semi合集-English.pdf - 第7431页
SEMI MF1529-1104 © SEMI 2004 12 15.1.1 The wafers in these tests were fabricated by five different p rocesses. The pur pose of these t ests is to demonstrate th e consistency of determination of sheet resistance uniformi…
SEMI MF1529-1104 © SEMI 2004 11
12.6.2 The current at each measurement site may be
kept at the value used for the first site, or it may be
readjusted for each site to give a specimen voltage
across the specimen sites that is within very tight limits.
Whenever the probes are lowered or the current is
changed, make sure that there is a short pause to ensure
that the probes, the current supply, and the
measurement voltage lines have stabilized before taking
data. Depending on the wafer type and the design of
the instrumentation, this may be from a fraction of a
second to several seconds.
12.6.3 Measure and record the temperature of the stage
to the nearest 0.1°C at least at the beginning and the
end of the measurements. Actual measurement of
temperature for each measurement site is preferred if
experience indicates that noticeable drift or fluctuation
(greater than 1°C) is likely to occur during the course of
the measurements.
13 Calculations
13.1 Calculate the Resistances from Each
Configuration for Each Site
13.1.1 Calculate the average of the forward voltage-
current ratio and the reverse voltage-current ratio for
Configuration A as follows to obtain R
A
:
)41(
)32(
)41(
)32(
2
1
r
r
f
f
A
I
V
I
V
R
(5)
13.1.2 Calculate the average of the forward voltage-
current ratio and the reverse voltage-current ratio for
Configuration B as follows to obtain R
B
(see Note 15):
)31(
)42(
)31(
)42(
2
1
r
r
f
f
B
I
V
I
V
R
(6)
13.1.3 The forward and reverse voltage-current ratios
must agree within 5% of the larger of these ratios both
for R
A
and for R
B
in order to be acceptable for use in
referee measurements.
13.2 Calculate Sheet Resistance at Each Site — For
each measurement site, calculate the sheet resistance,
R
s
, as follows:
Aas
RKR (7)
where:
2
2
872.7173.25696.14
B
A
B
A
a
R
R
R
R
K (8)
13.3 Reviewing the Data — Many applications of
uniformity testing involve acquisition of large amounts
of data. It is generally beneficial to review these data
before interpreting them simply as manifestations of
sheet resistance non-uniformity. A useful technique is
to plot the values obtained for R
s
as a time sequence,
that is, in the order the data for the site were taken for
these sites. Examination of this plot may be correlated
with measurement site coordinates to reveal unexpected
edge proximity effects or regions of the wafer that have
unusual non-uniformity that may need to be re-probed
with a different site-selection plan. Examination may
also reveal one or more points that are highly
inconsistent with adjacent points, possibly indicating
problems due to vibrations or spots of surface
contamination. Note such observations, but do not
delete or edit the data, unless agreed upon by all parties
to the test.
13.4 Analysis and Summary of Data — The sheet
resistance data may be analyzed and summarized in a
number of ways as agreed upon by parties to referee
test, or as appropriate for process control or other
applications. These may include, but are not limited to:
listing of all data with site coordinates, contour plots of
the deviations from average value, and distributional
statistics of the data.
14 Report
14.1 The report shall include the following information:
14.1.1 Operator's name, date and time of
measurements,
14.1.2 Wafer identification number and description,
14.1.3 Identification of instrument used, by
manufacturer, serial number, and model number,
14.1.4 Identification of probe used, by manufacturer,
serial number and probe spacing, probe tip radius and
material, and probe force specifications,
14.1.5 Statement of site location plan used,
14.1.6 Initial and final temperature at wafer stage for
the measurement sequence,
14.1.7 Summary of sheet resistance data as chosen in
connection with Section 13.4,
14.1.8 Notations about aberrant measurement sites as
identified by any data screening procedures employed,
and
14.1.9 Data from probe qualification tests.
15 Precision and Bias
15.1 Precision — Single laboratory values for
repeatibility of the measurement of wafer uniformity
were obtained from a series of tests run in 1991. For
these tests, five types of wafers with average sheet
resistance values from about 15 to about 400 were
tested, as detailed in Related Information 1.
SEMI MF1529-1104 © SEMI 2004 12
15.1.1 The wafers in these tests were fabricated by five
different processes. The purpose of these tests is to
demonstrate the consistency of determination of sheet
resistance uniformity with dual-configuration four-
probe measurements, not to determine the uniformity
achievable by a given process. The five processes
utilized are capable of both better and poorer uniformity
than seen here.
15.1.2 The repeatability of uniformity values given here
represent good measurement practice, but the
consistency of values for each wafer might have been
even better if all measurements had been taken over a
shorter time interval.
