US3057996A

US3057996A - Method and apparatus for operating an analytical mass spectrometer with a getter-ion pump

Info

Publication number: US3057996A
Application number: US59901A
Authority: US
Inventors: Ernest W Boyer
Original assignee: Continental Oil Co
Current assignee: ConocoPhillips Co
Priority date: 1960-10-03
Filing date: 1960-10-03
Publication date: 1962-10-09
Anticipated expiration: 1979-10-09
Also published as: GB991902A; BE622048A

Description

Oct. 9, 1962 Filed Oct.

E. W. BUYER METHOD AND APPARATUS FO R OPERATING AN ANALYTICAL MASS SPECTROMETER WITH A GETTER-ION PUMP MECHANICAL PUMP 3 Sheets-Sheet 1 I VACUUM GUAGE SOURCE ION PUMP INVEN TOR.

ERNEST W BUYER A TTORNE Y Oct. 9, 1962 E. w. BOYER 3,057,996

METHOD AND APPARATUS FOR OPERATING AN ANALYTICAL MASS SPECTROMETER WITH A GETTER-ION PUMP Flled Oct. 3, 1960 3 Sheets-Sheet 2 EXHAUST RATE FOR ETHANE IOO l I I l 80 L E G E N o D/Fl-T PUMP VAC. 101v M/e 28 g 60 MA? 30 I X a 40 Q.

O 2 4 6 8 IO l2 l4 l6 I8 20 TIME-MINUTES DIFFERENTIAL PUMPING EFFECT ON Hg-He BLEND 700 Y 500 I 2 400 L E e E N o i D/FE PUMP VAC. /0/\/ g 300 M/e 2a M/e 30 20o 0 I l I I l I l l l O 2 4 6 8 IO l2 l4 l6 I8 20 TIME-MINUTES IN ENTO ERNEST W BUYER A TTORNE'Y 3,057 LYTICAL 3 Sheets-Sheet 3 E. W. BOYER TUS FOR OPERATING AN ANA WITH A GETTER-ION PUMP MASS SPECTROMETER METHOD AND APPARA EXHAUST RATES FOR METHANE-BUTANE BLEND ATTORNEY Oct. 9, 1962 Filed Oct. 3. 1960 m 4 m o M W N V w m m M M m V n \.N D 3 N E W Y G U B E n. B L H W I D n 2 m and M AM 4 E C C M. n n ITI n a. ll a A on n 1- .2 \\\\I\. otuutl L\ \Iq lunuo-o a ooo 9 8 6 5 FIG. 5

nited Sates Paten 3 057 996 METHOD AND APPAriArus FOR OPERATING AN ANALYTICAL MASS SPECTROMETER WETH A GETTER-ION PUMP Ernest W. Boyer, Ponca City, Okla, assignor to Continental Oil Company, Ponca City, 02:122., a corporation of Delaware 7 Filed (let. 3, 19:39, Ser. N0..59,%l 7 Claims. (Cl. 250-413) This invention relates to the adaptation of an ion vacuum pump to ananalytical mass spectrometer.

This application describes the method for connecting an ion pump with an analytical mass spectrometer. A more detailed description of the novel hardware used to connect the ion pump tothe analytical mass spectrometer is contained in an application entitled Method and Apparatus for Connecting a Getter-Ion Pump to an Analytical Mass Spectrometer by ErnestW. Boyer, Harrell T. Ford, and Ernest E. McKelvey, filed concurrently with this application and assigned to the same assignee.

Since the development of the analytical mass spectrometer the diflusion pump has been the source for obtaining the high vacuum required for its proper operation. While the diffusion pump has been more than adequate in obtaining a satisfactory vacuum for the proper performance of the spectrometer, the industry has been plagued with the length of time that the pump requires to obtain an adequate vacuum, especially with condensable light hydrocarbon. Maximum or minimum nitrogen levels in the cold traps fail to solvethe problem. Even when a programmed nitrogen level was used, the solution was far from satisfactory. The problem is further compounded by a compound such as ethane which is condensed to a volatile liquid under trap conditions and is in equilibrium with cold trap and source region. Under these conditions, it causes a variable background which cannot be eliminated by simply subtracting the background as measured before recording a sample spectrum.

