U.S. patent application number 09/756515 was filed with the patent office on 2001-11-22 for determination of thickness and impurity profiles in thin membranes utilizing spectorscopic data obtained from ellipsometric investigation of both front and back surfaces.
Invention is credited to Herzinger, Craig M., Tiwald, Thomas E..
Application Number | 20010042832 09/756515 |
Document ID | / |
Family ID | 26879705 |
Filed Date | 2001-11-22 |
United States Patent
Application |
20010042832 |
Kind Code |
A1 |
Herzinger, Craig M. ; et
al. |
November 22, 2001 |
Determination of thickness and impurity profiles in thin membranes
utilizing spectorscopic data obtained from ellipsometric
investigation of both front and back surfaces
Abstract
Disclosed is spectroscopic ellipsometer system mediated
methodology for quantifying thickness and impurity profile defining
parameters in mathematical models of impurity profile containing
thin membranes having two substantially parallel surfaces which are
separated by a thickness, wherein the spectroscopic ellipsometer
system operates in near-IR and IR wavelength ranges.
Inventors: |
Herzinger, Craig M.;
(Lincoln, NE) ; Tiwald, Thomas E.; (Lincoln,
NE) |
Correspondence
Address: |
JAMES D. WELCH
10328 PINEHURST AVE.
OMAHA
NE
68124
US
|
Family ID: |
26879705 |
Appl. No.: |
09/756515 |
Filed: |
January 9, 2001 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60183977 |
Feb 22, 2000 |
|
|
|
Current U.S.
Class: |
250/341.4 ;
356/364 |
Current CPC
Class: |
G01N 21/211
20130101 |
Class at
Publication: |
250/341.4 ;
356/364 |
International
Class: |
G01J 003/447; G01N
021/21 |
Claims
We claim:
1. A method of quantifying thickness and impurity profile defining
parameters in impurity profile containing thin membranes,
comprising providing, and obtaining ellipsometric data from both
first and second sides of an impurity profile containing thin
membrane, and providing a mathematical model of said impurity
profile defining parameters comprising membrane thickness and
impurity profile defining parameters, then regressing said
mathematical model onto data obtained from each side of said
impurity profile containing thin membrane by a selection from the
group consisting of: utilizing the data sets obtained from front
and back of the thin membrane simultaneously; utilizing the data
sets obtained from front and back of the thin membrane
independently; and utilizing the data sets obtained from front and
back of the thin membrane both independently and simultaneously. to
evaluate said membrane thickness and impurity profile defining
parameters.
2. A method of quantifying thickness and impurity profile defining
parameters in impurity profile containing thin membranes comprised
of two substantially parallel surfaces that are separated by a
thickness, said method comprising, in any functional order, the
steps of: a. providing an impurity profile containing thin membrane
comprised of two substantially parallel surfaces that are separated
by a thickness, and providing a spectroscopic ellipsometer system
capable of producing spectroscopic data sets at at least one angle
of incidence of a beam of electromagnetic radiation to a surface of
said impurity profile containing thin membrane when it is mounted
in said spectroscopic ellipsometer system; b. determining a range
of wavelengths over which the impurity profile containing thin
membrane is essentially transparent and the effect of the presence
of said impurity profile has essentially negligible effect; c.
determining a range of wavelengths over which the impurity profile
containing thin membrane is essentially transparent, but over which
the effect of the presence of said impurity profile has a
non-negligible effect; d. utilizing substantially wavelengths in
the range determined in step b., by an approach selected from the
group consisting of: reflection ellipsometry; and transmission
ellipsometry; obtaining a spectroscopic data set; e. utilizing
substantially wavelengths in the range determined in step c., by
reflection ellipsometry as applied to one surface of said impurity
profile containing thin membrane, obtaining a spectroscopic data
set; f. utilizing substantially wavelengths in the range determined
in step c., by reflection ellipsometry as applied to a surface of
said impurity profile containing thin membrane offset from that
utilized in step e. by said thickness, obtaining a spectroscopic
data set; g. providing a mathematical model for said impurity
profile containing thin membrane including a parameter that
quantifies thickness; h. providing a mathematical model for said
impurity profile containing thin membrane including parameters that
quantify impurity profile defining parameters; i. using the
spectroscopic data set obtained in step d., regressing the
mathematical model provided in step g. thereonto to evaluate the
parameter that quantifies thickness; j. using the thickness arrived
at in step i. and the spectroscopic data sets obtained in at least
one of the steps e. and f., simultaneously regressing the
mathematical model provided in step h. thereonto to evaluate the
parameters that quantify the impurity profile.
