Semiconductor device manufacturing: process – With measuring or testing – Electrical characteristic sensed
Reexamination Certificate
2000-12-29
2001-09-11
Niebling, John F. (Department: 2812)
Semiconductor device manufacturing: process
With measuring or testing
Electrical characteristic sensed
C438S018000, C438S514000, C073S105000
Reexamination Certificate
active
06287880
ABSTRACT:
BACKGROUND OF THE INVENTION
Semiconductors are fabricated by taking a basic semiconductor substrate and doping the surface of that substrate with impurities. In some cases, the materials are implanted directly upon the surface of the substrate. In other cases, the impurities are previously incorporated into the substrate itself.
One technique of implanting a second material into a substrate is that of ion beam implantation. In this technique, an ion beam is directed to the surface of the substrate, and the ions, traveling at high velocity, penetrate to a certain, predetermined depth in the substrate itself before stopping. By implanting these impurity ions, and through subsequent thermal treatment, the conductivity of the semiconducting material is increased. These changes permit the formation of integrated circuits in which electrons can be conducted in closely spaced paths throughout the substrate.
As integrated circuit technology develops, the width and depth of these implanted regions decreases. By reducing the size of the implanted regions, gates and traces on the substrate can be spaced closer and closer together. Just a few years ago, the depth of ion implantation was typically on the order of 0.5 to 1 microns. In recent years, the implanted regions have been reduced in depth to below 0.1 microns.
As these shallow implants are developed, it is necessary to determine the depth profile of the implanted regions. In particular, the depth, width, and concentration of ions implanted need to be carefully monitored. Only by providing experimenters with this research data can they further improve their processes.
A traditional technique for determining the characteristics of an implanted region is Spreading Resistance Profiling, or SRP. An apparatus
1
for performing SRP is shown in FIG.
1
A. In this technique, a small sample
2
having an implanted region
3
is prepared by making an angular cut in a surface
4
of the doped region
3
to create a beveled region or surface
5
, and mounting the prepared substrate in a sample holder (not shown). SRP machine
1
includes two probes
6
,
7
spaced closely together that are moved generally simultaneously across the angular beveled region of prepared sample
2
to obtain data at various locations of the sample. As probes
6
,
7
are moved across the surface to several different locations, they are pressed down firmly against the surface of the substrate in order to make good electrical contact with the substrate. SRP probes are typically on the order of 1-5 microns in radius at the point or points they contact the substrate.
With probes
6
,
7
making electrical contact with the sample, a voltage is applied across these electrodes, and the amount of current flow through the electrodes is measured. Knowing both the current and voltage, the resistivity of the sample as a function of depth can be determined. The relationship of resistivity to ion concentration is well known to those skilled in the art. Therefore from a plot of resistivity data versus depth, the implant characteristic or doping concentration can be measured and results delivered to process engineers. Notably, due to the size of electrodes
6
,
7
, however, they are spaced relatively far apart.
In order to make good contact with the surface of the prepared sample, the pressure applied to the surface by the SRP probe is substantial. The compression or indent of atoms that results from pressure applied to the SRP probe causes the electrical characteristics of the sample in that region to change. In addition, the two probes are spring or pneumatically actuated and move, ideally, simultaneously toward and away from the surface of the prepared sample. Due to mechanical limitations of the SRP machine, including the lack of a controlled force on each probe, indent depth is often inconsistent. Since the bevel is created to extend the depth dimension laterally, preferably at a shallow angle to facilitate making measurements at highly precise depths, any increase or other inconsistency in indent depth between the probes (also arranged laterally) defeats the precision of the measurement created at least in part by the shallow bevel.
In order for the resistance data to be usable the probes must have a constant spacing throughout the measurement. Alignment between probes is critical and a line connecting the probe tips must be perpendicular to the bevel direction or the sampled material from the two probes will be at different depths.
Another drawback is that the probes may wear differently. If one probe tip is worn down and the other probe tip is not, the electrical resistance between one probe tip and the surface of the substrate will be much greater than the resistance between the other probe tip and the substrate. In addition, the difference in contact area between the two probes will cause different amounts of deformation of the substrate where each probe contacts the surface. The differing deformation will change the electrical characteristics of the doped substrate and cause the SRP to produce erroneous readings. All of these characteristics of SRP cause bad test results.
In another technique, shown in
FIG. 1B
, a scanning proximity microscope
8
is used to control the force on the probe or probes
9
so that the above limitations are reduced. The proximity microscope provides force control for the contact on the sample
10
, and thereby eliminates some of the errors associated with unequal contacts and varying contact depths. The forces may be small, thus allowing use of a smaller tip
11
, typically 5-500 nanometers in radius. However, with current proximity microscope technology it remains difficult to use two probe contacts. The current state of the art uses a “back” contact somewhere on the sample which may induce errors due to series resistance and current flow through nearby heavily doped or metallic regions.
What is needed, therefore, is an improved method and apparatus for determining the electrical characteristics of a doped semiconductor. It is an object of this invention to provide such a method and apparatus.
SUMMARY OF THE INVENTION
The first embodiment of the invention is directed to a method of determining the electrical characteristics of a doped semiconductor substrate, machining the top surface away to provide a beveled surface that is disposed at an angle with respect to the top surface, wherein the beveled surface exposes a plurality of dopant densities at a plurality of depths within the substrate, fixing an electrode at a first electrode location to the substrate, perpendicular to the bevel plane, moving an electrically conductive probe to a first probe location on the beveled surface, applying a first electrical potential across the probe and electrode wherein the first electrical potential is sufficient to generate a first current through a first current carrying path defined between the first electrode location and the first probe location, wherein the first electrical potential and the first current collectively define a first electrical characteristic of the substrate along the first current carrying path and recording a first value indicative of the first electrical characteristic.
According to another aspect of the invention, the method includes moving the probe to a second probe location on the beveled surface, applying a second electrical potential across the probe and electrode wherein the second electrical potential is sufficient to generate a second current through a second current carrying path defined between the first electrode location and the second probe location, wherein the second electrical potential and the second current collectively define a second electrical characteristic of the substrate along the first current carrying path, and recording a second value indicative of the second electrical characteristic.
The first and second electrical potentials may be the same, and the first and second currents different. In this case, the first and second values saved will be indicative of the first and second currents, respectivel
de Wolf Peter
Erickson Andrew Norman
Niebling John F.
Nilles & Nilles S.C.
Stevenson Andre′ C
Veeco Instruments Inc.
LandOfFree
Method and apparatus for high resolution profiling 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 Method and apparatus for high resolution profiling in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for high resolution profiling in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2475421