semi合集-English.pdf - 第7470页
SEMI MF1617-0304 © SEMI 2003, 2004 6 calculated as 2.8 times th eir respective standa rd deviations ( s r , S R ). 14.1.1 SIMS measurements were m ade on samples taken from an iron contaminated silicon wafer to estimate …
SEMI MF1617-0304 © SEMI 2003, 2004 5
contaminants from the free surface or native oxide layer
on the specimen should ordinarily reach a maximum
within the first 1 nm of the depth profile, and decrease
monotonically thereafter. If this general profile shape
is not obtained, repeat the measurement on a fresh
surface of the specimen.
11.2.2 When the impurity signal has either: (1)
decreased to at least <1% of the maximum, (2) or
reached a constant background count rate, then measure
and record the silicon matrix ion count rate on the
appropriate detector.
11.2.3 Measure the sputter rate under the analysis
conditions. This can be done by measuring the SIMS
crater depth with a stylus profilometer or equivalent
instrument if available, and combining the depth
measurement with the recorded time elapsed during the
sputter profile. If the analytical craters are too shallow
to be measured by available instrumentation, the sputter
rate must be monitored during a measurement series by
repeated analysis of a previously characterized standard
sample.
11.2.4 Measure the BLANK under the same analytical
conditions as the unknown specimens.
12 Calculations
12.1 The RSF for an element in a matrix can be
determined experimentally from the SIMS profile of a
reference material containing a known areal density of
the impurity of interest as follows:
()
NISId
tDNI
bi
m
−
=RSF (1)
where:
D = the areal density of the impurity in
atoms/cm
2
,
N = the number of data cycles in the profile,
I
m
= the matrix isotope secondary ion intensity,
(counts/s),
d = the sputtered depth, cm,
SI
i
= the sum of the impurity isotope secondary
ion-counts over the depth of the profile,
I
b
= a constant background intensity of the
impurity isotope, and
t = the analysis time for the species of interest,
s/cycle.
12.2 The calibration of concentration in a SIMS profile
is calculated as follows:
=
EM
FC
I
I
C
m
i
i
RSF (2)
where:
C
i
= the impurity atomic concentration,
atoms/cm
3
,
I
i
= the isotope secondary in intensity, counts/s,
I
m
= the matrix isotope secondary ion-intensity,
counts/s,
RSF = relative sensitivity factor, atoms/cm
3
, and
FC
EM
= the ratio of matrix intensities on the faraday
cup (FC) to the electron multiplier (EM),
when two detectors are used.
12.3 The conversion of data cycles or time into depth is
completed by measuring the crater depth and the total
time (or data cycles) of the sputtering that formed the
crater. The depth scale is assumed to be linear with
time or data cycles.
12.4 Calculate the RSFs and impurity concentration
versus depth from Equations 1 and 2.
12.5 Calculate the surface areal density by integrating
the concentration versus depth profile after any
subtraction of a constant background count rate that
may be appropriate.
13 Report
13.1 Report the following information:
13.1.1 The instrument used, the operator, and the date
of the measurements,
13.1.2 Identification of test, BLANK, and reference
specimens,
13.1.3 Calibration procedure used,
13.1.4 Method used to discriminate molecular ion
interferences,
13.1.5 Surface areal density of the impurity, and
13.1.6 Surface areal density of the impurity for the
BLANK specimen.
14 Precision and Bias
7
14.1 Precision — Tables 1, 2, and 3 summarize the
statistics from a spoke-wheel round robin including 4
test samples and 10 laboratories. Each laboratory was
provided with a spin coated reference standard of
surface sodium, aluminum, and potassium. The 95%
repeatability, r, and reproducibility, R, limits are
7 Support data are available from SEMI Headquarters. Request
SEMI MF1617 round robin report (formerly ASTM RR: F01-1010).
SEMI MF1617-0304 © SEMI 2003, 2004 6
calculated as 2.8 times their respective standard
deviations (s
r
, S
R
).
