semi合集-English.pdf - 第4109页

SEMI F60-0301 © SEMI 2001 5 analyst subjectivity associated with the curve fit procedure. If it is necessary to dev ia te from the curv e fit parameters o utlined above in order to obtain a satisfactory curve fit it shou…

SEMI F60-0301 © SEMI 2001 4

standard. A minimum sputtering etch cleaning of the

SPC sample of ≥ 10 nanometers is recommended in

order to remove surface contamination, the oxide

passive layer and establish equilibrium. Note that the

degree of preferential sputtering and other ion etching

artifacts are a function of the matrix material and the

ion gun conditions.

13.4 Calibration frequency of the application specific

SPC should be designed to establish a record of the

reproducibility and accuracy of the measurement, as

reflected by the nominal composition of the 316L

standard. A minimum frequency of one per week is

recommended, or immediately prior to performing a

measurement in accordance with this test method.

13.5 A sputter rate calibration shall be performed prior

to an ESCA profile measurement conducted in

accordance to this test method. The ion sputter rate

determination shall be made using standard thin films

of 100 nanometers or less of SiO

on Si. The method of

determining the thickness of the oxide film shall be

based upon the oxygen concentration profile. The

oxide – substrate interface shall be specified as the

point at which the oxygen concentration decreases to ½

its maximum value in the SiO

film, ignoring the first

sputter cycle.

13.6 Multiple samples may be analyzed following a

single sputter rate determination if the ion gun

parameters have not been adjusted and the ion gun

performance is documented as stable over the period in

question.

14 Procedure

14.1 As-Received Surface Analysis— The sample is to

be mounted in accordance with manufacturer's

recommendations and in a manner consistent with ultra-

high vacuum surface analytical procedures. Some of

these practices are detailed in ASTM-E1078. The area

to be analyzed should be mounted parallel to the sample

holder surface so that TOA is well known.

14.1.1 Place the sample in the ESCA introduction

chamber for pump down. Transfer to the analytical

chamber at the manufacturer's recommended base

pressure. Align the sample with respect to the X-ray

beam and analyzer so that optimum count rate from the

desired analytical location is obtained. The use of a

collimated high voltage electron beam to align the

sample should be avoided, as this may pyrolyze surface

carbon and potentially alter surface oxide chemistry.

The surface area to be analyzed should be free of

visible particles and large defect features, if possible.

14.1.2 A beam size as close to 1 mm as possible

should be used to ensure measurement of a

representative surface. If surface curvature is great

(e.g., < 1/4" tubing), a smaller beam may be employed.

Elemental survey data (0–1100 eV) are to be measured

from the sample surface to determine the elements

present and their approximate surface abundance. A

high throughput analyzer setting may be used to obtain

a signal to noise ratio (S/N) sufficient to detect common

surface contaminants such as sulfur and phosphorus at

one atomic percent or better.

14.1.3 Using instrument settings sufficient to provide

a FWHM peak width of 1.0 eV or less on Au (4f

7/2

measure chromium, iron, and carbon regions. A typical

region width is 20 eV. Suggested ranges are as follows:

Cr (2p

3/2

) from 570 to 590 eV, Fe (2p

3/2

) from 700 to

720 eV, and C (1s) from 275 to 295 eV. Signal to noise

ratios of greater than 20 are suggested.

14.1.4 A consistent method of data reduction of high-

resolution surface spectra is necessary in order to

provide meaningful comparison of the relative Cr:Fe

ratios.

14.1.5 The spectra shall be charge compensated with

respect to the maximum of the Cr (2p

3/2

) set to 577 eV.

An iterated Shirley method of background subtraction

shall be applied to the Fe (2p

3/2

) and Cr (2p

3/2

) spectral

regions, using a minimum of three iterations.

14.1.6 The curve fit regions shall typically extend

from:

Cr (2p

3/2

) 570–582 eV

Fe (2p

3/2

) 704–717 eV

14.1.7 The curve fit peak parameters should be

initialized as follows:

Cr (2p

3/2

) Peak Position FWHM % gaussian

1 574 1.5 80

2 576.5 2.0 80

3 577.5 2.0 80

Fe (2p

3/2

) Peak Position FWHM % gaussian

1 707 1.0 80

2 708 1.0 80

3 710 1.4 80

4 711.5 1.4 80

5 713 1.4 80

14.1.8 The curve fit routine should apply the following

tolerances for the band limit parameters of all peaks

defined above:

FWHM delta

≤ 0.2 eV

% gaussian delta

≤ 10%

Curve fit position delta

≤ 0.25 eV

14.1.9 An automated form of data reduction is

recommended to enhance reproducibility and minimize

SEMI F60-0301 © SEMI 20015

analyst subjectivity associated with the curve fit

procedure. If it is necessary to deviate from the curve

fit parameters outlined above in order to obtain a

satisfactory curve fit it should be noted in the formal

report. Possible sources for the anomalous behavior

include an excessive concentration of iron oxide or a

sufficiently aged sample surface, such that the oxide

surface chemistry and structure has modified.

14.2 Depth Profile Analysis— The ESCA depth

profile may be performed following acquisition of the

initial as-received surface survey scan and high-

resolution spectra using the same sample x-y position

and take-off angle of 35°.

14.2.1 The depth profile may be acquired in either

continuous mode (simultaneous sputter etching and data

acquisition) or alternating sputter etch/data acquisition

mode. The following acquisition windows are

recommended:

Spectral line Lower limit Range

C 280 eV 15 eV

O 525 eV 15 eV

Cr 570 eV 15 eV

Ni 848 eV 15 eV

Fe 700 eV 25 eV

14.2.2 If only an elemental compositional profile is

acquired, any pass energy may be selected as long as

the remaining acquisition parameters defined by the

number of sweeps, step size and dwell time result in ≤

1.0at% detection of the matrix elements followed in the

depth profile.

