GB2518181A - Barrier Testing - Google Patents

Abstract

A method of measuring a permeation efficiency of a barrier comprises:      placing a test sample 23 in a first chamber 22; and providing a reference gas 46 to a second chamber 44 separated from the first by a barrier 66, which reference gas has a plurality of stable isotopes. An equilibrium quantity of particles of the at least one stable isotope is maintained. Mass spectrometer readings in relation to the at least one stable isotope and the permeate are obtained, and correlated with the maintained quantity of at least one stable isotope particles, to obtain an absolute value for a quantity of the permeate, to establish a quantity of permeate passing through the barrier per unit time. The method may overcome problems associated with drift. A method of achieving a selected pressure in a chamber comprises providing the chamber with a first gas at a first pressure through an orifice plate with a first conductance C1, the diameter being selected to achieve molecular flow. The speed and conductance C2 of a pump 47 are adjusted to achieve molecular flow, and the first pressure and first and second conductance are adjusted to establish and maintain the selected pressure.

Description

BARRIER TESTING

In many technologies there is a need to protect sensitive features of a product from, for example. substances in the atmosphere, such as water, oxygen, even nitrogen or other substances. Conventionally such sensitive features are protected from exposure to the atmosphere, or to contaminating substances, by bather layers. The barrier layers used must be suitable for the degree of protection desired, in terms of component materials and thickness and the like, and so, when developing a suitable barrier layer, it is necessary to know the sensitivity and effectiveness of such a barrier layer in relation to the protection provided in respect of the contaminants which are of concern.

For example, a barrier layer may be provided over an electronic component. or a portion of an electronic component, to protect the component from water, oxygen or other substances. Such a barrier layer usually comprises a thin film or membrane made from a material which provides the functionality desired, i.e. it provides a barrier to the passage, from one side of the barner to the other, of particular undesirable substances, usually gaseous water or oxygen, although other substances are also contemplated. Therefore when choosing a protective barrier layer it is essential to know exactly how effective the barrier ayer is likely to be for the degree of protection sought.

It is known to test barriers to detect the effectiveness of the barrier in preventing passage of a variety of contaminants from one side of the barrier to another, and test equipment has been developed for this purpose.

Figure 1 shows an example of such test equipment, comprising a first chamber 2 for h&ding a test sample, the first chamber separated from a second chamber 4 by a barrier 6. The second chamber 4 contains a monitor 8, or is adjacent a further chamber 10 which includes the monitor 8. A test sample 7 is provided to the first chamber 2 and the monitor 8 is arranged to detect how much of the sample material 7 reaches the monitor 8, i.e. how efficient the barrier 6 is at preventing passage of the test material from one side of the barrier to the other. This is conventional.

Conventionally the monitor may be a mass spectrometer and many types of mass spectrometer are available, both magnetic and non-magnetic. Magnetic mass spectrometers are large and expensive and so non-magnetic mass spectrometers are mostly considered to be more suitable. One suitable non-magnetic mass spectrometer is a quadrupole mass spectrometer, although other types of mass spectrometer are contemplated.

When initially placed in such test equipment a barrier may include within or on itself contaminants which will influence the readings taken. For example the bather is 1 0 likely to contain contaminants such as water or oxygen either within the body of the bather or on a surface of the barrier, and other contaminants such as Nitrogen or other materials may also be present. Such contaminants leave the barrier by outgassing and diffusion for a period of time after the barrier is introduced into test equipment, leading to delay in achieving a reliable reading for ban'ier efficiency.

Figure 3 is a typical example of the sorts of readings obtained by the test equipment in 0) Figure 1, in which the contaminant detected drops quickly initially and then drops 0 more slowly to achieve a steady, continuing rcading, or tail'. The initial high values If) are generally due to outgassing and diffusion of contaminant present on the barrier O 20 itself, with the steady, continuing reading relating to permeation. It can take a considerable time before contaminants from the barrier itself no longer contribute to any measurements taken, and to achieve a confidence that measurements actually relate to passage of species in chamber 2 through the barrier 6 into chamber 4 or 10 for measurement by the quadrupole mass spectrometer, i.e. to be confident that the efficiency of the barrier itself is being measured. This means that if a new barrier is developed it can take a considerablc length of time to establish, using conventional means, how effective the barrier is, Any deficiencies will therefore take even longer to correct and further testing will be needed which will extend the test time even further.

