semi合集-English.pdf - 第4625页

SEMI C3-0699 © SEMI 19 86, 1999 6 The introduction of a gaseous sa mple into a carrier gas stream is accomplished by m ea ns of a sample injection valve. Valves f or this purpose can be any of several types, alt hough t …

SEMI C3-0699 © SEMI 1986, 19995

4.2.5.3 Inspect the sampler vessel for any obvious

physical defects, such as dents, gouges, bent fittings,

etc. Since the sampler is a pressurized vessel, it should

be removed from service if any damage is apparent.

4.2.6 Sampling Procedure

4.2.6.1 Remove cover.

4.2.6.2 Loosen inlet and outlet fittings dust caps

located on side of the vessel.

CAUTION: SAFETY GLASSES AND

PROTECTIVE GLOVES ARE REQUIRED WHEN

OPERATING THIS EQUIPMENT. SAMPLING

SHOULD BE DONE ONLY IN WELL

VENTILATED AREAS.

4.2.6.3 Relieve pressure in vessel thro ugh vents in

dust caps by cautiously opening inlet sampling valve.

4.2.6.4 When gauge indicates atmosph eric pressure

and flow ceases, close inlet sampling valve and remove

dust caps.

4.2.6.5 Connect fill hose to supply tan k and inlet

fittings of sampler.

4.2.6.6 Open supply tank outlet valve. Gas and liquid

will begin to flow from the sampler outlet.

CAUTION: WHEN OXYGEN IS SAMPLED,

SOME ADDITIONAL PRECAUTIONS MUST BE

OBSERVED. IF THERE IS NO FACILITY SUCH

AS A CLEAN CONCRETE PAD ON WHICH THE

LIQUID OXYGEN CAN EVAPORATE SAFELY,

IT WILL BE NECESSARY TO CATCH THE

LIQUID OXYGEN IN A CLEAN, PREFERABLY

SEAMLESS ALUMINUM BUCKET. USE A LINE

WITH MINIMUM INSIDE DIAMETER OF ONE-

HALF INCH TO DIRECT THE LIQUID OXYGEN

INTO THE BUCKET. FURTHERMORE, IT IS

IMPORTANT THAT THERE IS NO

RESTRICTION IN THIS LINE BECAUSE THE

BACK PRESSURE DEVELOPED MAY BE

SUFFICIENT TO CAUSE THE LIQUID IN THE

CUP TO OVERFLOW INTO THE LARGE

CHAMBER. ON WARMUP, THE PRESSURE IN

THE SAMPLER COULD BECOME EXCESSIVE,

PARTICULARLY IF THE SAMPLE WERE

SUBCOOLED, NECESSITATING THE

REPLACEMENT OF THE SAFETY RELIEF

DEVICE.

4.2.6.7 Allow sampler to cool until a s teady flow of

liquid appears at outlet.

CAUTION: AVOID CONTACT WITH THE

FLOW OF CRYOGENIC LIQUID. THE

EXTREMELY LOW TEMPERATURE CAN

CAUSE PAINFUL INJURIES.

4.2.6.8 Open inlet sampling valve completely to allow

liquid to enter sampling cup.

4.2.6.9 After 30 seconds, close sampling valve.

4.2.6.10 Close supply tank outlet valve.

4.2.6.11 Disconnect fill hose.

4.2.6.12 Invert sampler for five minutes to allow

sampling cup to empty and provide vaporization of

liquid.

4.2.6.13 At ambient temperature the sa mpler should be

at 400-500 psig, indicating that a good sample was

obtained. A lower pressure would indicate a leak in the

sampler or that the cup was not filled with liquid. If

there is any doubt, release gas from sampler and take

another sample.

4.2.6.14 Re-install inlet and outlet fitting caps. Do not

tighten.

4.2.6.15 Affix a tag identifying the product to the inlet

sampling valve handwheel.

4.2.6.16 Install cover.

4.3 Gas Phase Sample — (To Be Determined)

4.4 Liquid Sample — (To Be Dete rmined)

4.5 Delivery to Analytical Instruments — (To Be

Determined)

5 Quantification (To Be Dete rmined)

5.1 Linear Response

5.1.1 Direct Comparison

5.1.2 Calibration Curve

5.2 Non-Linear Response

5.2.1 Direct Comparison

5.2.2 Calibration Curve

6 Gas Chromatography

6.1 The analysis of many gaseous impurities in bulk

and specialty gases is done by means of a gas

chromatography separation of impurities from the gas

matrix and quantification using a broad range of

sensitive and sometimes selective detectors. This

section will provide a guide for the format and

representation of such a procedure. Figure 1 is designed

to represent the format for submission of such

methodology.

