semi合集-English.pdf - 第5254页
SEMI M41-1101 © SE MI2000, 2001 11 Parameters (Units) V alue AS TM Test Method or Meas ureme nt Procedure Acceptanc e To-be -bonded Surf ace Cleanliness: Metals (/c m 2 ) Note C AAS, ICP-MS, TXRF (F15 26-95 ), SIMS (F16 …
SEMI M41-1101 © SEMI 2000, 2001 10
Parameters (Units) Value ASTM Test Method or
Measurement Procedure
Acceptance
Resistivity (ohm-cm) Note C F43-93, F84-93,
F1527-94
Note C, or Certified by
Wafer Manufactures
Conductivity Type Note C F42-93 Note C, or Certified by
Wafer Manufactures
Oxygen Concentration (/cm
3
) Note C F1188-93a,
F1619-95
Note C, or Certified by
Wafer Manufactures
Carbon Concentration (/cm
3
) Note C F1391-93 Note C, or Certified by
Wafer Manufactures
Surface Cleanliness: Metal
Contamination (/cm
2
)
Note A, C AAS, ICP-MS,
TXRF (F1526-95),
SIMS (F1617-98)
Note A, C
Surface Cleanliness: Particle
Density (/wafer)
Note A, C Light Scattering
Tomography (F1620-96)
(SEMI M34)
Note A, C
Surface Roughness (nm) Note A, C AFM
(SEMI M34)
Note A, C
Carrier Lifetime (µsec) Note C µ-PCD Method
(F1535-94)
Note C
Crystalline Alignment of Top
Silicon Film to Base Wafer (
o
)
Note C X-ray Diffraction
(F847-94)
Note C
Surface Feature
(Haze, Scratch, etc)
None F154-94, F523-93, F1726-
97
Must be measured on each
wafer
OSF Density (/cm
2
) Note C Optical Metrology
(F1727-97)
Note C, or Certified by
Wafer Manufactures
Buried Oxide (BOX)
Thickness (µm) 3–5 Ellipsometry (F576-95)
or
Reflective Spectroscopy
(SEMI M34)
Tolerance is ± 5%;
Note C, or Certified by
Wafer Manufactures
Location of Bonded Interface Inside Oxide
(Lower Surface)
TEM Certified by Wafer
Manufactures
Void Density (/cm
2
) None Scanning Acoustic
Tomography, Optical Defect
Inspection
Note C
Oxide Defect Density (/cm
2
) Note C I-V on Capacitor,
Cu Decoration
(SEMI M34)
Note C
Dielectric Breakdown Voltage
(V)
Note C I-V on Capacitor Note C
Interface States (/cm
2
) Note C, H C–V Technique Note C
Fixed Charge Density (/cm
2
) Note C, H C–V Technique (F1153-92) Note C
Bonding Strength (kg/cm
2
) Note C Tensile Strength Note C, or Certified by
Wafer Manufactures
Base Silicon Wafer
Surface Orientation Note C F26-87a Note C
Resistivity (ohm-cm) Note C F43-93, F84-93, F1527-94 Note C
Conductivity Type Note C F42-93 Note C
Fiducial Axis Orientation
(Flat/Notch)
Note C F671-90, F1152-93 Note C
SEMI M41-1101 © SEMI2000, 200111
Parameters (Units) Value ASTM Test Method or
Measurement Procedure
Acceptance
To-be-bonded Surface
Cleanliness: Metals (/cm
2
)
Note C AAS, ICP-MS,
TXRF (F1526-95),
SIMS (F1617-98)
Note C
Back Surface Finish Note C Optical Metrology Note C
Note A: Same as the standard of regular silicon wafer
Note B: The value is of 150 mm wafers, and is determined according to wafer diameter.
Note C: To be determined by negotiation between wafer users and suppliers
Note D: Reflective spectroscopy or FT-IR is recommended for top silicon film of less than several µm (about 7 µm), and FT-IR for top silicon
film of more than several µm (about 7 µm).
Note E: Tolerance of ± 0.5 µm is recommended for top silicon film of less than several µm (about 7 µm), and ± 1.0 µm for top silicon film of
more than several µm (about 7 µm).
