JPH06342027A - Method and apparatus for evaluating wire - Google Patents

Abstract

PURPOSE:To highly accurately measure EM life/wire temperature characteristics by measuring a resistance value and disconnection time of a wire for each temperature, calculating a temperature of the wire from the resistance value and a temperature coefficient of resistance which has been measured and determining a relation between the disconnection time and the wire temperature. CONSTITUTION:A temperature of a temperature control base 1 is maintained at a set temperature by a controller 5, constant current is supplied 7 to a wire 3, and cumulative time is counted by a timer simultaneously. Voltage V across the wire 3 is detected by a probe 4, a resistance value R of the wire 3 is calculated from the voltage V, disconnection is determined when the voltage V exceeds a threshold Vth, and the resistance value R and disconnection time H are stored. Then a temperature T of the wire 3 is calculated from TCR (temperature coefficient of resistance) which has been measured and the measured resistance value R. Then an inverse of the wire temperature T and a logarithm of the disconnection time H are plotted to obtain a characteristic curve. An activation energy Ea and a proportional constant A are determined from the characteristic curve to calculate EM(electro-migration) life H (average failure time).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a wiring evaluation method and an evaluation apparatus, and more particularly, to an apparatus for estimating the life of an aluminum wiring laid on a semiconductor wafer.

[0002]

2. Description of the Related Art The electromigration (hereinafter referred to as EM) life of Al wiring has been conventionally estimated by assembling it in a package and then applying a constant current at a constant temperature for a long time in a thermostatic chamber. ~ 1000 hr was required. Proceedings of the 38th JSAP Joint Lecture Meeting No. 2 (1991), p. 639,
A method for evaluating the activation energy of aluminum wiring in a short time has been proposed.

This evaluation method applies a constant current density Jc to a large number of samples each of which has a wiring pattern under test having different heat dissipation characteristics, and conducts heat conduction simulation (three-dimensional heat dissipation characteristics analysis) and temperature coefficient of resistance for each sample. The wiring temperature is calculated based on the above, and the disconnection time of each sample is measured to obtain the relationship between the wiring temperature and the disconnection time for each sample. The characteristic line indicating the relationship is obtained, and the activation energy Ea is estimated from the slope rate.

Therefore, the EM life (mean failure time MT
Since F) is a function value of the current density J, the wiring temperature T, and the activation energy Ea, if the activation energy Ea calculated by the above-described evaluation method is used, a desired wiring temperature (for example, 150 ° C.) is obtained. ), The EM life at any current density J can be estimated if the relationship between the current density J and the wiring temperature T is known by the heat conduction simulation.

[0005]

However, in the evaluation method proposed above, it is necessary to change the heat dissipation characteristics for each sample.
For this reason, a radiation fin having a different shape is provided on a part of each aluminum wiring to be tested, whereby the radiation characteristic is changed and the wiring temperature T is changed. However, according to this wiring temperature changing method, it is necessary to perform heat conduction simulation (three-dimensional heat dissipation characteristic analysis) for each sample to calculate each wiring temperature, and as a result, each sample is prepared and their heat conduction is performed. The simulation had to be performed, which was a great burden on the work.

In addition, such a change of the shape of the heat radiation fin has a limit in the change of the obtained wiring temperature, and as a result, the accuracy of the slope rate of the obtained characteristic line, that is, the activation energy Ea is lowered. . The present invention has been made in view of the above problems, and an object thereof is to provide a wiring evaluation method and a wiring evaluation apparatus that are light in work load and can obtain highly accurate EM life / wiring temperature characteristics.

[0007]

According to the wiring evaluation method of the present invention, a constant current is passed through while maintaining an object to be tested in which wiring is laid on an insulator at a plurality of predetermined temperature states. Each resistance value and each disconnection time of the wiring for each temperature state is measured, the temperature of the wiring is calculated based on the resistance value and the resistance temperature coefficient of the wiring measured in advance, and each of the disconnection time and each The feature is that the relationship with the temperature of the wiring is determined.