As stated, as the equipment is testing the efficiency of the barrier it is necessary to reach a position where the amount of contaminant measured is coming from the sample in chamber 2 rather than from the barrier itself For ultra barriers with permeation rates of the order of io-to I 06 glm2lday. the measurements can take up to 4-5 days to get to a position where there is no significant reading for contaminants on or in the barrier, so that all readings relate to substances which have passed through the barrier. Therefore the timescale for conventional barrier test equipment is comparatively long.

An additional problem is that mass spectrometers may be subject to drift': a drift in a reading of up to 50% per day for a, for example. quadrupole mass spectrometer is not unknown, and is mainly caused by drift in various components of the mass spectrometer, for example the electron multiplier detector, mass selection device, or other elements.

Electron multipliers rely on electron stimulated surface emissions, and as such are dependent on the level of gas present. In particular, electrons are absorbed at the surface of the electron multiplier and then emitted therefrom. The gas present influences the amplification factor' of the electron multiplier, as do a number of other factors, including the pressure, temperature, and electron energy among others.

Sometimes the various drift elements relevant to the amplification factor cancel each other out, and sometimes they reinforce each other.

This is the case for all non-magnetic mass spectrometers, including quadrupole mass spectrometers. however it is not a problem for magnetic mass spectrometers. as the ion energy for such devices is sufficiently high (-1000 volts) that significant deviation is not generally the problem it is for non-magnetic mass spectrometers. However, magnetic mass spectrometers are very large, expensive, and power hungry, whereas non-magnetic mass spectrometers are smaller, easier to manage, much less expensive and use much less power. For these reasons, among others, it is more desirable to use non-magnetic mass spectrometers if possible, although a number of problems must be addressed if non-magnetic mass spectrometers are to be relied upon.

One such problem is that quadrupole mass spectrometers are an unsuitable detector for these kinds of measurements without repeated calibration, which adds further delay and complication to the process. For example, each calibration step is likely to include, inter alia, a pumping step to completely remove the calibration gas from the system to avoid false readings and this generally takes some time.

Thus current methods for detecting the efficiency of a barrier present a number, and a variety, of problems.

The present invention seeks to reduce or overcome the disadvantages of pnor arrangements including those set out above.

Preferred embodiments of the invention will now be described by way of example and with reference to the following drawings in which:-Figure 1 shows an arrangement for testing a barrier effectiveness in accordance with

the prior art,

Figure 2 shows a barrier with contaminants within and on a surface, Figure 3 shows a graphical representation of readings of H20 over time provided in C accordance with the arrangement of Figure 1, i_c Figure 4 shows test equipment arranged in accordance with the present invention, and O Figure 5 shows a detail of the test equipment of Figure 4.

Figure 4 shows a chamber 22 separated from a further chamber 44 by a bather 66, the barrier 66 being the barrier to be tested. Chamber 22 may be provided with a test sample 23. Chamber 44 includes means 45 for providing a gas 46 to chamber 44 and also a pump 47, for example a throttled pump with a pump speed, for removing gas from chamber 44 and maintaining a desired pressure regime. A mass spectrometer, for example a quadrupole mass spectrometer, is positioned in a further chamber 100 with an ion source 101, the further chamber 100 also containing a further pump 102 for example an un-throttled vacuum pump with a pump speed, for maintaining a desired pressure regime in thither chamber 100.

The arrangement is such that, to test the efficacy of the barrier 66, a test sample is placed in chamber 22 and the amount of the test sample detected by the mass spectrometer is monitored over time.

However there are a number of problems associated with the use of a mass spectrometer, such as a conventional quadrupole mass spectrometer, when testing an ultra barrier to determine permeation. One problem is that readings provided by the mass spectrometer are subject to drift, as discussed above.

Figures 4 and 5 relate to an apparatus and method directed to overcoming problems associated with this drift.

Chamber 44 is provided, by conventional means 45, with a gas that is inert with respect to said permeate species, for example a gas which has a mass spectrum that does not interfere with the mass spectrum of the permeate species, and which also has at least one minor stable isotope. Preferably the stable isotope is present in the naturally occurring provided inert gas in suitable proportion, such that, with the provided inert gas provided to chamber 44 at a selected pressure, the quantity of the minor stable isotope at that pressure is of a similar order of magnitude to the quantity 0) of permeate species expected to be present in chamber 44 due to permeation through O barrier 66. In general it is contemplated that the provided inert gas has minor stable isotopes which naturally comprise between 100 to 5,000 ppm of the provided inert 0 gas, and provide corresponding partial pressures to the pressure in chamber 44. It is contemplated that the pressure selected for maintaining the provided inert gas in second chamber 44 should be 50 to 1000 times the partial pressure in chamber 44 due to the permeate species, present in chamber 44 due to permeation of the membrane under test, and typically 100 times, for best effect. Conventionally, with the volume of chamber 44 and the pressure and temperature of the gas known, the quantity of the minor stable isotope present in the chamber may be calculated. By this means an absolute value for the quantity of the minor stable isotope may be obtained.