6.2 Sample Introduction — By its nature, gas

chromatography implies the use of carrier gases into

which the gaseous sample is injected and carried into

the separation medium, either solid or liquid in nature.

SEMI C3-0699 © SEMI 1986, 1999 6

The introduction of a gaseous sample into a carrier gas

stream is accomplished by means of a sample injection

valve. Valves for this purpose can be any of several

types, although the most commonly used are either

rotary or diaphragm valves using loops of stainless steel

tubing of a known volume. It is not necessary to

diagram such sample introduction systems for a

procedure unless the introduction system is unique to

the application by way of special conditions for safety

or other reasons critical to the accuracy of the sampling

technique.

6.3 Separations — It is an acceptable and common

practice to quantitate several gaseous impurities with a

single gas chromatographic separation. Occasionally a

single injected sample volume will undergo

"multidimensional" separation to achieve the necessary

analysis. Rotary valves, diaphragm valves or Dean's

pressure switching again are used to move impurities

from one separation medium to another, or from one

column to another. Where multidimensional techniques

are used, they should be accompanied by flowpath

diagrams which clearly indicate valving and their

appropriate positions. The valving sequences used

should have adequate explanation to clarify the

separations.

6.4 Columns — Columns used are generally of 3.2

mm (1/8 in.) o.d. and 2.2 mm (0.085 in.) i.d. stainless

steel of varying lengths and packed with one of many

solid supports. Columns may be specified by length and

packing material if these standard dimensions apply. If

another dimension or material is used for the column

itself, it should be specified by material type, and by

o.d. and i.d. in mm. Packing materials should be

specified by material, % coating and coating type if

applicable, and mesh size. Capillary columns are

assumed to be fused silica unless otherwise specified

and should be specified by i.d. in mm, film thickness in

µm, and length in meters. Column temperatures and

applicable program rates should be given for each

independently heated zone in degrees Celsius.

6.5 Carrier and Support Gases — Carrier and support

gas flow rates for all separations should be specified.

Carrier and support gases used should fall into the

general purity requirements in Table 2 and purity need

not be specified unless the analysis has specific purity

requirements for safety, accuracy, or component

lifetime. Flows should be specified in mL/min or

L/min, or as linear velocity for open tubular capillary

columns.

6.6 Detectors — Detector technolo gy should be

specified, and alternative detection may be assumed

equivalent for detectors of like selectivity and

sensitivity. Table 2 gives a reference of relative

detector sensitivity and is provided as a guideline only.

Detector parameters are less likely to be easily

formatted since detector technologies vary widely in

their specific parameters. All elements critical to

achieving like sensitivity must be included, such as

temperature, make-up gases and their flows, flame,

combustion or reaction gas ratios or flows, voltages,

currents, or any additional settings as outlined by the

detector manufacturer.

6.6.1 Detection limits must be speci fied for each

impurity for the actual methodology used. Because the

lower detection limit can be greatly influenced by

retention time for any impurity, it is important that the

detection limit be derived for the impurity within the

context of the method, and not solely on the

detectability of the detector. Detection limits are

assumed to be calculated on a mole/mole basis.

6.7 Operating Procedures — The standard procedure

assumed for the gas chromatographic analysis of most

gases is as follows:

1. Inject the calibration standard onto the column

using a gas sampling valve. Record the retention

time(s) and peak area(s) for all impurities detected.

2. Analyze the sample to be tested in the same

manner as the calibration standard.

3. Repeat 1.

4. Compare the average peak area of the calibration

standard with that of the sample being tested. (See

Calculation of Concentration, in Section 6.9.)

6.7.1 The method should specify an y additional valve

switching or special parameters which are necessary for

the successful analysis of the impurities specified for

the procedure. It must also specify the order of elution

if multiple impurities are detected.

6.8 Calibration — Calibration for these analytical

procedures is generally by external, single point

calibration. Calibration gas mixtures should be

specified and are assumed to be within the range

specified in Section 2.26. Calibration standards are

assumed to be made in a balance of gas representative

of the sample unless otherwise specified. If calibration

techniques other than external, point calibrations are

used, they must be described.

6.9 Calculation of Concentration — The calculation of

the concentration of impurity within the sample is based

on a comparison of the average peak area of the

impurity in the calibration standard, to the average peak

area of the impurity within the sample, based on the

formula:

Sample Peak Are

Standard Peak Are

Concentratio

of Standard

Concentratio

of Sample

SEMI C3-0699 © SEMI 1986, 19997

It is assumed that the results of such a calculation must

not exceed the specification for the product being

analyzed.

