semi合集-English.pdf - 第3887页

SEMI F27-0997 © SEMI 1997, 2003 2 ambient ai r or non-dry gas . Thus, it is necessary to consider two main situation s: 1. The “initial dr y-down” situation, which is determined by the moistu re content of the component …

SEMI F27-0997 © SEMI 1997, 2003 1

SEMI F27-0997 (Reapproved 1103)

TEST METHOD FOR MOISTURE INTERACTION AND CONTENT OF

GAS DISTRIBUTION SYSTEMS AND COMPONENTS BY

ATMOSPHERIC PRESSURE IONIZATION MASS SPECTROMETRY

(APIMS)

This test method was technically reapproved by the Global Gases Committee and is the direct responsibility

of the North American Gases Committee. Current edition approved by the North American Regional

Standards Committee on September 16, 2003. Initially available at www.semi.org October 2003; to be

published November 2003. Originally published September 1997.

1 Purpose

1.1 This test will determine the quantity of removable

moisture and the degree of interaction with trace

concentrations of gas phase moisture, of gas

distribution systems and components. APIMS is

currently the technique of choice for such tests because

it is essentially the only commercially available method

capable of ppt moisture analysis and because of its

superior response time. This method may provide

guidelines for the application of other techniques with

similar detection limits and response time to APIMS

which are not commercially available at this time.

1.2 The results of this test can be used for qualitative

ranking of systems and components and can also be

used, by a sufficiently sophisticated user, as input for

numerical simulation of distribution system behavior.

2 Scope

2.1 System and Component Types — This procedure

applies to in-line components to be used to contain

electronics grade materials in semiconductor gas

distribution systems. The following components are

expected to yield meaningful results when tested

according to the present method: tubing, connectors

(fittings), particle filters, valves (check, relief, shut-off,

and metering), regulators, flow-through

transducers/sensors, mass flow controllers and meters.

Components with dead volumes, such as pressure

gauges, can be tested according to this method, but the

results will be difficult to interpret. Additional criteria

besides those considered here need to be developed for

testing components with dead volumes.

2.1.1 As gas phase moisture levels are dominated by

desorption from surfaces, the procedure is expected to

be most useful for components of large surface area

such as particle filters and tubing. Components with

very small wetted surface areas may have moisture

interactions which are too small to measure using this

test. Check valves and relief valves can be tested only if

their operating parameters are consistent with the test

conditions.

2.1.2 Tubing samples must be rather long (3–4 m) to

yield useful results, but they can be bent into a U shape

in order to be accommodated into a practical test bench.

Radii of curvature should be no less than 6 times the

internal diameter of the tube and the minimum number

of bends should be used. Most ultra-high purity

components are currently supplied with metal gasket

connectors, so the test bench employed should be fitted

with mating connectors. This type of fitting is usually

welded to tubing, which is not recommended as it

introduces a potential source of uncontrolled variability

to the experiments. Instead, a compression fitting on the

tube and a suitable adapter should be used. The

compression fitting should not be disconnected after

initial connection until the test series is over. In this

way, deterioration in integrity of the connection can be

avoided. The same type of gasket to compression fitting

adapter should be used for all tubing samples.

2.1.3 Purifiers require special test procedures and are

not addressed here.

2.1.4 Simple systems consisting of components

connected in line can be tested by the present method.

Complex systems (i.e., those with more than one

potential inlet and/or outlet) will show performance

which varies depending on the test configuration

(which inlets and outlets are chosen). Testing of such

systems is not addressed by the current document.

2.2 Gases — The procedure will be carried out in

nitrogen. The results will provide a ranking with respect

to moisture contribution which may be applied with due

caution to components intended for use in other gas

streams. Other “inert” gases will have different purging

characteristics and may dry a component more quickly

or slowly. Reactive gases may react chemically with

moisture. Considerations relating to corrosion

resistance are outside the scope of the present

document, although the test procedure may prove useful

in corrosion studies.

2.3 Operating Situations — Moisture contribution from

a component may be the result of contamination arising

in its manufacture, or from subsequent exposure to

SEMI F27-0997 © SEMI 1997, 2003 2

ambient air or non-dry gas. Thus, it is necessary to

consider two main situations:

1. The “initial dry-down” situation, which is

determined by the moisture content of the

component as received with the effects of

manufacturing process and design, surface quality,

pretreatment and packaging convoluted together.

2. The “Response to upset” situation, which is

determined by the amount of moisture taken up by

the component and subsequently released in any

exposure after receipt.

NOTICE: This standard does not purport to address

safety issues, if any, associated with its use. It is the

responsibility of the users of this standard to establish

appropriate safety and health practices and determine

the applicability of regulatory or other limitations prior

to use.

3 Limitations

3.1 This test method allows the determination of

moisture interactions which can be used, for example,

to rank components in order of decreasing moisture

interaction. Because different degrees of moisture

interaction are desirable in different situations, selecting

the “best” components requires consideration of how

they will be used in a given distribution system, either

qualitatively or through numerical simulation of

distribution system behavior.