15.1.3 A standard deviation, (in percent), was used to
represent the uniformity values obtained for each test
on each of the wafers. While it is common to give
repeatability values as a standard deviation, in order to
avoid confusion from duplication of terminology, the
repeatability of the repeated determinations of non-
uniformity for each type wafer are given in Table R1-1
as the range of standard deviation values obtained in the
original tests.
15.1.4 A straightforward propagation of errors based on
the performance specifications, in order to estimate the
precision, does not appear possible. Required
instrument resolution, power supply stability, and probe
performance qualification are designed to allow a
relative accuracy and precision of better than 0.1% for
all measurement sites on a wafer. However, for wafers
with high uniformity (for example, standard deviation
of all measurement site values on the order of 0.2%),
even these requirements may not ensure good two-
party, or multilaboratory agreement on the wafer
uniformity.
15.2 Bias — A statement of bias cannot be made
because there are no semiconductor reference artifacts
with a known level of non-uniformity.
16 Keywords
16.1 epitaxy; four-point probe; ion implant;
metallization; polysilicon; sheet resistance; silicon
SEMI MF1529-1104 © SEMI 2004 13
RELATED INFORMATION 1
SINGLE LAB TEST RESULTS
NOTICE: This related information is not an official part of SEMI MF1529. It was derived from information
developed during the original preparation of the standard in ASTM Committee F-1 in 1994. This related
information was approved for publication by full letter ballot procedures on August 16, 2004.
R1-1 One wafer from each of five different thin film
types was tested for uniformity. Each was 100 mm
diameter. A measurement test-site plan using 81 test
locations located at wafer center and on four equally
spaced circles with a maximum radius of 38.1 mm, as
described in SEMI MF1618, was used for all
measurements. The measure of uniformity for
measurements on each wafer was the standard deviation
of the 81 measured values expressed as a percent of the
mean measured value. For each of the wafers, 11 to 13
tests of uniformity were run over periods that ran from
3 to 6 months, depending on the individual wafer.
R1-1.1 Temperatures of the wafer stage were
monitored and recorded. Average sheet resistances
were found to change as a function of temperature but
no corresponding corrections of sheet resistance values
were made because the temperature coefficients of
resistivity were not known for the films involved. The
assumption was made that the determination of
uniformity should be minimally affected as long as the
temperature stayed relatively constant during the course
of each measurement run.
R1-1.2 Results of this test are summarized in Table
R1-1. The highest temperatures used for each wafer
were achieved by deliberate elevation of room
temperature in order to determine whether temperature
affected average sheet resistance or uniformity value.
Temperature dependence of average sheet resistance
was seen for wafer Types 1, 2 and 5. No dependence of
sheet resistance uniformity on temperature was seen.
R1-2 An additional test involved three measurements
of uniformity on each of three different wafers from the
same five thin-film fabrication processes. These tests
were completed within three days. They were also
associated with evaluation of the repeatability
capability of the sheet resistance measurement
instrumentation. For this evaluation of the instrument,
50 measurements were made on each wafer by taking
10 very closely spaced measurements in each of five
relatively widely separated locations on the wafer. The
assumption was made that in each group of ten, the
measurements were so closely spaced compared to the
sampling volume of the four-point probe that any
variability experienced was a measure of the equipment
repeatability (imprecision) and was not caused by
variations in the thin film itself. The actual
measurements of wafer uniformity used the same 81-
point test pattern as used for the wafers in Section R1-1
.
R1-2.1 The measurements from this test, summarized
in Table R1-2, show that for each process, the three
wafers have a different average sheet resistance, even
when measured at the same nominal temperature and
they also have rather different levels of non-uniformity
as determined from the standard deviations of the 81
measurements. The variation of these run-to-run
standard deviations for a given wafer is seen to be small
compared to the typical wafer-to-wafer standard
deviations for the different wafers from a given process.
Table R1-1 Summary of Sheet Resistance Uniformity Tests Made by Dual-Configuration Four-Point Probe
Measurements at One Laboratory on One Wafer from Each of Five Thin Film Fabrication Processes
Process
Nominal
Resistance,
Number of Test
Runs
Range of
Temperatures, °C
Range of Mean
Values,
#1
Range of
Uniformity Values,
%
(1) Boron Implant
1.3 × 10
13
/60 keV
570 13 19.35–28.45 566.3–574.6 1.83–1.89
(2) Phosphorus Implant
5 × 10
14
/100 keV
144 13 20.21–27.73 143.4–144.4 0.81–0.86
(3) Arsenic Implant
5 × 10
15
/60 keV
26 13 20.94–32.48 26.33–26.78 0.67–0.69
(4) Polysilicon
600-nm thickness
900° deposition temperature
14.5 11 21.41–26.00 14.48–14.55 0.24–0.24
(5) n/p-epitaxy 3 m/1.2 cm
4,000 11 20.23–30.92 4021–4301 1.53–1.62
#1
Measurements were not corrected for changes in temperature.