3. A method of quantifying thickness and impurity profile defining
parameters in impurity profile containing thin membranes comprised
of two substantially parallel surfaces that are separated by a
thickness, said method comprising, in any functional order, the
steps of: a. providing an impurity profile containing thin membrane
comprised of two substantially parallel surfaces that are separated
by a thickness, and providing a spectroscopic ellipsometer system
capable of producing spectroscopic data sets at at least one angle
of incidence of a beam of electromagnetic radiation to a surface of
said impurity profile containing thin membrane when it is mounted
in said spectroscopic ellipsometer system; b. determining a range
of wavelengths over which the impurity profile containing thin
membrane is essentially transparent and the effect of the presence
of said impurity profile has essentially negligible effect; c.
determining a range of wavelengths over which the impurity profile
containing thin membrane is essentially transparent, but over which
the effect of the presence of said impurity profile has a
non-negligible effect; d. utilizing substantially wavelengths in
the range determined in step b., by an approach selected from the
group consisting of: reflection ellipsometry; and transmission
ellipsometry; obtaining a spectroscopic data set; e. utilizing
substantially wavelengths in the range determined in step c., by
reflection ellipsometry as applied to one surface of said impurity
profile containing thin membrane, obtaining a spectroscopic data
set; f. utilizing substantially wavelengths in the range determined
in step c., by reflection ellipsometry as applied to a surface of
said impurity profile containing thin membrane offset from that
utilized in step e. by said thickness, obtaining a spectroscopic
data set; g. providing a mathematical model for said impurity
profile containing thin membrane including parameters that quantify
thickness and impurity profile defining parameters; h. using all
obtained spectroscopic data sets, simultaneously regressing the
mathematical model thereonto to evaluate the parameters that
quantify thickness and the impurity profile defining parameters.
Description
[0001] This Application is a Continuation-In-Part of Provisional
Application Ser. No. 60/183,977 filed Feb. 22, 2000.
TECHNICAL FIELD
[0002] The present invention relates to non-destructive
characterization of sample systems, and more particularly to
spectroscopic ellipsometer system(s) mediated methodology for
quantifying thickness and impurity profile defining parameters in
mathematical models of impurity profile containing thin membranes
comprised of two substantially parallel surfaces which are
separated by a thickness, wherein said spectroscopic ellipsometer
system(s) operates in near-IR and IR wavelength ranges.
BACKGROUND
[0003] In view of developing open stencil lithography mask
technology which utilizes open stencil lithography masks formed
from thin silicon membranes, (which are typically formed by pn
junction stop-etch techniques), a need exists for a non-destructive
approach to characterizing thin membrane thickness and impurity
profiles in impurity profile containing thin membranes comprised of
two substantially parallel surfaces that are separated by a
thickness of about 100 microns or less.
[0004] A Search of Patents has revealed U.S. Pat. No. 4,472,633 to
Motooka which describes use of linearly polarized infrared light to
investigate semiconductor wafers. Plots of Ellipsometric PSI vs.
Ellipsometric DELTA, as a function of Angle of Incidence and/or
Wavelength, for various carrier density profiles and depths are
determined. Ellipsometric data obtained from a sample wafer is then
utilized to plot Ellipsometric PSI vs. Ellipsometric DELTA, as a
function of Angle of Incidence and/or Wavelength, and the results
compared to the known plots. Close correlation between sample wafer
and a known Ellipsometric PSI vs. Ellipsometric DELTA, as a
function of Angle of Incidence and/or Wavelength, is indicative of
the sample having a doping profile and depth similar to that of the
wafer from which the known Ellipsometric PSI vs. Ellipsometric
DELTA data was obtained. Data, is described as obtained utilizing
monochromatic light, even though different wavelengths are used in
succession where wavelength is the independent variable.
