JPH10185878A - Dielectric breakdown measuring device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a dielectric breakdown measuring method that does not require a MOS structure for measuring dielectric breakdown in the measurement of dielectric breakdown characteristics to an insulating film. SOLUTION: An SPM(scanning probe microscope) starts to impress a voltage on a measurement point. First, the SPM monitors a current passing through an oxide film as increasing the value of impressing voltage. Second, the SPM determines whether the oxide film 3a suffers a dielectric breakdown according to the relation in magnitude between the value of a current obtained by monitoring and the first reference value set in advance. Then, at the time when the SPM determines that the oxide film 3a suffers a dielectric breakdown, the SPM outputs data on the value of a voltage impressed on the oxide film 3a of this time. This outputted voltage is a dielectric breakdown voltage. In this way, it is possible to measure a dielectric breakdown voltage through the use of an SPM.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dielectric breakdown measuring method for measuring dielectric breakdown of an insulating thin film of a semiconductor device.

[0002]

2. Description of the Related Art In general, a conventional dielectric breakdown measuring method is applied to a device formed by laminating a silicon substrate, an insulating thin film and a MOS structure for dielectric breakdown measurement between a gate electrode of the MOS structure and the silicon substrate. Is a method of measuring the dielectric breakdown characteristics of an insulating thin film by applying a voltage to the insulating thin film. There have been many research results on this dielectric breakdown measurement method.

[0003] As a dielectric breakdown measuring method, there are measuring methods relating to time zero dielectric breakdown and temporal dielectric breakdown. The measurement method relating to the time zero dielectric breakdown and the time-dependent dielectric breakdown is a very important measurement method from the viewpoint of evaluating various factors affecting the dielectric breakdown voltage and the long-term reliability against the dielectric breakdown.

[0004] Time-dependent dielectric breakdown means that the insulating thin film does not break down at the moment when a voltage or current stress is applied, but if the stress is continued, the insulating thin film breaks down after a certain period of time. It is a phenomenon that does. On the other hand, the time zero dielectric breakdown is a dielectric breakdown that is evaluated in a short time by using a method that is generally performed, such as measurement by applying a lamp voltage with a curve tracer. About both, "Submicron Device" by Mitsumasa Koyanagi (Maruzen Publishing, 1988, pp. 68, 78)
In detail.

[0005]

However, such a conventional dielectric breakdown measuring method has a problem that a MOS structure is required. Since a MOS structure is required, dielectric breakdown depends on the material and shape of the gate electrode, process factors (for example, etching conditions), and the like. Further, since a MOS structure is required, there is a problem that in-line monitoring of dielectric breakdown characteristics cannot be performed during a semiconductor device manufacturing process.

The present invention has been made to solve these problems. In measuring the dielectric breakdown characteristics of an insulating thin film, a MOS structure for measuring the dielectric breakdown is not required. It is an object of the present invention to obtain a dielectric breakdown measuring method that can be easily monitored.

[0007]

Means for Solving the Problems According to a first aspect of the present invention, there is provided a dielectric breakdown measuring method for measuring a dielectric breakdown voltage of a sample having an insulating thin film exposed on the surface, the method comprising: A first step of applying a predetermined voltage between the probe to be measured and the sample while increasing the voltage, and measuring a current flowing between the insulating thin film and the probe, and the current reaches a reference value. A second step of determining the predetermined voltage at the time as a dielectric breakdown characteristic, and using a scanning probe (SPM) as a mechanism for applying the predetermined voltage.
Microscopy) is used.

[0008] The problem solving means according to claim 2 of the present invention is as follows.
A dielectric breakdown measuring method for measuring the dielectric breakdown resistance of a sample having an insulating thin film exposed on the surface, wherein a predetermined voltage is repeatedly applied between the probe and the sample facing the insulating thin film,
A first step of measuring a current flowing between the insulating thin film and the probe, and a second step of determining the number of times the predetermined voltage is repeatedly applied until the current reaches a reference value as insulation breakdown characteristics. And an SPM (Scanning P) as a mechanism for applying the probe and the predetermined voltage.
robe Microscopy).

According to a third aspect of the present invention, there is provided:
The method further includes a third step in which the probe moves on the insulating thin film, wherein the first and second steps are executed for a measurement point reached by the third step, and the third step is executed. Before the predetermined voltage is set to zero.

