ASTM F398-92(2002)
(Test Method)Standard Test Method for Majority Carrier Concentration in Semiconductors by Measurement of Wavenumber or Wavelength of the Plasma Resonance Minimum (Withdrawn 2003)
Standard Test Method for Majority Carrier Concentration in Semiconductors by Measurement of Wavenumber or Wavelength of the Plasma Resonance Minimum (Withdrawn 2003)
SCOPE
This standard was transferred to SEMI (www.semi.org) May 2003
1.1 This test method covers determination of the wavenumber of the plasma resonance minimum in the infrared reflectance of a doped semiconductor specimen, from which the majority carrier concentration can be obtained.
1.2 This test method of determination of the wavenumber minimum is nondestructive and contactless. It is applicable to n- and p-type silicon, n- and p-type gallium arsenide, and n-type germanium.
1.3 This test method gives a relative measurement in that the relation between the wavenumber of the plasma resonance minimum and the majority carrier concentration is empirical. Such relations have been established for the several cases summarized in Annex A1. These relations are based upon determinations of the plasma resonance minimum by the procedure of this method and determinations of the Hall coefficient according to Test Methods F76 (Section 2) or resistivity according to Test Methods F43 or Test Method F84 (Section 2) as appropriate.
1.4 These relations have been established over a majority carrier concentration range from 1.5 X 10 to 1.5 X 10 cm -3 for n-type silicon, from 3 X 10 to 5 X 10 for p-type silicon, from 3 X 10 to 7 X 10 for n-type germanium, from 1.5 X 10 to 1 X 10 for n-type gallium arsenide, and from 2.6 X 10 to 1.3 X 10 cm -3 for p-type gallium arsenide.
1.5 These relations can be extended or developed for other materials by measuring the wavelength of the plasma resonance minimum according to this procedure on specimens whose majority carrier concentration has been determined by other means.
1.6 This test method is applicable to both bulk and diffused material. However, since there is some controversy over the effects of variations of junction depth on the measurement, it should be applied to surface concentration measurements on shallow (1 μm or less) diffusions only on a relative basis unless there is experimental corroboration of the results under the conditions of interest.
1.7 This standard does not purport to address all of the safety problems, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use.
General Information
Relations
Standards Content (Sample)
NOTICE: This standard has either been superceded and replaced by a new version or discontinued.
Contact ASTM International (www.astm.org) for the latest information.
Designation: F 398 – 92 (Reapproved 2002)
Standard Test Method for
Majority Carrier Concentration in Semiconductors by
Measurement of Wavenumber or Wavelength of the Plasma
Resonance Minimum
This standard is issued under the fixed designation F 398; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (e) indicates an editorial change since the last revision or reapproval.
INTRODUCTION
As originally published, this test method was written for infrared instruments with output in units
of wavelength. Calibration data, as included, were based on infrared measurements in units of
wavelength and on a combination of concentration scale methods including Hall effect, neutron
activation analysis, and four-probe resistivity. These latter measurements were converted, using the
empirical data of Irvin, from resistivity to “impurity concentration.” The result of this procedure is
calibration data that is a mixture of carrier concentration and impurity concentration values. While the
differences between them is large only at very high concentrations, there are values in the original
calibration plots for silicon (Fig. A1.1 and Fig. A1.2) that are above currently accepted solid-solubility
limit values.
Common practice current infrared spectrophotometers is to give output in units of wavenumbers
−1
(cm ). In order to relate this test method and its original calibration to current infrared units without
introducing numerical error (the calibration relation is not analytically invertible to wavenumbers), the
direct substitution of 10 000/wavenumber for wavelength is given in the analysis equation and in the
tables of coefficients.
This test method is not believed to have wide current use in the semiconductor industry. However,
because this test method may be useful for some applications and because the calibration data
contained herein is believed to be available nowhere else in the archival literature, this test method is
being retained as a standard.
1. Scope determinations of the plasma resonance minimum by the
procedure of this method and determinations of the Hall
1.1 This test method covers determination of the wavenum-
coefficient according to Test Methods F 76 (Section 2) or
ber of the plasma resonance minimum in the infrared reflec-
resistivity according to Test Methods F 43 or Test Method F 84
tance of a doped semiconductor specimen, from which the
(Section 2) as appropriate.
majority carrier concentration can be obtained.
1.4 These relations have been established over a majority
1.2 This test method of determination of the wavenumber
18 21
carrier concentration range from 1.5 3 10 to 1.5 3 10
minimum is nondestructive and contactless. It is applicable to
−3 18 20
cm for n-type silicon, from 3 3 10 to 5 3 10 for p-type
n- and p-type silicon, n- and p-type gallium arsenide, and
18 19
silicon, from 3 3 10 to 7 3 10 for n-type germanium, from
n-type germanium.