[0005] Another U.S. Pat. No. 4,807,994 to Felch et al., describes a
non-ellipsometric method of mapping ion implant dose uniformity.
Monochromatic Electromagnetic radiation with a bandwidth of not
more than 1 nm, (chosen for sensitivity to sample parameters being
measured), which has interacted with a sample in Reflectance or
Transmission, is monitored by a Spectrophotometer and the results
compared to previously obtained similar data regarding film
thickness and ion implant doses, and similarities determined.
[0006] U.S. Pat. No. 5,900,633 to Solomon et al., describes a
non-ellipsometric approach to analyzing patterned samples which
involves irradiating a spot which includes first and second pattern
regions, measuring eminating radiation, providing known reference
spectrum/spectra and comparing measured spectral data thereto to
evaluate parameters of layers in said two pattern regions.
[0007] U.S. Pat. No. 5,486,701 to Norton et al., describes a
non-ellipsometric approach simultaneously utilizing wavelengths in
both UV and Visible wavelength ranges to enable calculating a ratio
thereof, which in turn is utilized to determine thin film
thicknesses.
[0008] U.S. Pat. No. 6,049,220 to Borden et al., describes
apparatus and method for evaluating semiconductor material. In a
major implementation thereof, two beams are caused to illuminate a
sample, one having energy above the bandgap and the other having
energy near or below the bandgap. The second beam, after
interaction with the sample, is monitored and change therein caused
by said interaction is indicative of carrier concentration. It is
noted that reflectance of an electromagnetic beam from a sample is
a function of carrier concentration.
[0009] Known relevant art includes Articles, P-N Junction-Based
Wafer Flow Process For Stencil Mask Fabrication", Rangelow et al.,
J. Vac. Sci. Technology B, November/December P. 3592 (1998); and
"Application of IR Variable Angle Spectroscopic Ellipsometry To The
Determination Of Free Carrier Concentration Depth Profiles", Tiwald
et al., Thin Film Solids 313-314, P661, (1998).
[0010] In view of known prior art, there remains need for accuracy
improving methodology for measuring impurity profiles in
substrates, which methodology utilizes electromagnetic radiation
with wavelengths in ranges for which the substrate is opaque and
transparent, and which method involves utilizing data obtained both
when electromagnetic radiation is caused to impinge on one surface,
and then the other surface of said substrate.
DISCLOSURE OF THE INVENTION
[0011] In a basic sense, the present invention comprises a method
of quantifying thickness and impurity profile defining parameters
in impurity profile containing thin membranes, comprising providing
an impurity profile containing thin membrane, and obtaining
ellipsometric data from both first (front) and second (back) sides
thereof, in combination with providing a mathematical model of said
impurity profile defining parameters which comprises membrane
thickness and impurity profile defining parameters, then regressing
said mathematical model onto data obtained from both sides of said
impurity profile containing thin membrane to evaluate said membrane
thickness and impurity profile defining parameters. Note that this
can include utilizing data in a procedure selected from the group
consisting of:
[0012] utilizing the data sets obtained from front and back of the
thin membrane simultaneously;
[0013] utilizing the data sets obtained from front and back of the
thin membrane independently; and
[0014] utilizing the data sets obtained from front and back of the
thin membrane both independently and simultaneously.
[0015] The present invention can more accurately be described as a
method of quantifying thickness and impurity profile defining
parameters in impurity profile containing thin membranes comprised
of two substantially parallel surfaces that are separated by a
thickness, wherein said method comprises, in any functional order,
the steps of:
[0016] a. providing an impurity profile containing thin membrane
comprised of two substantially parallel surfaces that are separated
by a thickness, and providing a spectroscopic ellipsometer system
capable of producing spectroscopic data sets at at least one angle
of incidence of a beam of electromagnetic radiation to a surface of
said impurity profile containing thin membrane when it is mounted
in said spectroscopic ellipsometer system;
[0017] b. determining a range of wavelengths over which the
impurity profile containing thin membrane is essentially
transparent and the effect of the presence of said impurity profile
has essentially negligible effect;
[0018] c. determining a range of wavelengths over which the
impurity profile containing thin membrane is essentially
transparent, but over which the effect of the presence of said
impurity profile has a non-negligible effect;
[0019] d. utilizing substantially wavelengths in the range
determined in step b., by an approach selected from the group
consisting of:
[0020] reflection ellipsometry; and
[0021] transmission ellipsometry;
[0022] obtaining a spectroscopic data set;
[0023] e. utilizing substantially wavelengths in the range
determined in step c., by reflection ellipsometry as applied to one
surface of said impurity profile containing thin membrane,
obtaining a spectroscopic data set;
[0024] f. utilizing substantially wavelengths in the range
determined in step c., by reflection ellipsometry as applied to a
surface of said impurity profile containing thin membrane offset
from that utilized in step e. by said thickness, obtaining a
spectroscopic data set;
[0025] g. providing a mathematical model for said impurity profile
containing thin membrane including a parameter that quantifies
thickness;
[0026] h. providing a mathematical model for said impurity profile
containing thin membrane including parameters that quantify
impurity profile defining parameters;
[0027] i. using the spectroscopic data set obtained in step d.,
regressing the mathematical model provided in step g. thereonto to
evaluate the parameter that quantifies thickness;
[0028] j. using the thickness arrived at in step i. and the
spectroscopic data sets obtained in at least one of the steps e.
and f., simultaneously regressing the mathematical model provided
in step h. thereonto to evaluate the parameters that quantify the
impurity profile.
[0029] An alternative embodiment of the present invention method of
quantifying thickness and impurity profile defining parameters in
impurity profile containing thin membranes which are comprised of
two substantially parallel surfaces that are separated by a
thickness, can also be recited as comprising, in any functional
order, the steps of:
[0030] a. providing an impurity profile containing thin membrane
comprised of two substantially parallel surfaces that are separated
by a thickness, and providing a spectroscopic ellipsometer system
capable of producing spectroscopic data sets at at least one angle
of incidence of a beam of electromagnetic radiation to a surface of
said impurity profile containing thin membrane when it is mounted
in said spectroscopic ellipsometer system;
[0031] b. determining a range of wavelengths over which the
impurity profile containing thin membrane is essentially
transparent and the effect of the presence of said impurity profile
has essentially negligible effect;
[0032] c. determining a range of wavelengths over which the
impurity profile containing thin membrane is essentially
transparent, but over which the effect of the presence of said
impurity profile has a non-negligible effect;
[0033] d. utilizing substantially wavelengths in the range
determined in step b., by an approach selected from the group
consisting of:
[0034] reflection ellipsometry; and
[0035] transmission ellipsometry;
[0036] obtaining a spectroscopic data set;
[0037] e. utilizing substantially wavelengths in the range
determined in step c., by reflection ellipsometry as applied to one
surface of said impurity profile containing thin membrane,
obtaining a spectroscopic data set;
[0038] f. utilizing substantially wavelengths in the range
determined in step c., by reflection ellipsometry as applied to a
surface of said impurity profile containing thin membrane offset
from that utilized in step e. by said thickness, obtaining a
spectroscopic data set;
[0039] g. providing a mathematical model for said impurity profile
containing thin membrane including parameters that quantify
thickness and impurity profile defining parameters;
[0040] h. using all obtained spectroscopic data sets,
simultaneously regressing the mathematical model thereonto to
evaluate the parameters that quantify thickness and the impurity
profile defining parameters.
[0041] The present invention will be better understood by reference
to the Detailed description Section of this Specification, in
conjunction with the Drawings.
SUMMARY OF THE INVENTION
[0042] It is therefore a purpose and/or objective of the present
invention to teach a method of evaluating thickness and impurity
profile describing parameters in an impurity profile containing
thin film, wherein ellipsometric data is obtained from both sides
of the impurity profile containing thin film, and a mathematical
model which contains parameters which describe the thickness and
impurity profile is then regressed onto data obtained from both
said sides of the impurity profile containing thin film, separately
and/or simultaneously.
