Electricity: measuring and testing – Impedance – admittance or other quantities representative of... – Lumped type parameters
Reexamination Certificate
2000-08-24
2002-10-08
Le, N. (Department: 2858)
Electricity: measuring and testing
Impedance, admittance or other quantities representative of...
Lumped type parameters
C324S719000, C324S1540PB, C438S017000
Reexamination Certificate
active
06462565
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention is directed to a wire technique of a semiconductor device, and more particularly to a measuring-pattern for a width of the wire and a method for measuring the width of the wire using the measuring-pattern.
2. Description of the Related Art
A process for a semiconductor should be performed with high accuracy, so after each step of the process, the manufactured should be subject to a test or an evaluation. Since a width of a wire, among many elements for a semiconductor device is one of significant factors affecting performance and stability of the semiconductor device, we must test as to whether or not the width of a manufactured wire is the same as that of a designed wire.
There are many types of measuring method for the width of wire. Among them, one approach is to directly measure the width of wire using SEM(Scanning Electron Microscope) and another approach is to use a resistance of the wire subjected to a measurement and a resistivity of a material constituting the wire wherein the resistance is determined through a four-probe test. The method using the SEM is to scan the pattern to be measured through accelerated electrons, to magnify the pattern using information contained in secondary electrons generated by the accelerated electrons and then to measure the width of the wire. However, when a sectional view of the pattern is not at a right angle, the width of the wire can not be accurately measured by the SEM. The volume of SEM is great, so a space for the SEM in a production line is large. Also, a cost for the SEM is much.
A technique for measuring the width W
1
of wire
10
whose thickness is ti will be described with reference to FIG.
1
. On the wire
10
are arranged four probes p
1
, p
2
, p
3
and p
4
each separated by constant interval S
1
. The outer two probes p
1
and p
4
are connected to a power source
11
and the inner two probes p
2
and p
3
are connected to a voltmeter
12
. When the current I from the power source
11
flows through the wire
10
, a voltage between the probe p
2
and the probe p
3
is measured at the voltmeter
12
. The resistance of the wire is calculated from the measured voltage using Ohm's law and then the width of the wire is computed by entering the calculated resistance to the following equation.
W
1
=(&rgr;*S
1
)/(R*t
1
), wherein &rgr; is resistivity of the wire
10
. That is, the resistivity of the wire
10
should be previously perceived in order to determine the width of the wire W
1
. For taking the gauge of resistivity, first a portion of the wire is defined in rectangular form of l×l
2
and then four probes are disposed at four corners of the defined rectangular, resepctively. One pair of probes with l
2
interval are connected to a power source and the other pair of probes with l
2
interval are connected to a voltmeter. The current I flows from the power source to the defined wire, and then a voltage V between the probes is measured at the voltmeter. Thereafter, the resistivity is calculated using Van der Pauw method. In detail, the resistivity is determined by the following equation.
&rgr;=(
V/I
)
f
(
l
1
/
l
2
),
where f(l
1
/l
2
) is a correction factor, approximate value determined as a function of average probe separation and average specimen diameter.
Since the approximate value is used in calculating the resistivity, a deviation of the resistivity is possibly apt to be large. In particular, where l
1
or l
2
is not enough larger than the thickness of the wire measured and the size of the probe, the deviation of the resistivity becomes enormously large. Accordingly, such resistivity inevitably results in a debasement in a measuring reliability of the width of the wire.
SUMMARY OF THE INVENTION
Accordingly, an object of the present invention is to accurately measure a width of a wire without using a resistivity of the wire.
In a view of the present invention, a measuring pattern for a width of a wire comprises a first pattern with a first width, a second pattern connected to the first pattern and having a second width wider than the first width and a third pattern connected to the second pattern and having a third width narrower than the first width. The first pattern, the second pattern and the third pattern are made of the same material. The second width is substantially more than ten times the first width and the third width is narrow such that current therethrough is uniform. Also, a joint between the first pattern and the second pattern and a joint between the second pattern and the third pattern have a selected slant angle so as to minimize a current turbulence phenomena. In another view of the present invention, a measuring pattern for the width of the wire is prepared. The measuring pattern includes: a first pattern with a first width; a second pattern connected to the first pattern and having a second width wider than the first width; and a third pattern connected to the second pattern and having a third width narrower than the first width. The first pattern, the second pattern and the third pattern are made of the same material. The first pattern through a power source is connected to the third pattern. A first pair of probes are disposed on the first pattern and then are connected to a first voltmeter. A distance between the first pair of probes is a first distance wider than the first width. A second pair of probes are disposed on the second pattern and then are connected to a second voltmeter. A distance between the second pair of probes is a second distance wider than the first width. Thereafter the first width using gauged voltage at the first voltmeter and the second voltmeter is determined. The first width is determined by the following equation.
W
=(
W*s/S
)*(
V/v
),
where w is the first width, W is the second width, s is the first distance, S is the second distance, v is voltage at the first voltmeter and V is voltage at the second voltmeter. The second width is substantially more than ten times the first width and the third width is narrow such that current therethrough is uniform. Also, a joint between the first pattern and the second pattern and a joint between the second pattern and the third pattern have a selected slant angle so as to minimize a current turbulence phenomena. Since evaluated by a distance between a pair of probes which are disposed on the first pattern and the second pattern connected to the first pattern, voltage between the probes and the width of the second pattern, the width of the first pattern is not
20
reflected by whether the sectional view of the tested object is rectangular or not.
Also, unlike the four-probe test method, since the present invention does not compute resistivity of material for the wire, the deviation of the width of the wire owing to the deviation of resistivity is not basically occured.
REFERENCES:
patent: 2659861 (1953-11-01), Branson
patent: 3974443 (1976-08-01), Thomas
patent: 4024561 (1977-05-01), Ghatalia
patent: 4751458 (1988-06-01), Elward
patent: 5049811 (1991-09-01), Dreyer et al.
patent: 0325269 (1989-07-01), None
patent: 1479869 (1977-07-01), None
Kim Jong Il
Kim Kil Ho
Shin Kang Sup
Hynix / Semiconductor Inc.
Le N.
Nguyen Vincent Q.
Selitto Behr & Kim
LandOfFree
Measuring pattern for measuring width of wire in... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Measuring pattern for measuring width of wire in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Measuring pattern for measuring width of wire in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2966085