Mass spectrometer readings for the minor stable isotope may then be taken and compared with the quantity of isotope present as calculated, to thereby allow the mass spectrometer reading of the permeate to be scaled appropriately to indicate the absolute quantity of permeate present in chamber 44.

Thus the isotope measurements may be used as a reference for measurements of permeate. In particular, as the minor stable isotope is present in a quantity that is similar to that expected for the permeate, and both the readings -for the minor stable isotope and permeate -will experience the same drift, then drift effects can effectively be eliminated from the measurements and the resulis obtained for the permeate can be relied upon as an abso'ute, rather than as a relative, measurement.

A suitable candidate for the provided inert gas is Argon which has a peak at 40 amu with minor stable isotope peaks at 36 amu and 38 amu; or nitrogen which has peaks at 28 and 14 amu with minor stable isotope peaks at 29, 30 and 15 amu. Other candidates are also suitable and contemplated and would be understood to be included by the skilled addressee.

There are many ways in which the pressure in the second chamber 44 may be established. Conventionally, pressure down to 103mb can be measured very accurately using low cost instruments, however measuring pressure below l0 mb using low cost instruments is more difficult and prone to error. More expensive equipment is available for accurate measurement of pressure below l0 mb for example a conventiona' spinning rotor gauge may be used to measure to lO' or l0 mb but below 10 the accuracy of even this equipment is poor, with errors significantly increasing the lower the pressure.

The present anangement provides for establishment of a pressure below lO3mb, more particu'arly the present anangernent provides for estabhshment of a pressure in the range io to 108mb and typically approximately S 10 mb.

As a first step gas in the chamber must be subject to molecular flow. Conventionally, for molecular flow the Knudsen number must be greater than 1. i.e.: Kn >1. in accordance with the following equation: (1) Kn = X/d Where: d = dimension of the physical length X = molecular mean free path length in the same units Particular advantages of a molecular flow regime are that in molecular flow the conductance is independent of the pressure and is proportiona' to the inverse square root of the atomic mass of particles of the gas. in particular: (2) C a 1!(atornic mass)"2 Such conductances relate to, in particular, the conductance of an orifice plate providing a gas to a chamber in molecular flow, or the conductance of a pump with a pump speed removing gas from such a chamber, to maintain an equilibrium quantity of particles in the chamber. Such an onfice plate and pump are set out in Figures 4 or 5.

In addition the conductance of an orifice plate such as that set out in Figures 4 or 5 is given by: (3) C= (v/4)*A where A is the area of the orifice in cm2, and is governed by the following equation: (4) A = md214 and v is the average gas velocity and is governed by the following equation: (5) v=(SKTIirm)h/2 where K is the Boltzmann constant and m is the molecular mass of the gas.

For air/nitrogen at room temperature this simplifies to: (6) C = 11.6*A 1/s Further, with the provided inert gas introduced into chamber 44 at a pressure P1, in a molecular flow regime, the pressure P2 in chamber 44 is governed by the following equation: (7) P2=Pl*Cl/C2 where Cl refers to the conductance of an orifice plate relating to the provided inert gas in Figure 4, and C2 refers to the conductance of throttled pump 47 pumping out the contents of chamber 44 with a selected pump speed. Thus by selecting or adjusting the pressure with which the gas is provided to chamber 44, and selecting the conductance of the orifice plate and pump speed, the pressure in chamber 44 can be set. This pressure can be a very low pressure as shown in the examples set out below.

As noted above and in accordance with equation (2): Cl a i/(atomic mass into chamber 44)1/2 C2 a l/(atomic mass out from chamber 44)1)2 This shows, with equation (7), a particular advantage of a pure molecular flow regime, that when calculating the pressure, P2, the mass terms inherent in conductance cancel out and do not need to be considered. Issues related to the mass component may not be significant where the masses in Cl and C2 are similar, for example when the atomic masses in Cl and C2 relate to 40Ar and 38Ar, however when the respective masses in Cl or C2 are very different, for example when one relates to Ar and the other to Il2 or H20 then problems in rdation to fractionation would need to be considered. In the present case as the mass components cancel out and fractionation may be disregarded. Problems associated with fractionation are conventional and will not need to be considered further.