BRIEF DESCRIPTION OF THE DRAWINGS
[0043] FIG. 1 shows a demonstrative Spectroscopic Ellipsometer
System as utilized to obtain data from thin doped membranes as
reported in this Specification.
[0044] FIG. 2 shows, for a first thin impurity profile containing
membrane, reflection mode ellipsometric DELTA vs. Wavelength in
microns, obtained utilizing an angle of incidence of 75 degrees,
for a thin film silicon membrane which was found to be 7.66 microns
thick.
[0045] FIG. 3 demonstrates actual reflection mode ellipsometric
DELTA data obtained utilizing an angle of incidence of 68 degrees
to the same impurity profile containing thin membrane that was used
to obtain data shown in FIG. 2.
[0046] FIG. 4 shows that, for the same thin film membrane used to
obtain the data in FIGS. 2 and 3, where IR range wavelengths are
utilized, where present, the impurities cause said (n) and (k) to
vary, but where said impurities are absent, (k) for instance,
quickly becomes essentially zero.
[0047] FIGS. 5 and 6 show PSI and DELTA vs. Wave Number, (where
Wavelength in Microns is obtained by dividing 10,000 by the Wave
Number), for a second impurity profile containing thin film
sample.
[0048] FIGS. 7 and 8 show Psuedo Dielectric Functions for said
second thin film sample.
[0049] FIG. 9 shows (%) Depolarization for said second thin film
sample.
[0050] FIGS. 10 and 11 show Real and Imaginary Dielectric Constants
as a function of depth into the thin membrane, wherein 0.0 is the
surface thereof near the Doping profile for said second thin film
sample.
[0051] FIG. 12 shows a Log (N) vs. Microns from the Doped Surface
of the Thin Membrane for said second thin film sample.
DETAILED DESCRIPTION
[0052] The present invention is a spectroscopic ellipsometer system
based method which utilizes wavelengths in the near-infrared,
(near-IR), (ie. 1-1.7 micron), and infrared (IR), (ie. 2-35
micron), ranges. It is noted that in the near-IR wavelength range,
silicon is essentially transparent and impurity, (ie. P-type
doping), effects are negligible, but that in the IR wavelength
range, while un-doped silicon remains essentially transparent, the
"doped" silicon becomes essentially opaque and reflective because
of the "metallic" presence of free carriers.
[0053] A demonstrative Spectroscopic Ellipsometer System as
utilized to obtain data utilized in the present work, is shown in
FIG. 1. Note the presence of a Source of Electromagnetic Radiation
(LS), a Polarizer (P) for producing a Polarized beam of
electromagnetic radiation (PPCLB), a Rotating Compensator (C),
Stage (STG) for supporting a Material System (MS), an Analyzer (A),
a Dispersive Optics (DO) and a Detector (DET) comprised of a
plurality of Detector Elements (DE), each of which is positioned to
intercept a different wavelength. Also indicated is Compensator
(C') to show that said Rotating Compensator can be placed on either
side of the Material System (MS). In most such systems only one (C)
or (C') is present, however. Further shown is a Focusing Lens (FE)
which can be present to converge an electromagnetic beam (EPCLB)
which passes through the Analyzer (A), onto the Dispersive Optics
(DO). In use, the Polarizer (P) and Analyzer (A) are typically set
to an azimuthal angle and held motionless, and the Compensator (C)
or (C') is caused to rotate while an electromagnetic beam (PPCLB)
is caused to impinge upon the Material System (MS) such that
spectroscopic data is collected by the Detector (DET).