17 19
1.5 3 10 to 1 3 10 for n-type gallium arsenide, and from
1.3 This test method gives a relative measurement in that
18 20 −3
2.6 3 10 to 1.3 3 10 cm for p-type gallium arsenide.
the relation between the wavenumber of the plasma resonance
1.5 These relations can be extended or developed for other
minimum and the majority carrier concentration is empirical.
materials by measuring the wavelength of the plasma reso-
Such relations have been established for the several cases
nance minimum according to this procedure on specimens
summarized in Annex A1. These relations are based upon
whose majority carrier concentration has been determined by
other means.
This test method is under the jurisdiction of ASTM Committee F01 on 1.6 This test method is applicable to both bulk and diffused
Electronics and is the direct responsibility of Subcommittee F01.06 on Silicon
material. However, since there is some controversy over the
Materials and Process Control.
effects of variations of junction depth on the measurement, it
Current edition approved May 15, 1992. Published July 1992. Originally
e1
should be applied to surface concentration measurements on
published as F 398 – 74 T. Last previous edition F 398 – 87 .
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.
NOTICE: This standard has either been superceded and replaced by a new version or discontinued.
Contact ASTM International (www.astm.org) for the latest information.
F 398 – 92 (2002)
shallow (1 μm or less) diffusions only on a relative basis unless 5. Significance and Use
there is experimental corroboration of the results under the
5.1 The measurement of the carrier concentration of a
conditions of interest.
semiconductor is a vital part of the development and manufac-
1.7 This standard does not purport to address all of the
ture of semiconductor devices and integrated circuits.
safety concerns, if any, associated with its use. It is the
5.2 An important use of this technique is the measurement
responsibility of the user of this standard to establish appro-
of the surface concentration of diffused layers. There is,
priate safety and health practices and determine the applica-
however, some controversy about the effect of variation of
bility of regulatory limitations prior to use.
junction depth on the correspondence between the concentra-
tion measured by this technique and the true concentration,
2. Referenced Documents
particularly for very shallow (1 μm or less) junctions.
2.1 ASTM Standards:
5.3 This technique also may be applied to the measurement
E 275 Practice for Describing and Measuring Performance
of bulk materials and to the study of nonuniformities across the
of Ultraviolet, Visible, and Near Infrared Spectrophotom-
surface of those materials. For such application the user should
eters
be certain that the surfaces are free of the films which normally
F 42 Test Methods for Conductivity Type of Extrinsic
appear on an as-grown crystal.
Semiconducting Materials
5.4 It must be noted that the reflectivity minimum measured
F 43 Test Methods for Resistivity of Semiconductor Mate-
by this procedure and associated with the plasma resonance
rials
occurs at a larger wavenumber value than that associated with
F 76 Test Methods for Measuring Resistivity and Hall
the actual plasma resonance. The latter occurs at a wavenumber
Coefficient and Determining Hall Mobility in Single-
3 value which can be calculated from the equation:
Crystal Semiconductors
F 84 Test Method for Measuring Resistivity of Silicon
2 2 2
5l 5 ~2p c! m*e/Ne
S D
3 P
n
Slices with an In-Line Four-Point Probe P
where:
3. Terminology
−1
n = wavenumber value of plasma resonance, m ,
P
3.1 Definitions of Terms Specific to This Standard:
l = wavelength of plasma resonance, m,
P
3.1.1 plasma resonance—plasma resonance is the transition
c = speed of light, m/s,
between low-frequency and high-frequency reflectivity condi-
m* = effective mass of free carriers, kg,
tions in an extrinsic semiconductor in which the mobile carriers
e = low frequency dielectric constant, F/m,
−3
and the immobile ionized impurities can be considered as a
N = density of free carriers, m , as measured by this
plasma. At high frequencies or short wavelengths, the fields are
method, and
varying too fast for the plasma to respond and the reflectivity
e = electronic charge, C.
approaches that of intrinsic silicon. At low frequencies or long
wavelengths, the fields vary slowly so that the plasma may
6. Interferences
respond and the reflectivity is high.
6.1 Highly polished surfaces can be obtained over highly
3.1.2 plasma resonance minimum—this is the minimum in
damaged regions; this can cause misleading results. Users of
reflectivity as a function of wavenumber associated with
this method should, therefore, assure themselves that the
plasma resonance.
polishing technique that they are using yields damage-free
surfaces in order to be certain that the measurement obtained
4. Summary of Test Method
using this technique represents the bulk properties of the
4.1 The majority carrier concentration of a semiconductor is
materials.
measured by determining the reflectivity of the semiconductor
as a function of wavenumber in the infrared region of the
NOTE 1—Pending establishment of a method of test for surface damage,
spectrum. This reflectivity has a high-wavenumber value the user may check for surface damage by successively measuring the
reflectivity spectrum of the specimen and chemically removing a layer
which is almost independent of carrier concentration and is
from the surface using a non-preferential etchant until the plasma
characteristic of the intrinsic semiconductor lattice. As the
resonance readings from successive measurements agree.
wavenumber decreases, the reflectivity decreases to a mini-
mum at a wavenumber, n which is characteristic of the
6.2 Since specimens may be inhomogeneous, users of this
PR
number of free carriers in the semiconductor. For decreasing
technique for a referee method should agree on the location on
wavenumbers, beyond n , the reflectivity increases rapidly,
each specimen where the measurement is to be taken.