Thus with molecular flow we can use conductance to establish pressure. which is very useful, particularly for the low pressure regime of interest herein.

In relation to the pressure in chamber 100, we can establish two further conductances, C3, which refers to a conductance of a Mass Spectrometer ion source and/or the connection to a Mass Spectrometer ion source, and C4, which refers to a conductance of an un-throttled vacuum pump with a pump speed in chamber 100.

The pressure P3 in chamber 100, containing the Mass Spectrometer, is then governed by the following equation 8: (8) P3 = P2C3/C4 The values needed for parameters Cl -C4 and P1 -P3 to ensure a molecular flow regime, in particular to ensure that pressure in chambers 44 and 100 may be calculated in accordance with the equations (1) -(8), may now be set.

To ensure the gas in chamber 44 is to be subject to molecular flow, and in accordance with the equations and the discussion of Knudsen number above, an appropriate size for the diameter d' of Cl (the orifice plate) can be set. It is noted that a value selected for "d" (in equations (1) and (4)) will influence the throttled pump speed selected, which determines the value of C2, and provides for a suitable ratio for Ci/C2.

Example 1:

If the orifice diameter d' is set to 330pm, or 3.3x102cm, then by equations (4) and (6) we have a value for Cl of 10Is. If we set the pump speed of pump 47 in chamber 44 such that the conductance is 201/s then we will have a value for the ratio Ci/C2 of 5x1ft4. If we then set the pressure P1 at the inlet to 0.1 mb and establish that the Knudsen number. Aid, is greater than 1, confirming we have molecular flow conditions, we can establish that the pressure P2 in chamber 44 is governed by equation (7) and will have a value: P2=5.OxlO mb.

lii a similar fashion, to establish the pressure P3 in chamber 100 we can set: C3 = 0.1 us, and C4 ? 300 us So that, with P2= 5.0xi0mb Then in accordance with equation (8): P3<1.7xlE8mb These values ignore outgassing from the chambers etc., and also assume that the pressure from the permeate species is low compared to the pressure of the provided inert gas.

In the case where C3 is a significant % of C2, say >1%, we need to replace C2 in equation 8 with CT given by: (9) CT=C2 -F C3

Example 2:

For the case where C3=2 1/s and C2=10 1/s then by equation (9) C2'=12 1/s. with an orifice diameter of 33 tm we have from equations (6) and (9) C1= l0 1/s and with P1=1 mb we can establish that the pressure P2 in chamber 44 is given by equation (7) and will have a value: P2= 8.33 106mb And with C4 »=300 1/s and C3 = 1 us we have P3P2*C3/C4 which gives P3 C 2.78 io mb With the equipment arranged in accordance with Figure 4, a sample in chamber 22, and a suitable inert gas provided to chamber 44. the quadrupole mass spectrometer in chamber 100 will provide measurements of the permeate species along with the stable isotope of the provided inert gas. The total gas pressure in second chamber 44 is also monitored, preferably using conductances as set out above, and the stable isotope readings corrected for any variation in pressure or temperature. The permeate species readings are then normalised to the pressure corrected stable isotope readings and the effects of detector drift eliminated.

The values available for the parameters are likely to be constrained in the case of, for example, C2 (the conductance of the throttled pump) to fall within the range of 5 -40 us: if less than 5 1/s then effectively no pumping is occurring, and if greater than 40 1/s then Ci would need to be significantly smaller to satisfy the requirements of equation (4), so small in fact that limited gas flows could enter chamber 44, and so on.

It is contemplated that typical values for the C1/C2 ratio are: i03 -106, with a typical value of 5xi04. A typical provided inert gas pressure range lying between I and 103mb, with a typical value of Sxlft2mb, will give a pressure in chamber 44 of -s 2.SxlO rnb.

A further problem which may be encountered in respect of the present alTangement relates to measurements taken by the mass spectrometer in relation to the two species, one being the stable minor isotope of the provided inert gas and the other being the permeate. In particular, the provided inert gas and the permeate species behave differently in the mass spectrometer and so calibration must take place so that the two sets of readings can be understood and interpreted correctly.

For example, each of the permeate and the stable isotope of the provided inert gas. on entry into the mass spectrometer, generate respective readings in amps/unit pressure.