4 Referenced Standard

4.1 SEMI Standard

SEMI C15 — Test Method for ppm and ppb Humidity

Standards

NOTICE: Unless otherwise indicated, all documents

cited shall be the latest published versions.

5 Terminology

5.1 Acronyms and Abbreviations

5.1.1 APIMS — Atmospheric Pressure Ionization Mass

Spectrometry

5.1.2 EPSS — Electropolished Stainless Steel

5.1.3 MFC — Mass Flow Controller

5.1.4 ppm — molar parts per million (µmole/mole).

The same as ppmv.

5.1.5 ppb — molar parts per billion (nmole/mole). The

same as ppbv.

5.1.6 ppt — molar parts per trillion (pimole/mole). The

same as pptv.

5.2 Definitions

5.2.1 induction time — the elapsed time between when

humidified gas is input to the test component and when

moisture is detected at the moisture analyzer. For a

component and test system which are perfectly

transparent to moisture the induction time is equal to

the residence time of the gas in the system.

5.2.2 peak height — the maximum moisture

concentration recorded when a moisture input of

predefined length and concentration is introduced to a

test component.

6 Required Equipment

6.1 APIMS — The APIMS used for moisture detection

can be of any type. In order to be considered valid, any

series of moisture tests shall reference two single point

calibrations, one at the beginning and one at the end of

the series. These calibrations shall both be made under

the same analytical conditions (flow, pressures, plate

voltages, etc.) as the tests and the results shall be within

5% of each other. Provided this condition is met, the

interval between calibrations may be left to the

discretion of the operator; however, it is suggested that

calibration should be carried out whenever the

equipment is moved and/or every two weeks of

operation.

6.2 Dry Gas and Moisture Generator — A source of

extremely dry nitrogen (less than 100 ppt moisture) and

a moisture generator capable of delivering up to 13 slm

nitrogen doped with 50 ppb moisture is required. This

generator may be the same as used to calibrate the

APIMS. The output of the generator shall be verified

according to SEMI C15.

6.2.1 Most ultra-high purity components are currently

fitted with metal gasket type connectors. The same type

of connector should, therefore, be incorporated into the

test bench for connection to the test components.

6.3 Test Blank — Any series of tests shall include the

results of testing a blank. The blank shall be the shortest

convenient length (no more than 1 m) of 1/4"

electropolished stainless steel (EPSS) tubing with

suitable fittings and/or adapters at either end to enable it

to be inserted in place of the test specimen.

6.4 Moisture Pulse Generator — A valving arrangement capable of switching rapidly between dry and humidified

nitrogen is also required. An example of such a design is shown in Figure 1. In this design, flow is maintained in

both the humidified and dry gas lines at all times. By simultaneously switching valves V

and V

, or V

and V

either humidified or dry gas is directed through the test specimen while the other gas stream is directed to vent. Note

that there is a bypass loop so that flow can be maintained to the APIMS when the specimen is removed. In this

design, all gas lines, but especially those lines between V

and V

and between V

and V

, should be as short as

possible and constructed of EPSS tubing of high quality. Maintaining gas lines at a constant temperature between 50

and 80°C wherever possible is also recommended.

Figure 1

Moisture Test Schematic

6.4.1 The bypass loop will contain a stagnant volume of gas during testing. To avoid exposing the APIMS to a large

moisture upset whenever it is fed the gas contained in the bypass loop, the loop should be thoroughly baked out (at ≥

200°C) and protected from atmospheric contamination thereafter. Valve V

should be such that some flow can be

maintained through the bypass loop and the test specimen simultaneously as well as through each separately. Use of

pneumatic valves to facilitate rapid and simultaneous switching is recommended.

6.4.2 Other arrangements than that shown may also be used. The moisture pulse generator should be designed so as

to give the fastest possible response of the blank to a change in input moisture level.

6.4.3 Use of a glove box or other such enclosure about the test specimen and adjacent piping, to minimize exposure

of the system to ambient contamination when removing and introducing the test specimen, will improve the

precision of moisture content measurements but not of moisture interaction measurements. Use of such an enclosure

is, therefore, left to the discretion of the operator.

6.5 Temperature Control and Measurement — A stable temperature during the test is of critical importance. The

test component shall be kept at 35°C. TESTS AT DIFFERENT TEMPERATURES CANNOT BE

COMPARED. Ideally, the blank and device under test should be maintained in a temperature-controlled chamber.

However, if this is impractical, heating tape can be used. Temperature shall be measured at 1 m intervals on tubing.

Measurement at one point is adequate for small components. Temperature control should be to ± 1°C. A continuous

record of temperature during the test shall be maintained.

7 Procedure

7.1 Blank Tests — A blank test shall be carried out after each calibration. If initial dry-down testing of components

is not of interest, the initial dry-down test of the test blank may be omitted and the test blank brought to equilibrium

with zero gas in whatever manner is most convenient, except that the test blank should not be heated above 200°C.

Permanent changes in moisture interaction have been observed at temperatures above this level.