semi合集-English.pdf - 第4871页

SEMI M1-0305 © SEMI 1978, 2005 47 R2-7.2 Flat Length sh ould be det ermined by SEMI MF6 71. I f flat di ameter is specified instead of flat lengt h, it can be determined by ¶6.2.1 of DIN 5 0441/4 or by a dial gauge met h…

SEMI M1-0305 © SEMI 1978, 2005 46

are made at room temperature and recombination lifetime if they are made at elevated temperature (50 to 75C).

The test specimens required for this test method can be made by procedures compatible with typical wafer

processing.

R2-5 Chemical Characteristics

R2-5.1 Oxygen Concentration in relatively high resistivity specimens (



> 1 to 3 ·cm at room temperature) should

be measured by infrared techniques. SEMI MF1188, JEITA EM-3504, and DIN 50438 all utilize approximately

normal incidence illumination and the IOC-88 calibration factor to determine the interstitial oxygen content. See

SEMI M44 for a discussion of other calibration factors and their relationship to IOC-88. SEMI MF1619 uses p-

polarized radiation, incident at the Brewster angle to reduce the multiple reflections from polished wafers. For more

heavily doped wafers, secondary ion mass spectrometry (SIMS, SEMI MF1366) and gas fusion analysis (GFA) can

be used to determine total oxygen content. There is no standardized method for performing GFA, and the

repeatability of this technique is generally poorer than SIMS.

R2-5.2 Radial Oxygen Variation should be determined by measuring the oxygen content by one of the above

methods at selected locations on the wafer defined in SEMI MF951. Usually measurements are made at the center

and at a single point on the primary flat or notch bisector 10 mm from the edge of the wafer opposite the fiducial;

this is known as test plan A-1.

R2-5.3 Carbon Concentration should be measured by SEMI MF1391, JEITA 56, and DIN 50438/2. The most

modern calibration coefficients are used in SEMI MF1391 and JEITA 56. For these methods, the test specimen

cannot be too heavily doped, and special thick test specimens may be necessary.

R2-5.4 Boron Contamination in heavily doped n-type silicon can be determined by SEMI MF1528.

R2-6 Structural Characteristics

R2-6.1 Dislocation Etch Pit Density, Slip, Lineage, Twins, and Swirl are usually displayed by etching and visual or

microscopic observation. JIS H 0609 is a comprehensive test method for carrying out this procedure with the use of

non-chromic etchants. SEMI MF1809 also recommends non-chromic etchants for this test. For the full procedure,

SEMI MF1809 must be used with other standards including SEMI MF1726 and SEMI MF1810. DIN 50434 is an

older, but comprehensive test procedure to observe these and other structural defects based on chrome-containing

etchants. It is also possible to observe these defects by x-ray topography using DIN 50443/1.

R2-6.2 Shallow Pits can be exposed by the relatively low temperature heating cycle and procedures in SEMI

MF1049 if they are present in sufficient density, but testing for shallow pits in production environments is usually

carried out using the temperature cycle in SEMI MF1727 followed by etching with etchants discussed in SEMI

MF1809, examination by SEMI MF1726, and counting by SEMI MF1810. They can also be observed by using JIS

H 0609.

R2-6.3 Oxidation Induced Stacking Faults (OISF) can be observed by using JIS H 0609, or by using the

temperature cycle in SEMI MF1727 followed by etching with etchants discussed in SEMI MF1809, examination by

SEMI MF1726, and counting by SEMI MF1810. OISF can also be observed by x-ray topography using DIN

50443/1 following thermal treatment according to JIS H 0609 or SEMI MF1727.

R2-6.4 Oxide Precipitates (Bulk Micro Defects, BMD) are generated with the use of a temperature cycle such as

those in SEMI MF1239. The amount of precipitation can be measured by the oxygen reduction method of SEMI

MF1239 or by direct observation by infrared that has not yet been standardized.

R2-6.5 Bulk Defects can be detected by x-ray topography in accordance with DIN 50443/1.