[0010] The problem solving means according to claim 4 of the present invention is:
(A) (a-1) A predetermined voltage is applied to another insulating thin film corresponding to the insulating thin film while increasing the voltage between the probe and the sample, and the other insulating thin film and the probe are applied. A fourth step of measuring a current flowing between the needle and (a-
2) a fifth step of obtaining the predetermined voltage when the current reaches a reference value in the fourth step;
(A-3) repeating the sixth step in which the probe moves on the another insulating thin film at a first distance with a first distance to obtain a plurality of the predetermined voltages; (B) repeating a first process for obtaining the result of 1 while updating the first distance to obtain a plurality of the first results;
A second distance is set in accordance with a criterion for dividing the plurality of first results into two, and (c) performing the first to third steps using the second distance.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION

Summary of the invention. Devices that can monitor the current-voltage characteristics (I / V characteristics) of insulating thin films include scanning probe microscopes (Scan).
ning Probe Microscopy (SPM). This SPM
Can monitor the I / V characteristics directly from the surface of the insulating thin film. That is, the SPM measurement does not require the MOS structure as described in the related art.

The contents of the measurement by the SPM are described in detail in, for example, Japanese Patent Application Laid-Open No. 8-62229. FIGS. 17 and 18 of the present application respectively correspond to FIGS. 2 (a) and 2 (b) of the above publication. With the atomic force acting between the probe 2 and the sample 3 kept constant, the first measurement point P on the surface of the sample 3 to be measured (for example, the oxide film 3 a) is moved below the probe 2.
The sample 3 to be measured is moved so that 1 comes. Then, at time t1, surface unevenness image data is measured, and at time t2, a sawtooth wave (triangular wave) of, for example, a sample voltage of 20 V is applied to the sample 3.
A voltage is applied to measure I / V characteristic data (for example, measure current values at 100 points from 0 V to a predetermined voltage value). When the measurement of the I / V characteristic data ends at time t3, at time t3
The sample to be measured 3 is moved so that the measurement point P2 on the surface of the sample 3 is located below the probe 2 between t and t4. After that, at time t5, surface unevenness image data is measured at the second measurement point P2, and at time t6, I / V characteristic data is measured. Hereinafter, measurement is similarly performed for the measurement points P3 and P4. In this manner, the SPM measures the surface unevenness image and the I / V characteristics at all of the many measurement points on the surface of the oxide film 3a, which is an insulating thin film, and forms a surface current distribution image based on these. Is possible.

However, it is not possible to measure the dielectric breakdown characteristics of the oxide film 3a using the contents described in the above publication as it is. The reason for this is that, in the measurement of the I / V characteristics, a certain specific voltage value is set first, and no further voltage value is applied.

Therefore, in the measurement by the SPM, the dielectric breakdown characteristics are measured by gradually increasing the value of the applied voltage until the dielectric breakdown occurs in the oxide film 3a. A preferred embodiment of the dielectric breakdown measuring method using the SPM will be described below.

Embodiment 1 FIG. 1 is an operation diagram illustrating a dielectric breakdown measuring method according to Embodiment 1 of the present invention.
FIG. 2 is a timing chart showing I / V characteristics measured by this dielectric breakdown measuring method. In FIG.
The vertical axis represents the sample voltage V applied to the measurement point, the minute current (FN current) I flowing at the measurement point, and the horizontal axis represents time t.
1 and 2, reference numeral 3 denotes a sample to be measured, and P1, P2, and P3 denote measurement points (pixel points). The observation area (S) of the surface of the oxide film 3a provided on the surface of the sample to be measured 3 is exposed.
Many measurement points (for example, 8 × 8 points, 6
4 × 64 points or 128 × 128 points).

FIG. 3 is a flowchart showing a dielectric breakdown measuring method according to the first embodiment of the present invention, and FIG. 4 is a block diagram showing a configuration of an SPM used in the dielectric breakdown measuring method. Hereinafter, the dielectric breakdown measuring method according to the present embodiment will be described with reference to FIGS. First, referring to step 101 of FIG. 3, the feedback loop of FIG. 4 is kept closed, and the interatomic force between the probe 2 provided at the tip of the cantilever 1 and the sample 3 is measured. Control to keep the force constant. The feedback loop is a loop including the cantilever 1, the photodiode detector 5, the filter 9, the switch 15, the differential amplifier 10, the integral amplifier 11, the proportional amplifier 12, the adder 13, the high voltage amplifier 14a, and the piezo element 6. It is. And
The computer 71 moves the sample 3 to be measured in the XY directions (two-dimensional direction) such that the measurement point P1 on the surface of the oxide film 3a is located below the probe 2 (SPM probe).

Next, referring to step 102, at time t1 in FIG. 2, the A / D converter 73 in FIG. 4 outputs a signal around the measurement point P1 at the output terminal T1 connected to the output side of the adder circuit 13. The signal of the surface unevenness image of the oxide film 3a is converted into surface unevenness image data. Then, the computer 71
The surface unevenness image data is stored in the RAM disk 72.

Next, referring to step 103, the computer 71 controls the high-voltage amplifier 14c via the D / A converter 78 and the filter 9a, and at time t2,
The application of the sample voltage V to the oxide film 3a at the measurement point P1 is started. The initial value of the sample voltage when executing step 103 is zero volt.

Next, referring to step 104, the computer 71 gradually increases the voltage value of the sample voltage V.
Specifically, the waveform of the sample voltage output from the high-voltage amplifier 14c controlled by the computer 71 is a sawtooth wave (triangular wave). Further, the high-voltage amplifier 14c sets the voltage value of the sample voltage V to a voltage step ΔV (for example, 0.05 V).
Just raise.

Next, referring to step 105, the preamplifier 22 amplifies the current I and converts it into data via the AD converter 77. Then, the computer 71 monitors the current I by taking in the data.

Next, referring to step 106, the computer 71 determines whether or not the current value of the current I obtained by monitoring is larger than a reference value Iref preset by the operator. If larger, the process returns to step 104. Therefore, steps 104 to 106 are repeated until the current value of the current I obtained by monitoring becomes larger than the reference value Iref. The reference value Iref needs to be larger than a critical current value for determining whether or not dielectric breakdown occurs in order to cause dielectric breakdown of the oxide film 3a. The current value is set by an operator based on the result of an experiment performed in advance.

Next, at step 106, when it is determined at time t3 that the current value of the current I obtained by monitoring is larger than the reference value Iref, the computer 71 refers to step 107 and t3
The voltage value V1 of the sample voltage V of FIG.
To be stored in the RAM disk 72 of the storage device. At the time t3, the voltage value of the sample voltage V is set to zero volt. The reason for this is to prevent an excessive current I from flowing between the probe 2 and the sample 3 via the oxide film 3a in a short time. This prevention does not affect the resistance of the oxide film 3a other than near the measurement point. The voltage value V1 stored in the RAM disk 72 corresponds to the measurement point P of the oxide film 3a.
1 is the dielectric breakdown voltage.

Next, referring to step 108, the computer 71 determines whether or not the surface unevenness image and the dielectric breakdown voltage have been measured for all the measurement points. If these have not been measured for all measurement points, step 10
Return to 1. When returning to step 101, the measurement point P
Similarly, with respect to 2, P3,..., The surface roughness image and the voltage values V2, V3,.
If all the measurement points have been measured in step 108, the process proceeds to step 109.

Referring to step 109, computer 71 reproduces the surface unevenness image of the observation area from the surface unevenness image data for all the measurement points stored in RAM disk 72. Further, the computer 71 edits the breakdown voltage data obtained at each measurement point, and displays a time zero breakdown histogram, which is a kind of breakdown characteristics, on the display 79 as an edited result. The time zero breakdown histogram is obtained by plotting the time zero breakdown voltage on the horizontal axis and the breakdown rate (also referred to as failure rate) on the vertical axis.

FIG. 5 is a graph showing an example of editing a time zero breakdown histogram. This figure shows two types of time-zero breakdown histograms collectively, and was manufactured under the same conditions using a measured sample 3 (silicon substrate) purchased from two different vendors (A, B). 5 shows a time zero breakdown histogram for an oxide film 3a (silicon oxide film). The dielectric breakdown rate on the vertical axis indicates the ratio of the number of measurement points at which dielectric breakdown occurred within a certain sample voltage range to the total number of measurement points in the same vendor, expressed as a percentage. The thickness of the oxide film 3a is 13 nm for both benders. The number of measurement points in the observation area for the time zero breakdown histogram is 16 × 16 for both vendors.