As shown in the conceptual diagram of FIG. 7, the wiring evaluation apparatus of the present invention comprises temperature maintaining means for maintaining a device under test in which wiring is laid on an insulator at a plurality of predetermined temperature states,
Energizing means for applying a constant current to the wiring, measuring means for measuring each resistance value and each disconnection time of the wiring for each temperature state, each resistance value and resistance temperature coefficient of the wiring measured in advance Based on the relationship between each wiring temperature and each disconnection time, a wiring temperature calculating means for calculating the temperature of the wiring based on the above, and an evaluation for estimating a disconnection time or activation energy of the wiring at an arbitrary wiring temperature. And means.

As the wiring, various conductor materials used for electric wiring other than aluminum and aluminum alloy are adopted.

[0010]

According to the present invention, the resistance value and the disconnection time H are measured by changing the ambient temperature of the sample while energizing the metal wiring having a predetermined shape with a constant current density Jc. Next, each wiring temperature T is calculated from the previously measured resistance temperature coefficient of the metal wiring and the resistance value. Then, a characteristic line indicating the relationship between the calculated disconnection time H and the wiring temperature T is calculated.

By doing so, the activation energy Ea and the EM life can be easily obtained from the obtained characteristic line.

[0012]

FIG. 1 shows an example of an evaluation device to which the present invention is applied. This evaluation device estimates the EM life and activation energy of aluminum wiring. A wafer 2 is placed on a temperature control base (heating means) 1, and an insulating layer (not shown) is placed on the wafer 2. Probe needles (measuring means) 4 and 4 are in contact with both ends of an aluminum wiring 3 having a predetermined pattern laid via a film.

A heater (not shown) and a temperature sensor (not shown) are built in the temperature control base 1, and the heater is energized from a DC power supply or a commercial AC power supply (not shown) through the controller 5. ing. The temperature sensor is arranged near the boundary surface with the wafer 2. The controller 5 is PCM controlled by a microcomputer (wiring temperature calculation means, evaluation means) 6, and the temperature sensor outputs a detected temperature to the microcomputer 6.

On the other hand, the probe needles 4, 4 are supplied with a constant current from a constant current source (energizing means) 7, and the voltage across the probe needles 4, 4 is an amplifier (measuring means) 8 and an A / D converter ( It is transmitted to the microcomputer 6 via the measurement means) 9. FIG. 2 is an enlarged view of details of main parts. The aluminum wiring 3 is heated by passing a constant current on the wafer 2, and the heat generated in the aluminum wiring 3 is radiated to the outside, particularly to the temperature control base 1.
Therefore, the temperature of the aluminum wiring 3 is controlled by adjusting the temperature of the temperature control base 1.

The evaluation method of this embodiment will be described below with reference to the flow charts of FIGS. First, the temperature control base 1 detected by the above temperature is set to a set temperature Ta = 25 ° C. (100). This temperature control is simply executed by feedback-controlling the on-duty ratio of the controller 5 according to the difference between the detected temperature and the set temperature Ta from the temperature sensor, but such control itself is well known.
Further description will be omitted.

Next, after the detected temperature reaches the set temperature Ta, the constant current source 7 is instructed to apply a constant current Ic to the aluminum wiring 3.
(Current density J = 7.7 × 10 ⁶ A / cm ² ) is supplied (102), and at the same time, the built-in timer is started to count the cumulative energization time (104). Next, after a sufficient time has passed for the temperature distribution state due to heat generation of the aluminum wiring 3 due to the energization to reach equilibrium, the voltage V across the aluminum wiring 3 is detected, and the resistance of the aluminum wiring 3 is determined from this voltage V. Value R
Is calculated (V / Ic) and stored (106).

Thereafter, the voltage V at both ends of the aluminum wiring 3 is detected, and when the voltage V exceeds the predetermined voltage value Vth, it is judged that the aluminum wiring 3 is broken (108), and the count value of the built-in timer ( That is, the disconnection time H) is stored, the timer is reset to 0, the counting is stopped (110), and the routine is ended. Thereby, a pair of the resistance value R and the disconnection time H is obtained.