R2.7 Dimensional Characteristics

R2-7.1 Diameter is presently very well controlled in silicon wafers that have been edge profiled with cam follower

equipment. For three-point measurements at locations defined by SEMI MF2074 and, except for positions on 150

mm diameter, n-type (100) wafers, by DIN 50441/4. The latter utilizes a measuring or projection microscope to

make the diameter measurements, but the former allows a dial gauge or any other equally accurate method of

dimensional measurement.

SEMI M1-0305 © SEMI 1978, 2005 47

R2-7.2 Flat Length should be determined by SEMI MF671. If flat diameter is specified instead of flat length, it can

be determined by ¶6.2.1 of DIN 50441/4 or by a dial gauge method as agreed upon between the supplier and

customer. Both SEMI MF671 and DIN 50441/4 take account of the possibility that the ends of the flat might be

rounded.

R2-7.3 Notch Dimensions should be determined by SEMI MF1152 (see Figure 5).

R2-7.4 Flat Orientation can be confirmed by SEMI MF847. There is no standardized method for determining the

crystallographic orientation of the diameter that bisects the notch.

R2-7.5 Edge Profile is usually determined by using one of the two templates (see Figure 6) with Method B of SEMI

MF928 or Procedure 2 of DIN 50441/2, both of which are nondestructive. The other two procedures in these

methods are destructive and not so widely used.

R2-7.6 Thickness is usually determined at the center point of the wafer with the use of SEMI MF1530, an

automated technique. Manual techniques originally used in the industry include SEMI MF533, JIS H 0611, and

DIN 50441/1.

NOTE 2: Test methods SEMI MF1530 (for thickness, thickness variation, and flatness), SEMI MF1451 (for sori), and SEMI

MF1390 (for warp) may not be suitable for use on large diameter wafers with polished back surfaces. New standardized test

methods for measuring these parameters on such wafers are under development.

R2-7.7 Total Thickness Variation (TTV) was originally determined with the use of manual 5-point techniques

covered in SEMI MF533, JIS H 0611, and DIN 50441/1. JIS H 0611 differs from SEMI MF533 and DIN 50441/1,

in that the measurements in JIS H 0611 are taken at the center and at 5 mm from the edge on diameters parallel and

perpendicular to the primary flat or notch bisector, while the measurements in SEMI MF533 and DIN 50441/1 are

taken at the center and at 6 mm from the edge on diameters 30 degrees and 120 degrees counterclockwise from the

bisector to the primary flat or notch (with the wafer facing front surface up). TTV can also be determined with the

use of SEMI MF657, which involves a continuous scan pattern over a portion of the wafer surface. Currently,

however, it is most frequently determined using SEMI MF1530, which involves an automated continuous scan

pattern over the entire wafer surface. In this case, the quantity determined is equal to the global flatness GBIR (see

Appendix 1 and Note 1, above).

R2-7.8 Surface Orientation can be determined by the x-ray methods of SEMI MF26, JEITA EM-3501, or DIN

50433/1, the optical method of SEMI MF26 or DIN 50433/2, or the Laue method of DIN 50433/3.

R2-7.9 For off-orientation {111} wafers, the orthogonal misorientation is specified for each of the wafer categories

in Tables 4 through 9. There is no standardized measurement method for this property so it should be determined by

a method agreed upon between supplier and customer.

R2-7.10 Bow can be determined with the manual methods SEMI MF534 and JIS H 0611. Currently, bow is not as

widely used a parameter as warp.

R2-7.11 Warp is currently most often measured with the use of SEMI MF1390, which is an automated method with

full surface scan and correction for gravitational sag. It can also be measured with the use of the contactless manual

method SEMI MF657, in which the prescribed scan pattern covers only a portion of the wafer surface in which there

is no correction for gravitational sag. As noted in Appendix 2, different reference planes are used for the two

methods. Because SEMI MF657 employs a back surface reference plane, the measured warp may include

contributions from thickness variation of the wafer. SEMI MF1390 employs a median surface reference plane and

is not susceptible to interferences from thickness variations. In general, the latter is preferred, especially for wafers

150 mm in diameter and larger (see Note 1, above).

R2-7.12 Sori , which is sometimes specified in lieu of bow or warp or both, can be determined by SEMI MF1451

(see Note 1, above).