FIG. 6 is a graph showing another editing example of the time zero breakdown histogram. In FIG.
The breakdown voltage data obtained at each measurement point on the sample 3 (silicon substrate) to be measured by the bender is converted into the dielectric breakdown voltage of the oxide film 3a (silicon oxide film). And an edge region in the same active region.

There have been many reports on the edge area. For example, in a manufacturing process of a semiconductor device, a non-uniformity and a stress of a laminated structure composed of Si and SiO ₂ in an edge region surrounded by an isolation oxide film, and a generation of a crystal defect or the like due to the non-uniformity or a stress are considered. is there. The difference in the distribution of the time-zero breakdown voltage as shown in the example of FIG. 6 is also considered to be caused by the non-uniformity, stress, and the like of the above-described laminated structure composed of Si and SiO ₂ .

As described above, the following can be analyzed by creating the time zero dielectric breakdown histogram. That is, as apparent from FIG. 5, the dielectric breakdown voltage of the oxide film 3a on the sample 3 to be measured of A-bender is lower than that of B-bender. The breakdown voltage of A-bender is 0.2% lower than that of B-Vendor.
V is shifted to the low voltage side. Also, as is apparent from FIG. 6, the dielectric breakdown voltage in the edge region is lower than that in the central region. Thus, the C-bender silicon substrate is:
It has location dependency on the degree of dielectric breakdown voltage. Also,
The reason why the distribution center and the spread (half width) of the distribution in the time zero dielectric breakdown histogram of FIG. 6 are slightly different from those of FIG. 5 is considered to be due to the sample 3 to be measured purchased from the C-bender. Furthermore, the distribution of the breakdown voltage distribution (the tail of the distribution) in the edge region in FIG. 6 is partially on the high voltage side because a part of the observation region exists on the thick isolation oxide film. It is thought to be.

Note that the distance between the measurement points for the time zero breakdown histogram for both vendors shown in FIGS. 5 and 6 is 567 nm, and therefore, there is no influence of the charge trapping region described in the second embodiment.

The effects of this embodiment are as follows. (1) The dielectric breakdown characteristics can be measured using the existing SPM. (2) In the dielectric breakdown measuring method according to the present embodiment, it is not necessary to prepare a MOS structure for dielectric breakdown measurement in advance. (3) In the dielectric breakdown measuring method of the present embodiment, since a MOS structure for dielectric breakdown measurement is not required, the breakdown characteristics depend on the material, shape, process factors, and the like of the gate electrode as in the conventional dielectric breakdown measuring method. Not done. (4) In the dielectric breakdown measuring method according to the present embodiment, since the SPM edits and displays the dielectric breakdown data in a histogram, the operator easily evaluates the sample 3 and the oxide film 3a based on the display. be able to. (5) In the dielectric breakdown measuring method according to the present embodiment, measurement on a nanometer scale is possible because SPM is used.

Embodiment 2 SPM measurement has the following drawbacks related to charge capture. 7 and 8 are conceptual diagrams showing the effect of injection of carriers (current I) into the oxide film 3a by the probe 2. As shown in FIG.
By passing a current I between the probe 2 and the sample 3 at 1, a charge trapping region 3b is generated in the oxide film 3a at the measurement point P1. The charge trapping region 3b has a certain width in the surface direction of the oxide film 3a, and changes the internal electric field in the oxide film 3a. The charge trapping region 3b is at the measurement point P
7, when the distance d between the measurement points P1 and P2 is sufficiently longer than the radius of the charge trapping region 3b, as shown in FIG.
There is no effect on the I / V characteristics for However, as shown in FIG. 8, if the distance d is shortened, for example, when the resolution needs to be increased, the charge trapping region 3b becomes the measurement point P2
Partially overlaps the area where the current I is radiated, affecting the measurement point P2, making the measurement of the I / V characteristic unstable. Further, the MOS structure for dielectric breakdown measurement described in the prior art also has a drawback related to charge capture. Hereinafter, a method for measuring the dielectric breakdown that solves this disadvantage will be described.

First, using the SPM, a plurality of time zero breakdown histograms are created for the A-bender dummy sample 3 by changing the distance d. As an example, distances d are 1134 nm, 454 nm, and 303, respectively.
Consider the case of creating four time-zero breakdown histograms of nm and 227 nm. These four values are set in the computer 71 by the operator. This time zero dielectric breakdown histogram is created by applying the first embodiment. The thickness of the oxide film 3a of the dummy sample 3 to be measured is 13 nm so as to correspond to the oxide film 3a of the sample 3 to be measured after setting a distance d described later.

FIG. 9 is a time zero breakdown histogram created in this example. In this time zero breakdown graph, the portion above 20% of the vertical axis is omitted. The time-zero breakdown histogram at a distance d of 1134 nm and the time-zero breakdown histogram at a distance d of 454 nm are almost the same. About 95% of the time zero breakdown histograms at distances d of 454 nm and 1134 nm were concentrated by 16.7 V. On the other hand, the distance d is 303 nm
The distribution of the time-zero dielectric breakdown histogram is shifted to a higher voltage side than in the case where the distance d ≧ 454 nm, and the variation becomes larger. In the time-zero breakdown histogram in which the distance d is 227 nm, the distribution shifts to the higher voltage side and the variation further increases. The phenomenon that the variation in the distribution is shifted to the higher potential side is caused by the charge trapping region 3b in the oxide film 3a. If the distance d becomes shorter, the current I is suppressed by the influence of the charge trapping region 3b, and This is because breakdown is less likely to occur on the low potential side.

The following can be seen from the time zero breakdown histogram of FIG. It can be estimated that the radius of the charge trapping region 3b in the oxide film 3a is at least about 300 nm. Therefore, when trying to measure the distribution of the local breakdown voltage of the oxide film 3a, it can be determined that it is difficult to obtain a resolution at a distance d of 300 nm or less. If the distance d between the measurement points is about 450 nm or more, the dielectric breakdown voltage of the oxide film 3a without the influence of the charge trapping region 3b can be measured. When the distance d is sufficiently longer than 450 nm as indicated by a time zero breakdown histogram in which the distance d is 454 nm or more, the minimum and minimum values of the breakdown voltage of the oxide film 3a are 16.5 V and 16.7 V, respectively. And the difference is almost uniform within 0.2 V (1%).

As described above, the operator grasps the radius (approximately 450 nm) of the trapped charge region 3b from the time-zero breakdown histogram of the dummy sample 3 and then sets the computer 71 to a value sufficiently longer than 450 nm as a reference value. Set to. The computer 71 measures the dielectric breakdown characteristics based on the processing shown in the flowchart shown in FIG. In step 101 of the processing, the computer 71 moves the sample 3 so that the distance d becomes a reference value set by the operator.

As described above, the operator obtains a plurality of time-zero breakdown histograms using the SPM, and determines whether or not a time-zero breakdown histogram that can be identified as the same is used as a criterion. Divide the zero breakdown histogram by two. Then, the distance d is set according to the criterion.

The effects of this embodiment are as follows. That is, (6) the measurement of the time zero dielectric breakdown in which the SPM is not affected by the charge trapping region can be performed.

Embodiment 3 FIG. 10 is an operation diagram illustrating a dielectric breakdown measuring method according to Embodiment 3 of the present invention. FIG. 11 is a timing chart showing I / V characteristics obtained by this dielectric breakdown measuring method. In FIG. 11, the vertical axis represents the sample voltage V applied to the measurement point, the minute current I flowing at the measurement point, and the horizontal axis represents time t. FIG.
0 and FIG. 11, P1 and P2 are measurement points, and a number of measurement points (for example, 8 × 8)
Point, 64 × 64 points, or 128 × 128 points).

FIG. 12 is a flowchart showing a dielectric breakdown measuring method according to the third embodiment of the present invention. Less than,
FIGS. 10 to 1 show the dielectric breakdown measuring method in the present embodiment.
2 and FIG. The flowchart of FIG. 12 adds step 201 to the flowchart of FIG. 3, and replaces steps 107 and 109 with steps 107 ′ and 1 respectively.
09 '. In step 106,
At time t3 in FIG. 11, when it is determined that the current value of the current I obtained by monitoring is larger than the reference value Iref, the process proceeds to step 107 ′, and otherwise, proceeds to step 201. Next, referring to step 201, the computer 71 determines whether or not the voltage value of the sample voltage V is larger than a reference value Vref set in advance by the operator. If it is larger, the process returns to step 103; otherwise, the process returns to step 104. With the addition of step 201, as shown in FIG. 11, the current I appearing at time t2 to t3 is repeatedly applied until the current I reaches the reference value Iref. Note that the reference value Vref is
Set within a range where time zero dielectric breakdown does not occur. This reference value Vref is an I / V measured experimentally in advance.
The operator may make settings with reference to the characteristics.