PR
approaching a value of unity.
6.3 Anomalously low mobility material may yield a mea-
4.2 Carrier concentration in a specimen of unknown doping
surement of concentration which does not agree with that
level but known type and composition, for example, n-type
obtained by other techniques.
silicon, can be determined by measuring n and relating this
PR 6.4 GaAs has a sharp reflection minimum due to lattice
−1
to the carrier concentration through an appropriate empirical
reflection (Reststrahlen Band) centered at 302 cm . Therefore
relationship.
measurements of p and n-type GaAs near the lower limit of this
test method should be interpreted carefully so that the Rest-
2 strahlen minium is not mistaken for a plasma resonance
Annual Book of ASTM Standards, Vol 03.06.
Annual Book of ASTM Standards, Vol 10.05. minimum.
NOTICE: This standard has either been superceded and replaced by a new version or discontinued.
Contact ASTM International (www.astm.org) for the latest information.
F 398 – 92 (2002)
7. Apparatus 8.2 The polishing technique that is used to prepare the
surface of the test specimen must minimize subsurface damage
7.1 Spectrophotometer, Dual-Beam, with either wavelength
(Note 1).
or a wavenumber presentation, under continuous dry-nitrogen
or dry-air purge, and with the following additional character-
9. Calibration
istics:
−1 −1
9.1 Spectrophotometer:
7.1.1 Wavenumber range from 5000 cm to 25 cm .
9.1.1 Refer to the instruction manual for the details of
Wavelength range from 2 to 40 μm. If the range of values is
operation of the instrument to be used.
smaller, the range of impurity concentrations covered by the
9.1.2 Determine that the wavelength accuracy and repeat-
method is reduced from that stated in 1.2.
ability are within the specifications for the instrument in
7.1.2 Provision for up to a ten-times scale expansion of the
accordance with Practice E 275.
reflectance measurement can be necessary for some materials.
7.1.3 Wavelength reproducibility of at least 60.05 μm as
NOTE 3—It is advised that a service representative of the manufacturer
determined according to Practice E 275.
be contacted if the accuracy and repeatability determined above are
−1
7.1.4 Spectral resolution of 2 cm or smaller at 1000 outside the manufacturer’s specifications.
−1
cm .
9.2 Reflectance Accessory:
7.2 Reflectance Accessory, compatible with the spectropho-
9.2.1 With the accessories to be used for reflectance mea-
tometer. This accessory must provide an angle of incidence
surements installed in each beam and with a front surface
upon the specimen of 30° or less. Additionally, there shall be
mirror on each reflectance attachment used, displace the scale
either an identical or symmetric accessory for the reference
towards zero by adjusting the zero control or some other
beam providing equal path lengths in both beams, or an
control which does not change the 0 to 100% span. Run a
attenuator for the reference beam. Whether a reflectance
double-beam 100% reflectance trace (Fig. 1).
attachment or an attenuator is used in the reference beam, the
9.2.2 The condition of the instrument is not acceptable for
combination of specimen reflectance attachment and reference
this method if this trace varies by more than 8 percentage
beam attachment must provide a 100% trace varying from a
points, peak to peak.
straight line by not more than 8% of a full scale (see 9.2).
9.2.3 If the trace varies by more than this amount run a
100% trace in double-beam transmission operation.
NOTE 2—While good optical practice would require the use of reflec-
9.2.4 If this trace varies by more than an acceptable value as
tance accessories in each beam so designed and installed as to maintain the
identical path length nature of a dual beam spectrophotometer, the data
stated by the manufacturer of the spectrometer, the spectrom-
from the round robin indicate that the use of an attenuator in the reference
eter is not operating properly and must be readjusted.
beam does not deteriorate the interlaboratory precision of the measure-
9.2.5 If this trace is within the manufacturer’s specification
ment. Therefore the use of a matched pair of reflectance accessories is not
the reflectance attachments are the cause of the error and either
required but is recommended.
they should be carefully aligned or the manufacturer contacted.
7.3 Aluminum Mirror, optical quality, scratch-free, front-
surface, for each reflectance attachment to be used.
10. Procedure
10.1 If the conductivity type of the specimen is unknown
8. Test Specimens
determine it at a convenient time before interpretation of the
8.1 The surface of the test specimen must be smooth and measurement, according to Test Methods F 76 or F 42.
highly reflecting, giving the appearance of a mirror, as is 10.2 Measure the reflectance of the specimen as a function
normally used for semiconductor processing. of wavenumber (Fig. 2, Trace 1). Scan speed must be slow
FIG. 1 Displaced 100% Double Beam Reflectance Curve with Front Surface Aluminum Mirrors on the Reflectance
Attachments in Each Beam
NOTICE: This standard has either been superceded and replaced by a new version or discontinued.
Contact ASTM International (www.astm.org) for the latest information.
F 398 – 92 (2002)
−1
NOTE 1—The instrument on which this spectrum was obtained provided a bilinear presentation in wavenumber, with a scale change
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.