Different species generate different amps for the same pressure, in particular H20 generates different amps/mb to Nitrogen, or Argon, and so on. If the readings are to be interpreted correctly a calibration step must be carried out.

One such calibration step directed to address such effects is to, at intervals, switch the gas in chamber 44 from a pure provided iner gas to one which has added to it a known amount of permeate, for example 100 ppm of the permeate. This addition of the iOOppm of permeate will have no measureable effect on the pressure reading. It is possible to sow-ce a gas with a known component of a contaminant species (for example, permeate) and national standards apply. It is contemplated that chamber 44 is isolated from the test chamber, and from any permeate proceeding through the barrier, before such a calibration step is carried out.

Preferably the proportion of contaminant (for example, permeate) in the switched gas would be similar to that expected from permeation of the permeate through the barrier.

There are many ways to accommodate switching gas in chamber 44 from a pure provided inert gas to one containing a known concentration of permeate.

One way is to include a valve in chamber 44, so that in one arrangement the valve allows pure provided inert gas into chamber 44, and in a second arrangement the valve is switched to allow contaminated provided inert gas into chamber 44.

lii each case, as stated, it is contemplated that during calibration no permeate will pass into chamber 44 through the barrier, and this can be accomplished by, for example, removing the test sample from chamber 22, or having a further valve in chamber 44 which can close to prevent permeate which has passed through the barrier entering chamber 44, or entering the relevant portion of chamber 44. and contributing to any measurement taken. Removing the test samp'e from chamber 22 has the disadvantage that permeate present in the barrier will continue to enter chamber 44 for a period of time, and also steps will need to be taken to remove the test sample which will be at least time consuming. With permeate no longer entering chamber 44, a measurement or a series of measurements of the pure provided inert gas can be taken with no peimeate species present. A further measurement or series of measurements may then be taken following provision of a contaminated inert gas to chamber 44, i.e. a gas with a known amount of permeate species (the contaminant) present. Calibration can then occur.

Such calibration can be performed relatively infrequently, for example weekly or 1 0 fortnightly, as, although there is a risk of a relative sensitivity drift within the quadrupole mass spectrometer in relation to the measurements of the two species, this occurs at a much reduced rate, by at least 10 to 100 times, compared to detector drift.

The inventors note that the graph in Figure 3 shows the measurement, on the y axis over time, of a species present in chamber 44, the value reflecting the relative contributions from outgassing and diffusion (both of products initially present on or in o the barrier), and permeation (of product passing through the barrier). The contribution due to species present in or on the barrier initially (the initial readings) is o much higher than that due to species passing through the barrier (the later readings), but we can assume it remains steady so that when the contribution from outgassing and diffusion has dissipated, the permeation contribution remains.

The inventors contemplate that filling the test chamber with a substance, for example Deuterium, which is naturally sufficiently rare that it is not generally present, would ensure that the amount of Deuterium present in Fig 3 is due to permeation only, and this might be useful if, for example, permeation of Hydrogen through the barrier was being tested, Due to its rarity a degree of confidence could be established that no Deuteriuni would be present on or in the barrier initially, and that none would have outgassed or diffused from the barrier, so that the only Deuterium detected by the mass spectrometer would be due to permeation from the test sample through the barrier. This would mean that confidence in any measurements would be high from the start and no delay would arise from any uncertainty as to the source of the species detected. In a similar fashion isotopes of, for example 02, or rare isotopes of other permeate species. could be used as test samples to provide the time advantage sought in measuring the efficiency of a balTier.

Such further means can also avoid the time lag associated with measuring a barrier efficiency to ensure that no readings related to substances on or in the barrier are being measured, however it is not certain that suitable isotopes will be available for use, or they may be prohibitively expensive.

The advantages of the present apparatus and method is to allow for measurement of a very low pressure in a simple way, to substantially eliminate concerns about drift in mass spectrometer readings, and to provide a confidence in the values obtained, by carrying out calibrations which are helpfully infrequent but reliable.

The invention is not limited to the embodiments disclosed.