In step 107 ', the computer 71 sets the number n1, n2,... Of the repetition of the sawtooth wave at the measurement points P1, P2,. To be stored. At time t3 'when it is determined that the current value of the current I obtained by monitoring is larger than the reference value Iref, the voltage value of the sample voltage V is set to zero V.

Referring to step 109 ', the computer 71 reproduces the surface unevenness image data of the observation area from the surface unevenness image data for all the measurement points stored in the RAM disk 72. Further, the computer 71 edits the voltage application frequency data and the breakdown voltage data obtained at each measurement point, and displays a temporal breakdown histogram, which is a kind of dielectric breakdown characteristic, on the display 70 as an edited result.
The time-dependent dielectric breakdown histogram is obtained by plotting the number of voltage applications on the horizontal axis and, for example, the number of defects (the number of data of voltage application times corresponding to the number of voltage applications) on the vertical axis.

FIG. 13 is a graph showing an example of editing a time-dependent dielectric breakdown histogram. In the figure, a plurality of time-dependent dielectric breakdown histograms at different reference values Vref are collectively displayed. The different reference values Vref are 14.5V, 15.0V, 15.5V, 16.0 respectively.
It is. The thickness of the insulating thin film on the sample substrate to be measured is 9 nm. As is apparent from FIG.
Is higher, the dielectric breakdown with time occurs with a lower number of times of voltage application, that is, it is easily broken.

FIG. 14 is a graph obtained by converting the temporal breakdown histogram of FIG. 13 into a cumulative failure rate graph. The cumulative failure rate on the vertical axis is obtained by the following equation. Cumulative failure rate = lnln (1− (1− (N0 / (N + 1)))) (Equation 1) In Equation 1, N0 is the total number of failures, and N is the number of failures at each application. From the figure, a clear difference is recognized with respect to the reference voltage Vref.

The effects of this embodiment are as follows. (7) It is possible to measure dielectric breakdown with time using SPM. (8) In the SPM dielectric breakdown measurement method, since the SPM edits and displays the data of the number of times of voltage application in a histogram or the like,
The operator can measure the sample 3 or the oxide film 3a based on the display.
Can be easily evaluated.

Embodiment 4 Next, a fourth embodiment will be described. First, as described in the second embodiment, after grasping the radius of the trapped charge region 3b from the time zero breakdown histogram, the operator sets the SPM to a value sufficiently longer than the radius.

Then, the computer 71 measures the dielectric breakdown characteristics based on the processing shown in the flowchart shown in FIG. In step 101 of the process,
The computer 71 moves the sample 3 so that the distance d becomes a value set by the operator.

The effects of this embodiment are as follows. That is, (9) it is possible to measure the dielectric breakdown with time without the influence of the SPM on the charge trapping region.

Embodiment 5 FIG. Embodiments 1 and 2
In the above, a relatively small section was attached to the cylindrical piezo element 6 as the sample 3 to be measured, but a semiconductor wafer, for example, can be used as the sample 3 instead of the section. This semiconductor wafer may be being processed during the manufacturing process of the semiconductor device. Therefore, the dielectric breakdown measuring method of the present invention can be applied as an in-line measuring method in a semiconductor device manufacturing process.

For example, FIG. 15 shows an example of a semiconductor wafer 19 being processed during a semiconductor device manufacturing process. Each code in FIG. 15 corresponds to each code in FIG. On a semiconductor wafer 19, semiconductor elements 20 such as transistors and resistors are formed.