14.1.1 SIMS measurements were made on samples
taken from an iron contaminated silicon wafer to
estimate the precision of the surface iron measurement.
Two SIMS instruments were used: a CAMECA IMS 3f
and a CAMECA IMS 4f. All measurements used a 3-
keV oxygen ion beam with an oxygen flood.
Measurements were taken over a one-year period. For
the CAMECA IMS 3f, 23 measurements were taken
with an average iron reading of 1.51 × 10
11
atoms/cm
2
and a 1 standard deviation of 0.148 × 10
11
atoms/cm
2
.
For the CAMECA IMS 4f, 20 measurements were
taken with an average iron reading of 1.71 × 10
11
atoms/cm
2
and 1 standard deviation of 0.188 × 10
11
atoms/cm
2
.
14.2 Bias — Bias cannot be estimated because there
are no accepted absolute standards. However, for
comparison only, the VPD/AAS results for the sodium,
aluminum, and potassium test samples are included in
Tables 1–3. A correlation between SIMS and TXF
quantitive results for iron is shown.
8
15 Keywords
15.1 aluminum; iron; potassium; silicon; SIMS;
sodium; surface contamination
Table 1 Summary Statistics for Sodium (Units of
10
10
Atoms/cm
2
)
Sample VPD/AAS X-Bar s
r
S
R
r R
A 10–15 10.68 2.240 2.791 6.273 7.816
B 29–32 33.64 3.339 4.931 9.348 13.81
C 115–121 112.1 16.19 19.79 45.32 55.41
E 0.6–4 0.665 0.359 0.861 1.005 2.410
Table 2 Summary Statistics for Aluminum (Units of
10
10
Atoms/cm
2
)
Sample VPD/AAS X-Bar s
r
S
R
r R
A 3 3.504 0.6865 1.107 1.922 3.100
B 7–8 9.425 0.7684 2.239 2.152 6.269
C 22–25 28.80 2.298 6.107 6.434 17.10
E not detected 0.795 0.390 0.760 1.093 2.128
8 Smith, S. P., and Metz, J., “Understanding the Correlation of
Surface SIMS and TXRF Measurements of Surface Metal
Contamination on Silicon Wafers,” Science and Technology of
Semiconductor Surface Preparation, edited by G. S. Higashi, M.
Hirose, S. Raghavan, and S. Verhaverbeke, Material Research
Society Symposium Proceedings, Vol 477 (Materials Research
Society, Pittsburgh, PA, l997) pp. 305–310.
Table 3 Summary Statistics for Potassium (Units of
10
10
Atoms/cm
2
)
Sample VPD/AAS X-Bar s
r
S
R
r R
A 7–8 7.829 1.642 2.761 4.596 7.731
B 22–23 24.11 3.607 5.943 10.10 16.64
C 92 82.37 12.77 19.63 35.74 54.97
E 0.1–2 0.407 0.162 0.344 0.454 0.965
NOTICE: SEMI makes no warranties or
representations as to the suitability of the standards set
forth herein for any particular application. The
determination of the suitability of the standard is solely
the responsibility of the user. Users are cautioned to
refer to manufacturer' s instructions, product labels,
product data sheets, and other relevant literature,
respecting any materials or equipment mentioned
herein. These standards are subject to change without
notice.
By publication of this standard, Semiconductor
Equipment and Materials International (SEMI) takes no
position respecting the validity of any patent rights or
copyrights asserted in connection with any items
mentioned in this standard. Users of this standard are
expressly advised that determination of any such patent
rights or copyrights, and the risk of infringement of
such rights are entirely their own responsibility.
Copyright by SEMI® (Semiconductor Equipment and Materials
International), 3081 Zanker Road, San Jose, CA 95134. Reproduction o
f
the contents in whole or in part is forbidden without express written
consent of SEMI.