Claims (24)

1. CLAIMS: I. A method of measuring a permeation efficiency of a balTier including: providing said barrier between a first and a second chamber; placing a test sample in said first chamber; providing a reference gas to said second chamber; wherein said reference gas is unreactive with respect to the test sample; wherein said reference gas has a plurality of stable isotopes; maintaining an equihbrium quantity of particles of said at least one stable isotope in said second chamber such that said quantity is within at least one order of magnitude of an expected number of particles of said test sample present in said second chamber due to permeation through said barrier; obtaining a mass spectrometer reading in relation to said at least one stable isotope; obtaining a mass spectrometer reading in relation to said permeate. and colTelating said mass spectrometer readings and said maintained quantity of at least one stable isotope particles to obtain an absolute value for a quantity of said permeate, to establish a quantity of permeate passing through said balTier per unit time.
2. 2. A method as daimed in claim I wherein said reference gas is provided at a predetermined pressure andlor a predetermined temperature.
3. 3. A method as claimed in claim 2 wherein said equilibrium quantity is calculated based on the universal gas equation.
4. 4. A method as claimed in claim 1, wherein said reference gas is selected based on at least one stable isotope comprising iQO to 5000 ppm of the reference gas when found in nature.
5. 5. A method as claimed in claim 1 wherein said gas comprises a gas selected from the group 8 elements, or Nitrogen.
6. 6. A method as claimed in claim 5 wherein said gas is Argon.
7. 7. A method as daimed in claim I wherein a mass spectrometer taking said mass spectrometer reading is in said second chamber.
8. 8. A method as claimed in claim 1 wherein a mass spectrometer taking said mass spectrometer reading is in a third chamber adjacent said second chamber.
9. 9. The method of dairn I further including a calibrating step comprising: isolating an internal volume of said second chamber from said permeate; evacuating said second chamber; providing said second chamber with said reference gas; taking a first mass spectrometer reading relating to said reference gas; evacuating said second chamber; providing said second chamber with said reference gas plus a known quantity of said permeate species; taking a second mass spectrometer reading relating to said permeate species, and calibrating said mass spectrometer by correlating said first and second readings.
10. 10. Apparatus for testing a permeation efficiency of a barrier including: means to separate a first chamber from a second chamber by said barrier; means to provide said second chamber with a reference gas; means to maintain an equilibrium quantity of said reference gas particles in said second chamber; a mass spectrometer for obtaining a reading for said quantity of a gas in said second chamber; means to correlate said reading with said equilibrium quantity; means to compare said reading with a mass spectrometer reading of a test sample provided to said first chamber and permeating said test balTier, and means to derive an absolute value for said permeate quantity in said second chamber dependent on said comparison.
11. 11. Apparatus as claimed in claim 10 further including means for maintaining a selected pressure andior a selected temperature in said second chamber.
12. 12. Apparatus as claimed in claim 10 further comprising: means for isolating an internal volume of said second chamber from said permeate; means to evacuate said isolated internal volume; selection means to provide said isolated internal volume with a first gas or a second gas; rneaim to correlate readings from said first gas and said second gas to calibrate said apparatus.
13. 13. Apparatus as claimed in claim 12 wherein said isolating means comprises a valve.
14. 14. Apparatus as claimed in claim 12 wherein said selection means comprises a valve.
15. 15. Apparatus as claimed iii claim 10 wherein said mass spectrometer is placed in said second chamber.
16. 16. Apparatus as claimed in claim 10 wherein said mass spectrometer is placed in a third chamber adjacent said second chamber.
17. 17. A method of achieving a selected pressure in a chamber, comprising: providing said chamber with a first gas at a first pressure through an orifice p'ate with a first conductance, a diameter of said orifice plate selected to achieve molecular flow; adjusting a pump speed and conductance of a pump in said chamber, said pump speed and conductance selected to achieve molecular flow; adjusting said first pressure and first and second conductance to establish and maintain said selected pressure.
18. 18. A method as daimed in claim 17 wherein said first gas is selected based on having a plurality of stable isotopes.
19. 19. A method as claimed in claim 17 wherein said first gas includes a predeteimined proportion of at least one stable isotope.
20. 20. A method as daimed in claim 19 wherein said proportion is between 100 and 5000 ppm of said gas.
21. 21. A method as claimed in any one of claims 17 -20, wherein said first and selected pressures are partial pressures of said stable isotope.
22. 22. An apparatus for achieving a selected pressure in a chamber comprising: an orifice plate with a first conductance; a pump with a pump speed and a second conductance; means to provide to said chamber through said onfice plate a first gas at a first pressure; means to adjust said first pressure and said first and second conductances to provide for m&ecular flow, and means to control said first pressure and said first and second conductances to achieve said selected pressure in said chamber.
23. 23. An apparatus as claimed in claim 22 wherein said orifice plate diameter is adjustable to adjust said first conductance.
24. 24. An apparatus as claimed in claim 22 wherein said pump speed is adjustable to adjust said second conductance.