Here, the semiconductor element 20 has a sectional structure as shown in FIG. 16, for example. Semiconductor wafer 1
9, a thick isolation oxide film layer 21 and a thin gate oxide film layer 22, which are insulating thin films, are formed. In the next step for processing the semiconductor wafer 19 shown in the figure among the steps in the manufacturing process of the semiconductor device, a polysilicon layer and an aluminum layer which are to be the gate electrodes of the transistors are formed. An inspection step is added before this step. That is, before the formation of the polysilicon layer and the aluminum layer, the thin gate oxide film layer 22 is formed.
The cantilever 1 and the probe 2 are brought close to the surface of the first embodiment, and the method described in the first embodiment, preferably the second embodiment, is applied to measure the breakdown voltage and create a time-zero breakdown histogram. Then, at this breakdown voltage, the operator checks whether there is a breakdown voltage equal to or higher than a predetermined value set in advance. Also, it is checked whether or not the created time-zero breakdown histogram has a failure rate equal to or higher than a predetermined value preset by the operator. As a result of the inspection, if the breakdown voltage is equal to or higher than a predetermined value, the next step, that is, a polysilicon layer, an aluminum layer, etc., which will be a gate electrode of the transistor is formed. On the other hand, as a result of the inspection, when there is no breakdown voltage equal to or higher than the predetermined value, FIG.
In the stage of the semiconductor wafer 19 shown in (1), this wafer 19 is determined to be defective and is excluded from the production line. Note that this inspection method can be incorporated into a semiconductor device manufacturing process (manufacturing line) and performed as an inline inspection, or the inspection can be performed outside the manufacturing process. As described above, in the present embodiment, the measurement target is the semiconductor device being manufactured, such as the semiconductor wafer 19.

As described above, during the process of manufacturing a semiconductor device, a time-zero breakdown histogram is created by measuring a breakdown voltage, and an inspection step for judging pass / fail of a semiconductor wafer is provided based on these. .

The effects of this embodiment are as follows. That is, (10) the inspection process relating to time zero dielectric breakdown is performed during the manufacturing process of the semiconductor device, so that it is possible to determine whether there is a defect in the manufacturing process, or to determine the semiconductor device manufacturing Can be evaluated, so that a highly reliable semiconductor device can be manufactured.

Embodiment 6 FIG. The sixth embodiment is the same as the fifth embodiment, except that the first and second embodiments applied to the fifth embodiment are replaced with a third embodiment, preferably a fourth embodiment.

The effects of this embodiment are as follows. That is, (11) an inspection process relating to the time-dependent dielectric breakdown is performed during the semiconductor device manufacturing process, so that it is possible to determine whether or not there is a defect during the manufacturing process, or to evaluate the semiconductor device manufacturing device from the time-dependent dielectric breakdown. Therefore, a highly reliable semiconductor device can be manufactured.

Modified example. In the first to fourth embodiments, the cylindrical piezo element 6 is used for scanning / moving the sample 3 to be measured. May have the same effect.

In the first to sixth embodiments, the probe 2 provided at the tip of the cantilever 1 is made of a conductive coating film 2.
a. However, the coat film generally has low resistance to overcurrent. Therefore, it is preferable to use a bulk conductive probe, such as a tungsten needle, for higher resistance.

In Examples 3, 4, and 6, the number of dielectric breakdowns with time was recorded. However, the cumulative defective rate may be displayed by converting the amount of applied charge proportional to the number of breakdowns. To play.

Further, in the measurement of the dielectric breakdown with time, a pulse voltage having the same peak voltage may be applied instead of the sawtooth wave voltage, and the same effect is obtained.

During the measurement of the dielectric breakdown characteristics,
It is desirable to keep the feedback loop shown in FIG. 4 closed. By keeping the closed state, the current I can be measured stably.

[0060]

According to the first aspect of the present invention, the time zero dielectric breakdown characteristic can be measured by using the existing SPM without using the MOS structure, and the MOS structure for dielectric breakdown measurement can be measured. Since there is no need for this, there is an effect that the dielectric breakdown characteristics can be easily monitored during the manufacturing.

According to the second aspect of the present invention, a time-dependent dielectric breakdown characteristic can be measured by using an existing SPM without requiring a MOS structure, and a MOS structure for dielectric breakdown measurement is not required. Therefore, there is an effect that the dielectric breakdown characteristics can be easily monitored during the manufacturing.

According to the third aspect of the present invention, it is possible to obtain the dielectric breakdown characteristics at a plurality of measurement points while avoiding the movement of the probe while causing the dielectric breakdown, and to perform the mapping of the dielectric breakdown characteristics. It has the effect of being able to.

According to the fourth aspect of the present invention, after setting the second distance which is not affected by the charge trapping region, the first to third steps are executed, whereby the dielectric breakdown voltage can be accurately evaluated. This has the effect.

[Brief description of the drawings]

FIG. 1 is an operation diagram illustrating a dielectric breakdown measuring method according to a first embodiment of the present invention.

FIG. 2 is a timing chart showing I / V characteristics obtained by the operation of FIG.