R2-7.13 Global Flatness can be determined by either capacitance measurements, as in SEMI MF1530, or by

multiple beam interference, as in DIN 50441/3. Generally, the former is considered to be more reliable, especially

as the need increases for measuring smaller flatness deviation (see Note 1, above). As noted in ¶R2-7.7, GBIR is the

same as TTV; other global flatness parameters are discussed in Appendix 1.

SEMI M1-0305 © SEMI 1978, 2005 48

R2-7.14 Site Flatness is generally determined by SEMI MF1530 (see Note 1, above). The most commonly used

site flatness parameter is SFQR; other site flatness parameters are discussed in Appendix 1. Although it is not

widely used, the scanning site flatness parameter SFSR was recently introduced. For this parameter, use a subsite

width, Wss, equal to 8 mm and orient the wafer so the effective scan direction is along the wafer’s y-axis as defined

in SEMI M20.

R2-7.14.1 SEMI M49 is a guide for specifying test equipment for use in determining thickness, shape, and flatness

parameters on wafers intended to be used to fabricate advanced integrated circuits.

R2-7.15 Nanotopography should be determined by SEMI M43. This guide gives a variety of options that can be

used, so it is essential to specify the various conditions that are desired in any given case. The conditions chosen

should be agreed upon between supplier and customer.

R2-8 Front Surface Chemistry

R2-8.1 Surface Metal Contaminants

R2-8.1.1 Sodium, aluminum, potassium, and iron can be measured by secondary ion mass spectrometry (SIMS),

inductively coupled plasma mass spectrometry (ICP/MS), or atomic absorption spectroscopy (AAS). SIMS has

been standardized as SEMI MF1617. The latter two methods are frequently combined with vapor phase

decomposition (VPD), but they have not yet been standardized.

R2-8.1.2 Potassium, chromium, iron, nickel, copper, and zinc can be measured by Total Reflection X-Ray

Fluorescence Spectroscopy (TXRF), ICP/MS, and AAS. TXRF has been standardized both with (SEMI M33 and

ISO 17706) and without (ISO 14706) use of VPD to preconcentrate the surface metal contaminants.

R2-8.1.2.1 VPD is chemical preconcentration of the surface metals using vapor phase HF to decompose the surface

native oxide and a water (or acid-spiked water) droplet to scan across the wafer dissolving the surface metals. The

recovery rate of this preconcentration method is dependent upon the chemistry of the surface metals and upon the

chemistry used for the preconcentration. An alternative preconcentration method to VPD is to scan an acid droplet

across the wafer surface.

R2-8.1.2.2 VPD/AAS is a single-element technique which is widely used in Japan. It is element-specific and very

sensitive. VPD/ICP-MS is a rapid multi-element technique which is a more recent development. It is also very

sensitive, but its reproducibility is dependent upon the injection process into the ICP-MS. VPD/TXRF is an even

more recently developed multi-element technique. It is also very sensitive, but its reproducibility is dependent on

the residue-drying process.

R2-8.2 Surface Organics can be measured by SEMI MF1982. This standard describes two methods; the method to

be utilized should be agreed upon between supplier and customer.

R2-9 Surface Inspection Characteristics

R2-9.1 Visual Inspection of either the front or back surface of wafers can be carried out in accordance with SEMI

MF523 or JIS H 0614. The following conditions should be used for examination under high intensity illumination:

 Background light intensity: 8 ± 2 fc (86 ± 22 lux),

 Angle (alpha): 45 ± 10, and

 Angle (beta): 90 ± 10.

See ¶6.8 for a discussion of which artifacts on the surface should be considered as defects. SEMI MF154 is a useful

guide for identifying structures and contaminants seen on silicon surfaces.

R2-9.1.1 Scratches—In inspecting for scratches, it is important to note that while macro-scratches can be seen under

both high intensity and diffuse illumination, micro-scratches can be seen only under high intensity illumination.

Therefore, to separate the two kinds of scratches, it is necessary to count the scratches observed under both kinds of

illumination. The count of scratches seen under diffuse illumination is the number of macro-scratches while the

difference of the counts seen under high intensity illumination and diffuse illumination is the number of micro-

scratches. Of course, if the total requirement is for no scratches of either kind, then examination under high

intensity illumination only is adequate.