Note F: The value is without the compensation method by backside oxide.
Note G: The value is with the compensation method by backside oxide.
Note H: This item can be neglected if the bonding interface is between BOX and base wafer.
7 Sampling Plan
7.1 Unless otherwise specified, ASTM Practice E 122
shall be used. When so specified, appropriate sample
sizes shall be selected from each lot in accordance with
ANSI/ASQC Z1.4. Each quality characteristic shall be
assigned an acceptable quality level (AQL) of lot
tolerance percent defective (LTPD) value in accordance
with ANSI/ASQC Z1.4 definitions for critical, major,
and minor classifications. If desired and so specified in
the contact or order, each of these classifications may
alternatively be assigned cumulative AQL or LTPD
values. Inspection levels shall be agreed upon between
the users and the suppliers.
8 Test Methods - Dimensions
NOTE 3: Detailed test procedures of each item should be
determined between the users and the suppliers.
8.1 Thickness of Top Silicon Film — The following
two methods are available for thickness measurement.
8.1.1 Reflective Spectroscopy — The light of visual
wavelength (400–800 nm) is introduced into top silicon
film by varying its wavelength continuously, and then
the reflective spectra is measured. When the light is
introduced into multi-layers of SOI wafers, reflection
occurs on the surface of top silicon film and the front
and backside of BOX. In such a case, the phase varies.
The final intensities of the light that reflects from top
silicon film surface are the sum of the intensity of light
that reflects from each layer. The thickness of top
silicon film and BOX makes optical path difference and
then results in phase difference that is dependent on its
wave length. The reflective light intensities, depending
on its wavelength, are measured. The reflective spectra
are defined as the ratio of reflective light intensity to
incident intensity. This spectra curve varies by the
thickness of top silicon film and BOX. The top silicon
film thickness is derived from the obtained spectra
curve by approximate calculation based on simulation
or by comparing with the database.
8.1.1.1 Reference: J.-P. Colinge, “Silicon-On-Insulator
Technology”, Kluwer Academic Publisher, 1991.
NOTE 4: Thickness of top silicon film and BOX layer are
limited to measure because of using visual light.
Example: Nanospec/AFT model : 210LCW, SP-FSC15
Top silicon film thickness: 0.01–15 µm
BOX thickness: 0.004–3 µm
NOTE 5: Optical constant is already known in each of
multilayers, and it should be constant in the whole layer.
8.1.2 FT-IR (Fourier Transform Infra-Red Spectro-
metry) — The reflectance spectrum of the specimen,
which exhibits successive maxima and minima
characteristics of optical interference phenomena, is
measured as a function of wavelength using an infrared
spectrophotometer. These maxima and minima are
observed when the optical path lengths of the infrared
beam, reflected from both the top silicon film surface
and the top silicon film–buried oxide interface, differ
by an integral number of half wavelengths.
Consequently, the thickness of top silicon film is
calculated using the wavelength of the extreme
maximum and minimum in reflectance spectrum, the
refractive index of Silicon and Silicon dioxide, and the
angle of incidence of the infrared beam upon the SOI
wafer.
Reference: F95 – Standard Test Method for Thickness
of Lightly Doped Silicon Epitaxial Layers on Heavily
Doped Silicon Substrates Using an Infrared Dispersive
Spectrophotometer
8.1.3 Definition of top silicon film thickness tolerance
— Thickness tolerance is defined below.
8.1.3.1 After top silicon film thickness is measured at
predetermined number of points within an SOI wafer,
the maximum and the minimum values are chosen, and
then the tolerance is defined as;
SEMI M41-1101 © SEMI 2000, 2001 12
Tolerance = Maximum value – Minimum value
NOTE 6: The location and numbers of measuring points
should be determined between users and suppliers.
8.1.3.2 In case of multi-points measurements (ex. a
few hundreds) within an SOI wafer, the tolerance is
defined as;
Tolerance = 3σ (3 times of the standard deviation)
8.1.3.3 Measurement exclusion area such as wafer
edge should be determined between users and suppliers.