FIG. 3 is a flowchart showing a dielectric breakdown measuring method according to the first embodiment of the present invention.

FIG. 4 is a block diagram illustrating a configuration of an SPM.

FIG. 5 is a graph showing one edited example of a time zero breakdown histogram obtained by the breakdown measurement method according to the first embodiment of the present invention.

FIG. 6 is a graph showing another edited example of a time zero breakdown histogram obtained by the breakdown measurement method according to the first embodiment of the present invention.

FIG. 7 is a conceptual diagram showing an effect of carrier injection into an oxide film by a probe.

FIG. 8 is a conceptual diagram showing the effect of carrier injection into an oxide film by a probe.

FIG. 9 is a graph showing an example of a time zero dielectric breakdown histogram used in the dielectric breakdown measuring method according to the second embodiment of the present invention.

FIG. 10 is an operation diagram illustrating a dielectric breakdown measurement method according to a third embodiment of the present invention.

FIG. 11 is a timing chart showing I / V characteristics obtained by the operation of FIG.

FIG. 12 is a flowchart illustrating a dielectric breakdown measuring method according to a third embodiment of the present invention.

FIG. 13 is a graph showing one edited example of a time-dependent dielectric breakdown histogram obtained by the dielectric breakdown measuring method according to the third embodiment of the present invention.

FIG. 14 is a graph obtained by converting the temporal breakdown histogram of FIG. 13 into a cumulative failure rate graph.

FIG. 15 is an operation diagram showing a dielectric breakdown measuring method according to a fifth embodiment of the present invention.

FIG. 16 is a sectional structural view showing an example of a semiconductor element formed on a wafer in the course of manufacture.

FIG. 17 is an operation diagram illustrating a method of measuring I / V characteristics by SPM.

18 is a timing chart showing I / V characteristics obtained by the operation of the SPM shown in FIG.

[Explanation of symbols]

Reference Signs List 1 cantilever, 19 semiconductor wafer, 20 semiconductor element, photodiode detector, filter 9, switch 15, differential amplifier 10, integrating amplifier circuit 11, proportional amplification.

Continued on the front page (72) Inventor Naoshi Nishioka 4-1-1 Mizuhara, Itami-shi, Hyogo Ryoden Semiconductor System Engineering Co., Ltd.

Claims (4)

[Claims] 1. A dielectric breakdown measuring method for measuring a dielectric breakdown voltage of a sample having an insulating thin film exposed on a surface, wherein a predetermined voltage is increased between a probe facing the insulating thin film and the sample. A first step of applying and measuring a current flowing between the insulating thin film and the probe; and a second step of obtaining the predetermined voltage when the current reaches a reference value as a breakdown characteristic. And a mechanism for applying the probe and the predetermined voltage.
A dielectric breakdown measurement method using PM (Scanning Probe Microscopy). 2. A dielectric breakdown measuring method for measuring a dielectric breakdown resistance of a sample having an insulating thin film exposed on a surface, wherein a predetermined voltage is repeatedly applied between a probe facing the insulating thin film and the sample. A first step of measuring a current flowing between the insulating thin film and the probe; and obtaining a number of times the predetermined voltage is repeatedly applied until the current reaches a reference value as insulation breakdown characteristics. And S is a mechanism for applying the probe and the predetermined voltage.
A dielectric breakdown measurement method using PM (Scanning Probe Microscopy). 3. The method according to claim 2, further comprising: a third step in which the probe moves on the insulating thin film, wherein the first and second steps are executed for a measurement point reached in the third step, and The method according to claim 1, wherein the predetermined voltage is set to zero before the step (c) is performed. 4. (a) (a-1) applying a predetermined voltage between the probe and the sample while increasing the voltage to another insulating thin film corresponding to the insulating thin film, A fourth step of measuring a current flowing between an insulating thin film and the probe; and (a-2) a fifth step of obtaining the predetermined voltage when the current reaches a reference value in the fourth step. And (a-3) a sixth step in which the probe moves on the another insulating thin film at a first distance over a first distance to obtain a plurality of the predetermined voltages. A first process for obtaining a first result, which is a statistic of the above, is repeated while updating the first distance to obtain a plurality of the first results, and (b) a determination of dividing the plurality of the first results into two. Setting a second distance corresponding to a reference; and (c) setting the first to the first with the second distance. Performing third step, breakdown measuring method according to claim 3, wherein.