nm, 110 nm, 130 nm, and 180 nm are input as the wiring intervals, respectively.

[Step 5]
After the input in step 4, the device simulator is operated, and the value of the capacitance between the wirings corresponding to the wiring intervals of 90 nm, 110 nm, 130 nm, and 180 nm is calculated.
A predetermined straight line is obtained by plotting the calculation result (calculated inter-wiring capacitance value) on the vertical axis and the reciprocal value of the design wiring interval on the horizontal axis. This straight line is f (D (Sn)) in FIG.

[Step 6]
It is compared whether F (Sn) obtained in step 2, that is, the slope of the straight line, and f (D (Sn)) obtained in step 5, that is, whether the slope of the straight line matches.
If they match, the relative dielectric constant of the low-k film G is output as it is, assuming that it is the input value (2.0) of the relative dielectric constant used in the calculation of the device simulator in step 5.
If they do not match, the process returns to step 3 to select the next hypothetical value 2.1 and thereafter proceed in the same manner. That is, the process is repeated until F (Sn) and f (D (Sn)) match.
In this embodiment, F (Sn) and f (D (Sn)) coincide with each other when Pk = 2.6 selected at the seventh time. Therefore, the dielectric constant of the inter-wiring insulating film that has undergone the manufacturing process in the present embodiment is 2.6. Since the design value of the relative dielectric constant of this film was 2.3, the relative dielectric constant deteriorated by about 13% due to the influence of the manufacturing process.

  When the equal wiring density comb pattern TEG is used as in the present embodiment, the relative permittivity is determined only by the inclination of the approximate line in FIGS. It is possible to measure the relative dielectric constant without depending on the value of the capacitance component. The influence on the measured value of the relative permittivity with respect to the shift of the finished wiring width is, for example, when all the wiring widths are shifted by a fixed amount, the influence on the measured value of the relative permittivity with respect to the shift of the wiring width is shown in FIG. The relationship is as shown by the solid line. The curve shown by the broken line in FIG. 9 shows the relative dielectric constant measured with respect to the wiring width shift when the relative dielectric constant measurement is performed using only one point data of the wiring interval 90 nm pattern without using this measurement method. It is an influence. The broken line changes more steeply than the solid line, and it can be seen that this measurement method can obtain an accurate measured value of the relative dielectric constant with respect to the wiring width shift. Even when the wiring width is shifted by a fixed ratio, the comparison between the present measurement method for measuring the relative permittivity and the measurement method using only one point data of the wiring spacing 90 nm pattern shows that the former is a solid line in FIG. In this case, the broken line changes more rapidly than the solid line, and it can be seen that an accurate measured value of the relative dielectric constant can be obtained by this measurement method. Note that, unless it is an equal wiring density comb pattern TEG, the relationship between the reciprocal of the capacitance value and the wiring interval does not have a linear distribution as shown in FIGS. It is necessary.

FIG. 11 is a schematic diagram of an apparatus in which the method of the present invention is implemented.
That is, steps 3 to 7 in FIG. 1 are executed by a computer described below.

In FIG. 11, 1 is a control means in the computer, 2 is an input means, and 3 is a storage means.
11 is a structure parameter value of the film thickness of the insulator in each TEG, the penetration depth of the conductor (Cu film) into the stopper film provided in the lower layer of the insulator, and / or the distance between the conductors (Cu film). , A relation X calculation means for obtaining a relation X with a capacitance value between conductors in the TEG.
Reference numeral 12 denotes a capacitance value calculating means for calculating a capacitance value based on the virtual value of the relative dielectric constant of the insulator and the structure parameter value (design value).
Reference numeral 13 denotes relation Y calculation means for obtaining a relation Y between the structural parameter value used for calculation by the capacitance value calculation means 12 and the calculated capacitance value.
Reference numeral 14 denotes a comparison unit that compares whether or not the relationship X obtained by the relationship X calculation unit 11 matches the relationship Y obtained by the relationship Y calculation unit 13.
4 shows that when the relation X matches the relation Y as a result of the comparison by the comparison means 14, the virtual relative dielectric constant value used for calculation of the capacitance value stored by the control means 1 is stored from the storage means 3. The output means outputs the read virtual relative dielectric constant value as the relative dielectric constant value of the insulator.

  Then, when the above steps are executed by this apparatus, the relative dielectric constant can be obtained.

Flow chart of the present invention TEG top view Cross section of TEG Device characteristics-wiring spacing graph Device characteristics-wiring spacing graph Wiring capacitance vs. wiring spacing graph Wiring capacitance vs. wiring spacing graph Wiring capacitance vs. wiring spacing graph Extracted k value vs. dimension shift graph Extracted k value vs. dimension shift graph Schematic diagram of the device of the present invention Conventional flowchart

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Control means 11 Relation X calculation means 12 Capacitance value calculation means 13 Relation Y calculation means 14 Comparison means 4 Output means

Patent applicant Next-generation semiconductor material technology research association
Representative Katsumi Udaka

Claims (2)

A method for obtaining a value of a relative dielectric constant of an insulator in a sample in which a conductor is provided in a predetermined pattern in the insulator,
Structural parameter value measuring step for measuring the thickness of the insulator in the sample, the depth of penetration of the conductor into the stopper film provided in the lower layer of the insulator, and / or the structural parameter value of the distance between the conductors When,
A capacitance value measuring step for measuring a capacitance value between the conductors in the sample;
Capacitance for calculating a capacitance value by a predetermined capacitance calculation simulator using a virtual value of the relative dielectric constant of the insulator and a structure parameter value corresponding to the measurement value of the structure parameter value measurement step A value calculation step;
The relationship X between the measurement value obtained in the structural parameter value measurement step and the capacitance value obtained in the capacitance value measurement step, the structure parameter value used in the capacitance value calculation step, and the A comparison step for comparing whether or not the relationship Y with the capacitance value calculated in the capacitance value calculation step matches;
And determining the virtual value of the relative dielectric constant when the relation X and the relation Y are matched in the comparison step as the relative dielectric constant value of the insulator. A method for obtaining the relative dielectric constant value. An apparatus for obtaining a value of a relative dielectric constant of an insulator in a sample in which a conductor is provided in a predetermined pattern in the insulator,
The thickness of the insulator in the sample, the depth of penetration of the conductor into the stopper film provided in the lower layer of the insulator, and / or the structural parameter value of the conductor spacing, and the conductor in the sample A relation X calculation means for obtaining a relation X with a measured capacitance value between;
A capacitance value calculating means for calculating a capacitance value based on a virtual value of a relative dielectric constant of the insulator and a structure parameter value;
A relationship Y calculating means for obtaining a relationship Y between the structural parameter value used for the calculation by the capacitance value calculating means and the calculated capacitance value;
Comparing means for comparing whether or not the relationship X calculated by the relationship X calculating means and the relationship Y calculated by the relation Y calculating means match;
When the relation X and the relation Y match as a result of the comparison by the comparison means, the virtual relative dielectric constant value of the corresponding calculated capacitance value is output as the relative dielectric constant value of the insulator. And a relative dielectric constant value obtaining apparatus.

Images