SEMI MF1618-1104 © SEMI 2004 1
SEMI MF1618-1104
PRACTICE FOR DETERMINATION OF UNIFORMITY OF THIN FILMS
ON SILICON WAFERS
This guide was technically approved by the Global Silicon Wafer Committee and is the direct responsibility
of the North American Silicon Wafer Committee. Current edition approved for publication by the North
American Regional Standards Committee on August 16, 2004. Initially available at www.semi.org
September 2004; to be published November 2004. Original edition published by ASTM International as
ASTM F 1618-95. Last previous edition SEMI MF1618-02.
1 Purpose
1.1 The fabrication of semiconductor, dielectric, and
metal thin films is a critical part of silicon integrated
circuit production. The variation of film properties
across the surface of a wafer can have significant
impact on the further processing of the wafer and on the
ultimate yield of acceptable chips from the wafer, as
well as on their reliability.
1.2 The purpose of this practice is to promote
commonality of approach to the analysis of uniformity
among all parties needing to generate or assess such
information, including manufacturers of the basic test
instrumentation to be used.
1.3 This practice is intended for process control,
research and development, and process equipment
evaluation purposes. It is intended for the benefit of
semiconductor device and equipment manufacturers
alike so that acquisition, reduction, and communication
of thin film data is consistent among various parties
who may need to concur on interpretation of the results
of a thin film fabrication-process step.
1.4 Measurement of the uniformity of one or more thin
film properties such as thickness, sheet resistance,
reflectivity, dielectric constant or index of refraction
enables the monitoring of a critical aspect of the results
of a given process step. This information can be used
to determine the behavior of individual process steps; it
can be used with similar information from earlier
process steps to determine their interaction with respect
to final product uniformity. Also, it can be used in
conjunction with historical data from the same process
step to determine loss of control of the piece of
equipment or the process cycle being used with respect
to producing product having a required level of
uniformity. Further, it can be used to determine the
adequacy of new or modified process cycles, materials
or equipment for a given film deposition or film
modification requirement.
1.4.1 It is common practice to monitor the value of a
simple statistic, such as the standard deviation, that
results from layer uniformity measurement data and to
compare the current value of the statistic with historical
values for the same layer formation process. An
increase of the current value over historical values is
taken as an indication of possible deterioration of the
quality of the layer formation process. It is then
common to convert the data into a contour or similar
map of uniformity to aid in diagnosing changes in the
process that caused the increase in standard deviation
(or similar statistic). This practice does not treat the
interpretation of the statistic resulting from data
analysis, nor does it give procedures for converting the
data to a uniformity map.
1.5 The non-uniformities of a film property on a given
wafer are primarily systematic, not random, in their
spatial shape or distribution and arise from spatially
systematic variations in such process variables as
temperature, gas flow, pressure, or electric field. As a
result, the simple statistic standard deviation that is
specified for analysis of data acquired with this practice
will not generally have the normal interpretation for the
standard deviation of a sample from a random
population. It is a figure of merit for comparing data
sets of a similar type, but it cannot be used for
computing confidence or tolerance intervals.
2 Scope
2.1 This practice covers a set of site distribution
patterns for measuring the uniformity of a property of a
thin film on a silicon wafer, as well as simple
procedures for analyzing and reporting the results of
those measurements.
2.2 This practice is intended for use as a template for
the evaluation of the uniformity of intrinsic film
properties such as thickness or composition, and also
film functional characteristics such as sheet resistance
and reflectivity. The resulting information may be used
to assess the uniformity of the film itself or of the layer
formation process. This practice is not directly
applicable to evaluating wafer-to-wafer or lot-to-lot
variations.
2.3 This practice is intended for use with any thin film
or layer type, or formation technique, for which basic
measurement instrumentation and capability exists that,
is appropriate to the film parameter of interest. This
practice is intended for layer growth and deposition
techniques such as epitaxy, implantation, thermal and