NOTE 7: Recommendable metrology
a) Reflective spectroscopy or FT-IR is recommended
• for top silicon film of less than several µm (about 7 µm),
• and FT-IR for top silicon film of more than several µm
(about 7 µm)
b) Tolerance is defined as the difference between the
maximum and the minimum value after measuring several
(ex. 9) points.
c) It is not necessary to measure the BOX thickness of SOI
wafer after bonding. It is OK to measure it before wafer
bonding.
d) In case of the above 1) ~ 3), the number of measurement
points and their location should be specified in case of several
points measuring, and the measurement exclusion area should
be specified in case of multi-points measuring.
8.2 Crystal Defect of Top Silicon Film
8.2.1 OSF (Oxidation induced Stacking Fault) — This
technique is applicable to the top silicon film of thicker
than 1.5 µm. OSF density is measured by preferential
chemical etching and microscopic observation.
Preparation of samples and measurement of OSF
density are as follows:
8.2.1.1 Sample Preparation — SOI wafers are oxidized
at 1,100°C, 1 h, in H
2
/ O ambient after the SC-1 and
SC-2 cleaning. Oxide is removed by ca. 25 % HF and
then the SOI wafers are preferentially etched by 1 µm,
applying JIS H 0609:1994(B), and then rinsed
thoroughly in distilled water and blown dry. JIS H
0609 defines the chromium-free preferential solution,
which is composed of HF, HNO
3
, CH
3
COOH and H
2
O.
8.2.1.2 Measurement of OSF Density — Samples are
examined by an optical microscope. The sample
surface is observed by magnification of 200 X, and
OSF is counted on SOI wafer within the scope along
the two lines, which are parallel and perpendicular to
the orientation flat (so called cross scanning). OSF
density is calculated from the count number and
scanning area.
8.3 Buried Oxide Defect
8.3.1 Cu Decoration Method — In case of Bonded
SOI, the buried oxide is usually formed by thermal
oxidation. Therefore, the defect of buried oxide is
taken into consideration only for the thin oxide cases.
Buried oxide defect such as pinholes can be evaluated
by Cu decoration method. This method has been
applied to the buried oxide film of less than 400 nm
thickness. Sample preparation and Cu decoration are
conducted by the following procedure. The top silicon
film on the buried oxide is removed by KOH solution,
and then cleaned and rinsed. The sample is set on a
gold-plated brass (Cathode) in the methanol solution.
On the other side, a copper plate (Anode) is placed 5
mm above the sample surface. Positive constant bias of
1–3 MV/cm (ex. 40–120 V for 400 nm oxide) is applied
to the copper plate for 5 minutes. Small leakage current
passes through the buried oxide defect, and
consequently copper precipitates on the defects.
Typical allowable defect density is < 0.1/cm
2
.
8.4 Metal Contamination
8.4.1 The surface metal contamination can be
measured by TXRF, AAS and ICP-MS methods.
8.4.2 TXRF (Total X-Ray Fluorescence) — Total X-ray
Fluorescence uses a low angle incident, and a tightly
collimated X-ray beam excites the characteristic X-rays
from impurity atoms near the sample surface. Usually,
the angle of X-ray incident is less than 0.1 degree. The
element identification and the amount of the element
can be obtained by measuring energy and intensities of
fluorescence X-ray. The instrument provides a map of
impurity element distribution. The surface metal
contamination (typically from Na to Zn) shall be less
than 10
11
cm
-2
in total.
NOTE 8: This TXRF method is conveniently used to detect
the metals on the SOI wafer surface.
8.4.3 AAS (Atomic Absorption Spectrophotometry) —
The elemental characteristic absorption of the atom is
measured by introducing sample solution as aerosol
into the flame and then spectral absorption through the
flame from the light source is detected by the
spectroscope. The flameless method, superior to the
flame method in the sensitivity, is now broadly used.
8.4.3.1 Sample Preparation — Careful sample
preparation is necessary for the precise measurement.
SOI wafer surface is exposed to HF vapor, and the
metals on the surface are collected as droplet. To
improve the sensitivity, the volume of collective
solution should be as tiny as possible and the HF drops
are rolled all over the surface in collective operation. In
case of precious metals, it is better to use other kinds of
collective solutions instead, since they are not dissolved
or collected by HF solution itself.