6.10 Notes — Any special requirements for any

section within gas method may be included here. Notes

should reflect the minimum requirements to

successfully achieve the desired detection limits with

the equipment specified.

1.1 Impurity analyzed - description of method (e.g., gas

chromatograph with thermal conductivity detector).

1.1.1 Detection Limit - (mole/mole)

1.1.2 Instrument Parameters

1.1.2.1 Column(s):

Adsorbent, mesh size, length in meters, (dimensions if

other than 1/8 in. by 0.085 in.)

1.1.2.2 Carrier Flow: mL/min, gas type

1.1.2.3 Support Gases:

1.1.2.4 Sample Volume: mL

1.1.2.5 Temperatures:

Detector: °C

Columns: °C

Other Equipment: °C

1.1.3 Calibration Standard - amount ppm (or ppb)

impurity, balance if other than helium.

1.1.4 Operating Procedure (Parameters)

1.1.4.1 Order of elution

1.1.4.2 Special instructions

1.1.5 Notes

1.1.6 Figures

Figure 1

Format for Gas Chromatographic Method

7 Special Analyses

7.1 Resistivity Measurement — The American Society

of Testing and Materials (ASTM) is developing a

method to measure resistivity in substrates. This

method may be referenced when it is finalized.

7.2 Metal Analysis of Gaseous Silicon Compounds

7.2.1 Because of the importance of t race elemental

impurities impacting silicon device characteristics,

there is an increasing need for the detection and

quantitative analysis of low-level impurities in the

silicon materials used in device fabrication. Some of

these materials are gaseous silicon-containing

compounds, such as silane and the chlorosilanes. For

elemental analysis by FTIR or PLS, a compound must

first be converted to a solid form such as a single

crystalline rod or disc. For the quantitative

determination of metals and some non-metallic

elements, the gases can be passed through a series of

bubblers containing high-purity DI water or dilute

acids. The aqueous media can be concentrated and

analyzed by suitable techniques such as ICP or AA

spectroscopy. In both cases, the recovery is assumed to

be 100% efficient or, in other words, there is no loss of

any impurity from the silicon source.

7.2.2 FTIR (Fourier Transform Infrared)

Spectroscopy, PLS (Photoluminescence Spectroscopy)

and NAA (Neutron Activation Analysis) are probably

the most useful tools for the detection and analysis of

trace quantities of elemental impurities in solid silicon

materials. NAA requires special silicon material

specimens to be submitted to neutron irradiation in a

nuclear reactor and the analysis facility to be approved

by the Nuclear Regulatory Commission (NRC) and

other government agencies. In contrast, FTIR and PLS

do not require special governmental regulations, and the

instrumentation is available on the open market.

7.2.3 An FTIR spectrometer is an infrared

spectrometer in which a Michelson Interferometer is

used in place of a grating or prism. The simplicity of

the Michelson Interferometer, with only one moving

part, an oscillating mirror, along with a He-Ne laser as a

reference, provides nearly absolute frequency accuracy.

This translates into a high spectral resolution. FTIR has

greater detection efficiency since the energy-wasting

slits required for dispersive spectrometers are not used.

With a microcomputer to control the functions and to

perform data processing, signal averaging is used to

improve the signal-to-noise ratio. All these features lead

to good reproducibility and rapid measurement results

compared with conventional IR techniques.

7.2.3.1 As FTIR is a transmission met hod, it requires a

single crystalline silicon specimen with flat, parallel,

mirror-polished surfaces. For oxygen in silicon, the

absorption peak at 9 microns is used for quantitative

analyses. For carbon in silicon, the absorption band at

16 microns is used. Both measurements may be

performed at room temperature. (The carbon

determination has strong interference from the silicon

phonon band, but can be resolved with high purity,

carbon-free standards.) For the determination of

shallow donors and acceptors, FTIR measurements

must be made at liquid helium temperatures, but the

through-put is low. Although the operation of today's

FTIR instrument is relatively simple, the interpretation

of the results through software and associated

problems/solutions is not. Quantitative analysis is based

upon the measured percent transmission at the

characteristic wavelength and compared to standards.

Detection limits at about 10-12°K are in the 10

to 10

atoms/cc for shallow donors and acceptors, 3.1 × 10

atoms/cc for oxygen, and 2 × 10

atoms/cc for carbon.