[0054] Preferred present invention methodology provides that
electromagnetic radiation of near-IR wavelengths be utilized in a
spectroscopic ellipsometer system to acquire a reflection or
transmission data set which is then applied, via mathematical
regression, to evaluate a thickness parameter in a mathematical
model of an impurity profile containing thin membrane. Actual
experimental results have been acquired utilizing a J.A. Woollam
CO. Inc. (M-2000 NIR)tm spectroscopic ellipsometer system
configured in a reflection mode, with the angle of incidence of the
electromagnetic beam to the investigated surface of the impurity
profile containing thin membrane being near the Brewster Angle
thereof, in acquiring said data set in this step. See FIG. 2 which
shows reflection mode ellipsometric DELTA vs. Wavelength in
microns, obtained utilizing an angle of incidence of 75 degrees,
for a thin film silicon membrane which was found to be 7.66 microns
thick. Note that back-side reflections cause said FIG. 2 data to
show interference related effects. In passing, it is noted that the
thickness of the impurity profile containing thin membrane is
closely related to the spacing between the cyclic peaks in said
data, as modified by the refractive index of the silicon. (Note,
"E" indicates data obtained while investigating the front of sample
and "Er" the Surface near the Doping).
[0055] With thickness of said impurity profile containing thin
membrane thus determined, the preferred present invention method
involves obtaining two reflection mode data sets utilizing
electromagnetic radiation of IR wavelengths in a spectroscopic
ellipsometer system. One of said reflection mode data sets is
obtained with the electromagnetic beam caused to impinge upon one
of two substantially parallel surfaces of an impurity profile
containing thin membrane, and the second data set is obtained by
causing the electromagnetic beam caused to impinge upon the other
of said two substantially parallel surfaces. Actual experimental
results have been acquired utilizing a J.A. Woollam CO. Inc.
(IR-VASE) (Registered Trademark), spectroscopic ellipsometer system
configured in a reflection mode, with the angle of incidence of the
electromagnetic beam to an investigated surface of the impurity
profile containing thin membrane being below the Brewster Angle
thereof, (eg. at 68 degrees), while acquiring said data set in this
step.
[0056] It is noted that where an impurity profile containing thin
membrane has the impurities concentrated near one of said two
substantially parallel surfaces but removed from the opposite
substantially parallel surface, that data obtained with the
electromagnetic beam impinging on said opposite surface will
produce data which show the results of interference based in the
presence of back-side reflections. Where the electromagnetic beam
is caused to impinge on the substantially parallel surface near the
impurities, acquired data does not demonstrate interference
effects, as said impurities cause the silicon to be essentially
opaque and reflective of said electromagnetic radiation, thus
preventing back-side reflections. FIG. 3 demonstrates actual
reflection mode ellipsometric DELTA data obtained utilizing an
angle of incidence of 68 degrees to the same impurity profile
containing thin membrane that was used to obtain data shown in FIG.
2. Note the onset of Interference prior the wavelength of
approximately 3.3 microns, in the data curve obtained when
investigating the substantially parallel surface near the
impurities, indicating an approach to the case similar to that
shown in FIG. 2. While FIG. 2 data was obtained with the near-IR
beam impinging on the opposite substantially parallel surface, (ie.
the substantially parallel surface removed from the impurities), in
the near-IR wavelength range the impurity effects are essentially
negligible, hence, backside reflections occur regardless of which
substantially parallel surface is investigated. It is also noted
that non-parallel surfaces, or uneven thickness, can cause
interference effects.
[0057] FIG. 4 shows refractive index (n) and extinction coefficient
(k) data for the impurity profile containing thin membrane that was
used in providing FIGS. 2 and 3. Note that the location identified
as (0) on the "X" axis, is the substantially parallel surface near
which is the impurity profile, and the downward vertical line at
the right indicates the opposite surface. FIG. 4 shows that where
IR range wavelengths are utilized, where present, the impurities
cause said (n) and (k) to vary, but where said impurities are
absent, (k) for instance, quickly becomes essentially zero.
[0058] In the work which produced the foregoing results, the
refractive index was modeled in the near-IR with as a
Cauchy-dispersion with a Urbach absorption tail. The doping effects
in the IR wavelength range were modeled by a Drude model as
described in the previously referenced Tiwald et al. article.
[0059] It is to be understood that while initial work determined
impurity profile containing thin membrane thickness in a first
regression utilizing near-IR wavelength range data, and used a
second regression based upon two data sets based on IR wavelength
range data, it is within the scope of the present invention to
obtain all described near-IR and IR wavelength range data and
simultaneously regress thereonto to simultaneously evaluate
mathematical model thickness and impurity profile describing
parameters.