Below, set temperatures Ta = 25 ° C., 50 ° C., 7
A large number of samples are prepared for each 5 ° C., and the resistance value R and disconnection time H are set for each sample by the same process as above.
To measure. Next, the wiring temperature T is calculated for each of these resistance values R (201). In order to calculate the wiring temperature T from each resistance value R, the TCR (resistance temperature coefficient) of the aluminum wiring 3 is calculated in advance.
Is measured in advance, and the wiring temperature T may be calculated from this TCR and each resistance value. The relational expression of the resistance value R, the wiring temperature T, and the temperature coefficient of resistance TCR is shown below.

[0019] Next, the reciprocal of the obtained wiring temperature T is obtained (202), the logarithmic value of the disconnection time H is obtained (203), the obtained value is Arrhenius plotted (204), and the characteristic line L1 is obtained based on it.
(See FIG. 3) is calculated (205).

Next, the activation energy Ea and the proportional constant A are determined from this characteristic line L1 and the following theoretical formula (20)
6), the routine ends. The activation energy Ea of the aluminum wiring 3 obtained in this experiment was 0.72 ev, which was in good agreement with the value 0.68 ev obtained by the conventional method. However, EM life H represents MTF (mean failure time),
w is wiring width, t is wiring thickness, J is current density, A is proportional constant, K is Boltzmann constant, Ea is activation energy, T
Is the wiring temperature (absolute temperature).

The characteristic line L1 thus obtained is shown in FIG.
On the other hand, the current density J = 1 × 10 ⁶A / cm²Oke
The experimental value P in the conventional method is also shown. Next, the characteristic line L1
(The current density J is 7.7 × 10⁶A / cm²) From the current tight
Degree J is 1 × 10⁶A / cm²The characteristic line L2
Ask. Current density J = 1 × 10⁶
A / cm²The characteristic line L2 at time is an estimated point obtained in the conventional experiment.
It agrees well with P.

As described above, according to this embodiment,
The activation energy Ea and the EM life H can be calculated with high accuracy despite the simple configuration and operation.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of an evaluation device of the present invention.

FIG. 2 is an enlarged sectional view of a main part of FIG.

FIG. 3 is a graph obtained by Arrhenius plot of measurement data.

FIG. 4 is a characteristic diagram showing the relationship between current density, base temperature, and wiring temperature.

FIG. 5 is a flowchart showing an embodiment of the evaluation method of the present invention.

FIG. 6 is a flowchart showing an embodiment of the evaluation method of the present invention.

FIG. 7 is a conceptual diagram of the present invention.

[Description of Reference Signs] FIG. 1 shows an example of an evaluation apparatus to which the present invention is applied. 1 is a temperature control base (temperature maintaining means), 3 is aluminum wiring (wiring), 4 is a probe needle (measuring means), 6 is a microcomputer (wiring temperature calculating means, evaluating means), 7 is a constant current source (energizing means) , 8 is an amplifier (measuring means), 9 is an A / D converter (measuring means).

Claims (2)

[Claims] 1. A resistance value and a disconnection of the wiring for each temperature state by applying a constant current while maintaining the DUT whose wiring is laid on an insulator in a plurality of predetermined temperature states. The time is measured, the temperature of the wiring is calculated based on the resistance value and the temperature coefficient of resistance of the wiring measured in advance, and the relationship between the disconnection time and the temperature of the wiring is determined. How to evaluate the wiring. 2. A temperature maintaining means for maintaining a device under test in which wiring is laid on an insulator in a plurality of predetermined temperature states, an energizing means for supplying a constant current to the wiring, and each of the temperature states. Measuring means for measuring each resistance value and each disconnection time of the wiring, wiring temperature calculating means for calculating the temperature of the wiring based on the resistance value and the resistance temperature coefficient of the wiring measured in advance, and Based on the relationship between each wiring temperature and each disconnection time,
An evaluation device for estimating a disconnection time or activation energy of the wiring at an arbitrary wiring temperature, and a wiring evaluation apparatus comprising: