GB1598426A

GB1598426A - Leaktesting sealed containers

Info

Publication number: GB1598426A
Application number: GB5211776A
Authority: GB
Original assignee: National Research Development Corp UK
Current assignee: National Research Development Corp UK
Priority date: 1977-12-13
Filing date: 1977-12-13
Publication date: 1981-09-23

Description

(54) LEAK-TESTING SEALED CONTAINERS (71) We, NATIONAL RESEARCH DE VELOPMENT CORPORATION, a British Corporation established by Statute, of Kingsgate House, 66 - 74 Victoria Street, London, S.W.1. do hereby declare the invention for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement: This invention relates to the leak-testing of nominally sealed containers and particularly to the detection of leaks in flamesealed glass ampoules of the kind used for the sterile storage of vaccine.

Visual inspection is adequate to detect an ampoule in which the seal is badly cracked or malformed but there remains a serious risk of bacterial contamination by way of any capillary leak which is invisible to the eye. In the most common method of inspection for such small leakage paths the ampoule is exposed to a dye in solution under pressure when the dye may penetrate visibly along the path. The method is effective over a period of tens of hours for a leak equivalent to a hole size of less than one micron if the structure of the capillary is favourable to liquid penetration.

In another method which has been proposed the space above the contents of an ampoule, normally air-filled, is filled with a highly mobile gas such as helium. After filling and sealing the ampoule is exposed to an external vacuum so that substantially the whole of the helium filling is lost through any leak of significant size. The ampoule is then exposed to the atmosphere and a high frequency spark is applied to the external wall of the ampoule, the presence of helium or of air being distinguished by judging the appearance of the internal discharge. This procedure is said to detect capillaries smaller than those found by dye testing owing to the high pumping rate of helium. The success rate of the procedure depends ultimately on the vigilance and discrimination of an operator working in a darkened room.

Sterility is so important, particularly in material for intraspinal injection, that ampoules are often re-tested before use in hospitals and a procedure is only satisfactory if it can be applied simply in such conditions and is not limited to the facilities of a manufacturer.

According to the invention in one aspect there is provided a method of testing for leaks in a nominally sealed container having at least a window portion which is optically transparent, the method comprising the initial inclusion in the container prior to sealing of a selected filling gas substantially at atmospheric pressure and the operations of exposing the sealed container for a predetermined period to an atmosphere including a contaminant gas at a pressure which is substantially higher than the filling pressure, exciting the emission spectrum of the filling gas and deriving a signal representing the value of intensity of the emission in a predetermined region of the spectrum, the filling gas and the contaminant gas being so selected that the signal level is strongly depressed in dependence on the quantity of the contaminant gas present in the container whereby the occurrence of the signal below a predetermined level indicates the presence of a leak path exceeding a predetermined size.

It will be apparent that the initial operation of the inclusion in the container of a selected filling gas prior to sealing may be carried out at a time and location remote from the carrying out of the subsequent operations.

The emission spectrum may be excited by inducing a high-frequency discharge in the filling gas.

The method is advantageously applied to containers in which the filling gas is substan tially at atmospheric pressure whereby gas contamination during exposure to atmospheric air is minimised.

The containers may be glass ampoules for sterile storage of materials, the manufacture of ampoules including the operations of introducing the material to be stored in the ampoule, admitting the selected filling gas to the ampoule to fill the remaining volume substantially at atmospheric pressure and sealing the ampoule.

According to a further aspect of the invention there is provided an apparatus for testing for leaks in a nominally sealed container, the free space in the container being filled with a selected gas substantially at atmospheric pressure and the container having at least a window portion which is transparent, the apparatus comprising means for exposing the container for a predetermined period to an atmosphere including a contaminant gas at a pressure substantially higher than the filling pressure, means for exciting the emission spectrum of the filling gas subsequent to such exposure, and means for deriving an output signal representing the value of intensity of the emission in a predetermined region of the emission spectrum, the filling gas and the contaminant gas being so selected that the signal level is strongly depressed in dependence on the quantity of the contaminant gas present in the container whereby the occurrence of the output signal below a predetermined level denotes the presence of a leak path exceeding a predetermined size.

The means responsive to the selected parameter may include an optical filter to transmit radiation only within a predetermined region of the spectrum and may further include means arranged to focus the radiation on to a photodetector.

In one embodiment of the apparatus there is further provided an input conveyor for automatically conveying containers in sequence from a loading position to a test position, means for presenting each container for test and means responsive to the output signal to cause the respective container to be conveyed to a first unloading position if the output signal lies above a predetermined value or to a second unloading position if the output signal lies below the predetermined value.

Means may be provided responsive to the output signal, when it lies below a predetermined value, to indicate the presence of a leak in the container.

Preferably means is provided for comparing the value of the output signal with a reference value representing the intensity of emission from a similar container free from contaminant gas, the predetermined value being a constant proportion of the reference value.

When the inert gas is argon the contaminant gas may be nitrogen, the predetermined region of the spectrum for such combination being within the range from 300 nm to 350 nm, and preferably close to 310nm.

Embodiments of the invention will now be described with reference to the accompanying drawings in which: Figure 1 represents graphically the effect of nitrogen contamination on emission intensity in the argon spectrum; Figure 2 represents diagrammatically an apparatus in accordance with the invention for use by an operator; Figure 3 represents diagrammatically a modified form of the apparatus of Figure 2; and Figure 4 represents diagrammatically an apparatus for automatic operation in accordance with the invention.

It is generally known in the analysis of mixed materials by spectroscopy that a study of the intensity of lines characteristic of a particular element (or compound) may be invalidated by interference from the spectrum of a different element and that experimental conditions should be arranged so that such interference regions may be avoided. In arriving at the present invention it has been appreciated that in a suitable gas the interference phenomenon can be employed to give a critical indication of the presence of a contaminant and that this result can be applied to the problem of leak-testing. An experimental investigation has shown that the intensity of selected lines in the emission spectrum from a spark discharge in argon at atmospheric pressure is strongly suppressed in the presence of nitrogen at a partial pressure of a fraction of a millimetre of mercury.Because nitrogen has a broad band spark spectrum it is to be expected that selected lines in the spectra of other gases will be similarly affected and also that contaminant gases other than nitrogen having many spectral lines will be effective as suppression agents.

Figure 1 shows the general form of the experimental result as a graph relating on a linear scale the observed intensity of a portion of the argon spectrum (on the vertical axis) to the percentage of nitrogen present (on the horizontal axis). A curve K which represents the observed intensity for a particular discharge disposition shows that suppression is virtually complete at an impurity level of 1% with the observed intensity increasing steeply below this level. It is convenient to select as a test criterion a level of impurity which would typically result from a few hours' exposure of a leaking ampoule having a filling gas at atmospheric pressure to a contaminant gas at a higher pressure.This value can be estimated by applying Poiseuilles Law to a notional leak equivalent to a hole diameter of 0.5cm in an ampoule of lmL gas volume. Calculation shows that for a particular wall thickness a time of about three hours at a pressure of three or four atmospheres is required to reach an impurity level of 0.25%. The gradient of curve K is very steep in the region of 0.25% impurity and the corresponding intensity has fallen to about 40% of the value for pure argon. A very sensitive test for small leaks is therefore provided by comparing the level of intensity observed for a suspect ampoule with a reference level observed for one containing substantially uncontaminated argon. An observed level which is only 50%or less of the reference level will then be a clear indication of a leak.

If the excitation conditions of the discharge should vary from one test to another the steep portion of the curve K may shift to a parallel position L but the intensity levels of interest will show substantially the same proportionality as before. It is expected that a small range of intensity about the reference level will be found as a result of the natural variation of discharge conditions between individual non-leaking ampoules but any observation which lies within this range will be distinguishable with certaintly from one lying at the 50% level. It is an advantage of the comparatively poor degree of discrimination demanded that the detection system may be made cheaply.It is considered unlikely that the sealing process will produce leaks of much smaller size than the postulated value of hole diameter equivalent to 0.5cm. The test can be made more stringent however by, for example, doubling the period of exposure to high pressure or by carrying out a further pressure test on any ampoule for which the observed intensity lies between 50% and some higher percentage of the reference level. Preferably however the simplicity of the measurement system should be preserved by employing only one or two selection levels.

Referring to Figure 2 an apparatus is shown for carrying out the form of test described above under the control of an operator. An insulating block 10 is arranged to hold a gas-filled ampoule 12 which has been exposed to a contaminant gas at high pressure. The ampoule 12 is inserted in a vertical position so that a metal clip 14 mounted from the block 10 makes contact with the wall of the ampoule 12 above the level of the contents over an extended arc which for convenience should not exceed 1800. The clip 14 is connected to a highvoltage, high-frequency supply from an oscillator or induction coil 16 to form a probe which excites a spark discharge in the gas-filling of the ampoule 12.The field gradient at an edge 18 of the clip 14 will normally be sufficient to cause breakdown of the gas but the clip 14 should not have any sharp points such that the gradient might puncture the glass wall of ampoule 12.

Light emitted from the discharge passes through the wall of ampoule 12 at the arcuate gap provided by clip 14 and is focused by a lens 20 into a beam which enters a photomultiplier 22. A filter 24 is interposed between the lens 20 and photomultiplier 22 so that only a selected region of the spectrum is admitted. For an ampoule filled with argon at atmospheric pressure and subsequently exposed to nitrogen at a substantially higher pressure the region of the spectrum close to 310nm is found to give good results. The effect of any oxygen which may also be present in the ampoule appears to be slight in this region. The value of the output current of the photomultiplier 20 is read on a meter 26 and compared by the operator with a value previously obtained for a leak-tight ampoule.The meter reading should be taken quickly and the discharge maintained in the ampoule for the shortest possible time. In the general case it is desirable to limit the duration of the discharge because ion bombardment of the inner faces of the glass wall of ampoule 12 may release gas which will interfere with the emission spectrum. There may also be particularly cases in which the contents of ampoule 12 will be damaged by unduly prolonged exposure to a discharge.

With reference to Figure 3, an alternative arrangement to that of Figure 2 is shown in plan view in which an intensity reference channel is provided. A reference ampoule 28 is mounted alongside the test ampoule 12 in a block 30 so that both ampoules are equally exposed to the high voltage source 16 by means of a two-lobed clip 32. The emission from ampoule 12 is observed as before via the lens 20 and photomultiplier 22 through a filter 24. Correspondingly ampoule 28 is observed via a lens 34, photomultiplier 36 and filter 38. A screen 40 is arranged between the two observing axes to confine the light emitted from each ampoule to its respective channel. The output from each photomultiplier is passed to a comparator 42 and any difference between the two values is indicated on the meter 26.

An apparatus of the form indicated by Figure 2 or Figure 3 would be very suitable for use in a hospital where the number of ampoules to be tested at any one time was small. Automatic operation would clearly be preferred for test following manufacture and a possible schematic layout of apparatus for such an application is illustrated in plan view in Figure 4. Ampoules 12 are loaded from a nitrogen pressure soak chamber 50 into cups 52 carried by a conveyor belt 54 which moves in indexed steps. Each ampoule 12 travels in an upright position with the constricted portion uppermost to a test position where it is presented for test while the conveyor 54 is stationary. The test procedure is essentially that described with reference to Figure 2.A block 10 with a clip 14 is situated at the test position adjacent to the conveyor 54 and the high voltage source 16 and photomultiplier 22 with lens 20 and filter 24 are disposed in accordance with Figure 2. The ampoule 12 is lifted from the respective cup 52 at the test position and held in the block 10 by a suction arm of electrically insulating material (not shown) which extends from an actuator 56. The discharge in the ampoule 12 is maintained for the shortest possible period necessary to obtain an output signal from the photomultiplier 22 at a comparator 58. A period of one second or less is thought to be sufficient to allow for stabilisation of the discharge.

The comparator 58 may hold for comparison a permanently stored value representing the intensity of emission from pure argon or for example a value obtained from a reference discharge at the beginning of each test load. The comparator 58 also contains a preset level representing the minimum acceptable test value as a proportion of the reference value. If the preset level is exceeded by the test value an output signal is passed from comparator 58 to the actuator 56 which causes the suction arm to lift the test ampoule from the block 10 and return it to the cup 52 on the conveyor 54. Ampoules carried to the end of the conveyor 54 then form the stock of accepted parts. If the test value is lower than the preset level a 'reject' signal from comparator 58 causes the suction arm to transfer the test ampoule from the block 10 to a second conveyor belt 60.

Ampoules carried by the belt 60 will be separated from the main stock for scrap or for retest. On completion of the cycle of movements required by the actuator 56 for each test the conveyor 54 is indexed to present the next ampoule at the test position.

The mechanical handling arrangements described with reference to Figure 4 are illustrative only. Other techniques available for the conveying and location of fragile articles may be employed in carrying out the invention.

The two-channel measurement system of Figure 3 may also be employed in the automatic conveyor arrangement of Figure 4 as an alternative to the single channel system described, the output from comparator 42 being then applied to the actuator 56.

The discharge in the reference channel may be excited only intermittently, the most recently obtained reference value being held by the comparator 42 for use in intermediate tests.

With general reference to the Figures 2, 3 and 4 the output from a comparator or direct from a photomultiplier may be applied to any suitable form of indicator, store or control device.

It will be apparent that the manufacture and testing of ampoules is intended to be co-ordinated so that ampoules will in practice be filled with the appropriately chosen gas when the drug or other content is introduced immediately before the ampoule is sealed. Since the gas filling is most conveniently admitted at atmospheric pressure (or substantially so) the rate of contamination of a leaking ampoule by atmospheric nitrogen during shelf-life is expected to be very slow.

For other kinds of container such as a glass or metal-and-glass vacuum envelope the gas filling may be introduced temporarily as part of the process of testing. The seal in this case could be made mechanically for ease of release following the test.

WHAT WE CLAIM IS: 1. A method of testing for leaks in a nominally sealed container having at least a window portion which is optically transparent, the method comprising the initial inclusion in the container prior to sealing of a selected filling gas substantially at atmospheric pressure and the operations of exposing the sealed container for a predetermined period to an atmosphere including a contaminant gas at a pressure which is substantially higher than the filling pressure, exciting the emission spectrum of the filling gas and deriving a signal representing the value of intensity of the emission in a predetermined region of the spectrum, the filling gas and the contaminant gas being so selected that the signal level is strongly depressed in dependence on the quantity of the contaminant gas present in the container whereby the occurrence of the signal below a predetermined level indicates the presence of a leak path exceeding a predetermined size.

2. A method according to Claim 1 in which the emission spectrum is excited by inducing a high-frequency discharge in the filling gas.

3. A method according to Claim 1 or Claim 2 in which the operation of deriving a signal includes optically filtering the emission to select the predetermined region of the spectrum and directing the filtered radiation to the surface of a photodetector.

4. A method according to any preceding claim in which the derived signal level is compared with a reference signal level representing a designated value for the uncontaminated filling gas to determine when the signal level is below the predetermined level.

**WARNING** end of DESC field may overlap start of CLMS **.

Claims

**WARNING** start of CLMS field may overlap end of DESC **. into cups 52 carried by a conveyor belt 54 which moves in indexed steps. Each ampoule 12 travels in an upright position with the constricted portion uppermost to a test position where it is presented for test while the conveyor 54 is stationary. The test procedure is essentially that described with reference to Figure 2. A block 10 with a clip 14 is situated at the test position adjacent to the conveyor 54 and the high voltage source 16 and photomultiplier 22 with lens 20 and filter 24 are disposed in accordance with Figure 2. The ampoule 12 is lifted from the respective cup 52 at the test position and held in the block 10 by a suction arm of electrically insulating material (not shown) which extends from an actuator 56.The discharge in the ampoule 12 is maintained for the shortest possible period necessary to obtain an output signal from the photomultiplier 22 at a comparator 58. A period of one second or less is thought to be sufficient to allow for stabilisation of the discharge. The comparator 58 may hold for comparison a permanently stored value representing the intensity of emission from pure argon or for example a value obtained from a reference discharge at the beginning of each test load. The comparator 58 also contains a preset level representing the minimum acceptable test value as a proportion of the reference value. If the preset level is exceeded by the test value an output signal is passed from comparator 58 to the actuator 56 which causes the suction arm to lift the test ampoule from the block 10 and return it to the cup 52 on the conveyor 54. Ampoules carried to the end of the conveyor 54 then form the stock of accepted parts. If the test value is lower than the preset level a 'reject' signal from comparator 58 causes the suction arm to transfer the test ampoule from the block 10 to a second conveyor belt 60. Ampoules carried by the belt 60 will be separated from the main stock for scrap or for retest. On completion of the cycle of movements required by the actuator 56 for each test the conveyor 54 is indexed to present the next ampoule at the test position. The mechanical handling arrangements described with reference to Figure 4 are illustrative only. Other techniques available for the conveying and location of fragile articles may be employed in carrying out the invention. The two-channel measurement system of Figure 3 may also be employed in the automatic conveyor arrangement of Figure 4 as an alternative to the single channel system described, the output from comparator 42 being then applied to the actuator 56. The discharge in the reference channel may be excited only intermittently, the most recently obtained reference value being held by the comparator 42 for use in intermediate tests. With general reference to the Figures 2, 3 and 4 the output from a comparator or direct from a photomultiplier may be applied to any suitable form of indicator, store or control device. It will be apparent that the manufacture and testing of ampoules is intended to be co-ordinated so that ampoules will in practice be filled with the appropriately chosen gas when the drug or other content is introduced immediately before the ampoule is sealed. Since the gas filling is most conveniently admitted at atmospheric pressure (or substantially so) the rate of contamination of a leaking ampoule by atmospheric nitrogen during shelf-life is expected to be very slow. For other kinds of container such as a glass or metal-and-glass vacuum envelope the gas filling may be introduced temporarily as part of the process of testing. The seal in this case could be made mechanically for ease of release following the test. WHAT WE CLAIM IS:

1. A method of testing for leaks in a nominally sealed container having at least a window portion which is optically transparent, the method comprising the initial inclusion in the container prior to sealing of a selected filling gas substantially at atmospheric pressure and the operations of exposing the sealed container for a predetermined period to an atmosphere including a contaminant gas at a pressure which is substantially higher than the filling pressure, exciting the emission spectrum of the filling gas and deriving a signal representing the value of intensity of the emission in a predetermined region of the spectrum, the filling gas and the contaminant gas being so selected that the signal level is strongly depressed in dependence on the quantity of the contaminant gas present in the container whereby the occurrence of the signal below a predetermined level indicates the presence of a leak path exceeding a predetermined size.

5. A method according to any preceding

claim in which the container comprises a glass ampoule, the manufacture of the ampoule including the operations of introducing the material to be stored in the ampoule, admitting the selected filling gas to the ampoule to fill the remaining volume substantially at atmospheric pressure and sealing the ampoule.

6. A method according to any preceding claim in which the filling gas is argon and the contaminant gas is nitrogen.

7. A method according to Claim 6 in which the predetermined region of the spectrum is narrowly distributed about the wavelength of 310 nm.

8. An apparatus for testing for leaks in a nominally sealed container, the free space in the container being filled with a selected gas substantially at atmospheric pressure and the container having at least a window portion which is transparent, the apparatus comprising means for exposing the container for a predetermined period to an atmosphere including a contaminant gas at a pressure substantially higher than the filling pressure, means for exciting the emission spectrum of the filling gas subsequent to such exposure, and means for deriving an output signal representing the value of intensity of the emission in a predetermined region of the emission spectrum, the filling gas and the contaminant gas being so selected that the signal level is strongly depressed in dependence on the quantity of the contaminant gas present in the container whereby the occurrence of the output signal below a predetermined level denotes the presence of a leak path exceeding a predetermined size.

9. Apparatus according to Claim 8 in which the means for exciting the emission spectrum comprises a high-frequency electrical source arranged to induce a discharge in the filling gas.

10. Apparatus according to Claim 8 or Claim 9 including an optical filter to transmit radiation only within the predetermined region of the spectrum and photodetector means responsive to such radiation to produce an output signal representing the intensity of the emission.

11. Apparatus according to any of Claims 8 to 10 including means for comparing the level of the output signal with a reference level related to the level characteristic of an uncontaminated gas filling and means responsive to the difference between the level of the output signal and the reference level to indicate the presence of a leak when the difference corresponds to an output signal below the predetermined level.

12. Apparatus according to any of Claims 8 to 11 including an uncontaminated gas filled reference container similar to the test container, like means for exciting an emission spectrum from each container and means for deriving the reference value for the emission intensity from the reference container.

13. Apparatus according to any of Claims 8 to 12 including means for automatically conveying containers in sequence, following exposure to the contaminant gas, means for presenting each container to the means for exciting the emission spectrum, and means responsive to the output signal to transfer the container to a first unloading position if the output signal lies above a predetermined level or to a second unloading position if the output signal lies below the predetermined level.

14. Apparatus according to any of Claims 8 to 13 in which the selected filling gas is argon, the contaminant gas is nitrogen and the predetermined region of the spectrum lies within the range from 300 nm to 350 nm.

15. Apparatus substantially as hereinbefore described with reference to and as shown in Figure 2 or Figure 3 or Figure 4 of the accompanying drawings.