[0060] As additional insight, it is mentioned that if an impurity
profile containing thin membrane has the impurities concentrated
centrally therewithin, data obtained by investigation of either of
said two substantially parallel surfaces with a spectroscopic
ellipsometer system configured in a reflection mode, and using IR
range wavelengths in a beam of electromagnetic radiation caused to
impinge thereupon, (eg. at below the Brewster Angle), will show
interference effects.
[0061] Table 1 summarizes actual experimentally arrived-at results
for the impurity profile containing thin membrane investigation of
which provided data shown in FIGS. 2-4.
1 TABLE 1 NATIVE OXIDE THICKNESS (TOP) 0.002 MICRON SILICON
DEPLETION REGION 6.0383 MICRON GRADED SILICON FREE CARRIERS 1.2517
MICRON NATIVE OXIDE THICKNESS (BOTTOM) 0.0061 MICRON
[0062] While the above described FIGS. 2-4 provide example to the
basic application of the present invention methodology, additional
thin membranes have also been investigated by similar techniques,
but wherein data was obtained at multiple Angles-of-Incidence, such
as 60, 65, and 70 degrees with respect to the thin membrane
surfaces. All data was then utilized in regression evaluation of
Thin Membrane related parameters.
[0063] FIGS. 5-12 show various data for one such additional Thin
Membrane system which is characterized as in Table 2:
2 TABLE 2 THERMAL OXIDE 0.0 MICRON SILICON DEPLETION 3.1199 MICRON
GRADED P-TYPE FREE CARRIER 1.6882 MICRON P-TYPE FREE CARRIER 0.0
MICRON THERMAL OXIDE 0.0010637 MICRON VOID 1.0 MICRON
[0064] Note in said Figures, "E" indicates data obtained while
investigating the front of sample, and "Er" the Surface near the
Doping. FIGS. 5 and 6 show PSI and DELTA vs. Wave Number, (where
Wavelength in Microns is obtained by dividing 10,000 by the Wave
Number). FIGS. 7 and 8 show Psuedo Dielectric Functions. FIG. 9
shows (%) Depolarization, (where (%) Depolarization is a measure of
how much COS2 (DELTA)+SIN.sup.2(DELTA) deviates from 1.0). FIGS. 10
and 11 show Real and Imaginary Dielectric Constants as a function
of depth into the thin membrane, (wherein 0.0 is the surface
thereof near the impurity profile). Finally FIG. 12 shows a Log (N)
vs. Microns from the Doped Surface of the Thin Membrane, where (N)
is concentration per centimeter cubed.
[0065] It is to be understood that the terminology "two
substantially parallel surfaces that are separated by a thickness"
can identify impurity profile containing thin membranes which are
not strictly comprised of two precisely parallel surfaces separated
by a strictly unvarying thickness, but that said terminology serves
to identify a material system which is primarily comprised of two
surfaces which have areas defined by effective length and width
dimensions which are significantly greater than a thickness
separating them.
[0066] Additionally, the terminology, "a range of wavelengths over
which the impurity profile containing thin membrane is essentially
transparent, but over which the effect of the presence of said
impurity profile has a non-negligible effect", does not strictly
exclude all wavelengths at which an impurity profile is somewhat
transparent, but only serves to identify a range of wavelengths
over which a beam of electromagnetic radiation comprised thereof
substantially reflects from said impurity profile.
[0067] It is mentioned that both the (IR-VASE) (Registered
Trademark), and the (M-2000 NIR)tm spectroscopic ellipsometer
systems are of Rotating Compensator design, hence both are able to
measure DELTA values at zero (0.0) degrees.
[0068] Finally, while Rotating Compensator Ellipsometer Systems
were used as examples in this Specification, any Ellipsometer
System which can provide the necessary data are to be considered
within the scope of the claims.
[0069] Having hereby disclosed the subject matter of the present
invention, it should be obvious that many modifications,
substitutions, and variations of the present invention are possible
in view of the teachings. It is therefore to be understood that the
invention may be practiced other than as specifically described,
and should be limited in its breadth and scope only by the
claims.
* * * * *