14.2.3 The composition profile should extend far

enough into the depth of the sample to reach the base

metal composition of the 316L material, but a minimum

of 100Å from the surface as referenced to the calibrated

sputter rate in SiO

15 Calculations and Interpret ation of Results

15.1 Most manufacturers supply software for

determination of elemental composition. The elemental

composition should be calculated using sensitivity

factors appropriate to the instrument, each element, and

resolution settings for each measurement.

15.2 The total Cr/Fe ratio is calculated by adding peak

areas from all species of each element, adjusting for

different numbers of scans and sensitivity factors, and

dividing the Cr result by the Fe result. The Cr oxide/

Fe oxide ratio is calculated in a similar manner, except

that only peaks 2 and 3 for Cr and only peaks 3 through

5 for Fe (oxide species) are used. The formulas are as

follows:

Total Cr/Fe = (Σ Cr peak areas)/# Cr scans/Cr sensitivity factor

(Σ Fe peak areas)/# Fe scans/Fe sensitivity factor

Cr Oxide/ = (Σ

Cr oxide peak areas)/# Cr scans/Cr sens. factor

Fe Oxide (Σ Fe oxide peak areas)/# Fe scans/Fe sens. factor

16 Reporting Results

16.1 As-Received Surface Results — A tabular

summary of the elemental composition of all elements

detected in the surface survey spectrum is to be

supplied with the associated elemental survey spectrum.

If sample preparation has been done by other than the

component manufacturer all tables of as-received

surface analysis results shall include a note stating:

“Sample preparation was not performed by the

component manufacturer. Results are not to be

interpreted as indicative of the component

manufacturer’s quality of cleaning and packaging

procedures.”

16.1.1 Tabular summaries of total Cr/Fe ratio and Cr

oxide/ Fe oxide are also to be supplied, with associated

Cr(2p

3/2

), Fe(2p

3/2

), and C(1s) narrow region spectra.

16.1.2 Acquisition parameters, including

manufacturer, model and TOA; X-ray source (Al or

Mg); beam size; and other pertinent settings are to be

supplied. Analyst identity and analysis date are also

required information. Each table and graph must be

clearly labeled with sample identification.

16.2 Depth Profile Results — Compositional depth

profile data plots shall display the atomic concentration

of Fe, Cr, Ni, C, and O at minimum versus the sputter

time or the equivalent depth from the calibrated sputter

rate in SiO

16.2.1 The method applied to determine the oxide

thickness shall be specified in the report. A commonly

accepted method of quantifying the oxide thickness is

determined as the sputter time/depth at which the

oxygen concentration decreases to ½ the maximum

value. The carbon thickness may be determined

similarly.

APPENDIX 1

DISCUSSION OF EFFECTS OF DEPTH OF ANALYSIS ON DEPTH

PROFILE ANALYSIS BY ELECTRON SPECTROSCOPY FOR

CHEMICAL ANALYSIS OF PASSIVATED STAINLESS STEEL

NOTE: This appendix is being balloted as an official part of SEMI F## by full letter ballot procedure, but the recommendation in

this appendix are optional and are not required to conform to this standard. This appendix is derived from presentations made to

the SEMI Surface Analysis Task Force.

A1-1.1 The purpose of this appendix is to describe the

ESCA (sometimes referred to as X-Ray Photoelectron

Spectroscopy (XPS)) technique and explain the

interpretation of the depth profile analysis with respect

to the structure and composition of the passive oxide

layer on stainless steel. The effect of the depth of

analysis on the measured chromium to iron ratio, the

oxide thickness, and the depth of enrichment is

discussed. The discussion applies to depth profile

analyses performed by Auger electron spectroscopy,

also.

A1-1.2 ESCA — The ESCA technique makes use of an

X-ray beam for a probe. The surface of interest is

illuminated with X-rays, typically Al kα radiation,

which has a well-defined energy of 1486.6 eV. When

an x-ray photon and an electron interact, the electron

adsorbs all of the energy of the photon, 1486.5 eV; it

cannot react with it partially. The electron then has

sufficient energy to leave the atom and move into the

matrix. The electron’s energy in the matrix is 1486.6

eV minus its binding energy to the atom from which it

escaped. Its binding energy is indicative of the element

or compound from which it came and the orbital from

which it was removed. Such an electron is termed a

“photoelectron”.

A1-1.3 In a typical ESCA instrument, a

monochromatic beam of x-rays is focused at a point of

interest on a surface. A cloud of electrons is generated

at the surface by the x-rays. This electron cloud is

made up of secondary electrons, Auger electrons and

photoelectrons. (NOTE: The generation of Auger

electrons is a byproduct of the process. If the

photoelectron was an inner shell electron, when it

leaves the atom it creates an ion in an excited state.

Once the ion is created it can de-excite via the Auger

process. The presence of Auger electrons in an ESCA

survey can be used to determine what elements were

present.) The mechanisms of ESCA photoelectron and

Auger electron generation are illustrated in Figure A1-

Figure A1-1

Schematic representation of the ESCA

photoelectron and Auger electron generation

processes.

A1-1.4 An electron lens is focused at the same point as

the x-ray beam and a portion of the electron cloud

enters the lens. From the lens, the electrons enter an

electron analyzer that measures the number of electrons

as a function of their kinetic energy. The kinetic energy

is subtracted from the incident x-ray photon energy, and

the resulting binding energy is then plotted as the x-axis

with the number of electrons at each energy on the y-

axis. An example is shown in Figure A1-2. The

binding energy can be indicative not only of the

element from which the electron came but also the

chemical bonding state of the element.