The ion pump ,on the other hand permits the elimination of the troublesome cold trap and thus permits direct coupling between the ion pump and source. The application of an ion pump to a mass spectrometer is not new; however, the use was limited solely to a spectrometer which was analyzing simple systems. Since the ion pump is species selective, thatis, will pump one entity faster than another, it was determined to be impossible to apply the pump to an analytical mass spectrometer that obviously requires the pumping of widely different compounds. The molecules having differentsizes and atomic structureswould be operated on bythe eleetronicstream of the pump differently. The pumping rate would, of course, be determined by notonly the molecular size, that is, how large an area the molecule presented. to the electronic stream, but also the electronic configuration. However, the ion pump creates new problems since it is species selective in its pumping rate. Thus, the ion pump will pump various entities at a different rate, thereby upsetting the ratio of the entities at the sourceand resulting in errors in the analysis of the mixtures under test. The ion pump has a further difficulty in that. if the entire system requires venting for any reason, such as maintenance on the source, the system must be opened to the atmosphere. When the system is again evacuated, the ion pump, is roughed to approximately 20 microns; when this pressure is reached, .the power supply to the ion pump is energized. The nature of the pump is such that at operation at these pressures (about microns), the pump heats up. At elevated temperatures, for some reason not completely understood, chemical entities previously pumped are released from surfaces within the pump. These chemical entities so released can migrate ice to and contaminate the spectrometer source. Thesedeposited entities will result in erroneous analysis of samples under test. Entities, as used in this specification, shall mean any particles or combination of particles (molecules, atoms, etc.) in a charged or uncharged state.

It is, therefore, an object of this invention todisclose a method for connecting an ion pump to an analytical mass spectrometer in a manner that will eliminate the species selectiveness of the ion pump.

It is a further object of this invention to provide a method for preventing entities from migrating from the pump back to the source and thereby upsetting the composition of the material under test. 7

It is a further object of this invention to disclose a method for modifying the source geometry of an analytical mass spectrometer so that its conductance will be sufficient to prevent species selectiveness of the ion pump.

This invention primarily features an analytical mass spectrometer which has .a source geometry which controls the conductance of the system sufficientlyso that the ion pump will not operate upon the source in a species selective manner. The source conductance is controlled by increasing the resistance to the flow of gases from the leak to the pump by inserting bafiles or other obstruction means near the..leak and. between the leak and the pump. The bafiles are preferably mounted in the source.

Further objects, features, and advantages of the invention will become apparent from the followingdescription and claims when read in view of the accompanying drawings, in which: i

FIG. 1 is apartial sectional drawing of-the hardware used to couple the ion pump to the spectrometer source;

1G. 2 is a graph showing the improvement of the exhaust ratefor a condensable lighthydrocarbonj FIG. 3 is a graph showing the comparison inthe species selectivity between the differential pump .andthe 'io'n pump for two radically different molecules; i

.FIG. 4 is a graphshowing the improvement in the exhaust rate of the ion pump underthe diffusion-pump using radically different hydrocarbon as best sampled; and

FIG. 5 illustrates other diaphragms used to restrict flow of entities. V

Referring to FIG. 1, an ion pump 10 is shown con nected' to an analytical mass spectrometer sourcefll through air-tight hardware which comprisesa cylindrical tube 12 which has a flange 13 adapted to mate with an output flange 14 of the ion'pump.

Flanges

13 and 14 are held together by any suitable means such as a plurality of bolts 15. A high vacuum valve 16 is connected to the end of cylindrical tube 12 opposite flange 13 by any suitable means such as shoulder'17. High vacuum valve 16 may be any suitable valve capable of maintaining an air-tight seal during operation in either a closed or open position. The type of valve 16 used is coinmonly available and briefly incorporates a valve seat 18, a valve head 19 adapted to seal the end of cylinder 12 when the valve head and seat are mated. A Itlireaded valve stem 20 passes through valve end 21 and has one end rotatably attached to valve head 19v and the other end rigidly attached to a suitable handle or nob 2i. In order to maintain an adequate seal at all times betwen the valve end 21 and valve head 19, a bellows 23 is sealably attached to the top of the valve and,to the periphery of the valve head. Thus, any air escap ing into the inner portion of the valve around the valvestem threads is confined within the bellows and does not enter the inside of cylindrical tube 12.

The output 24 of valve 16 isconnected to a cylindrical portion 25. Portion 25 includes a pair of pipes 26 and 27, respectively, attached to its periphery. The first input 26 is connected to a vacuum gauge measuring means 28 and the second input 27 is connected to a valve 29. The output of cylindrical portion is connected through a cylindrical portion 30 to spectrometer source 11. For convenience, cylindrical portion 30 is here shown to contain a labyrinth comprising a plurality of battles 31 supported within cylindrical portion 30 by any suitable means. Also connected to cylindrical portion 12 is a second valve 32 which is shown to be a valve similar to valve 16 and, therefore, will not be described. The size of valve 32 permits rapid evacuation of the chamber Within cylindrical tube 12; however, it is obvious that other types of valves may be used for either valve 32 or valve 16 providing they are sufiiciently air-tight during their operation in either the open or closed position. Connected to the outlet of valve 32 is a pipe 33 which is connected to the exhaust exit 34 of a mechanical pump 35. A pipe 36 is connected between the outlet of valve 29 and exhaust exit 34 of mechanical pump 35. A valve 37 is connected between pipe 36 and the air to permit venting of the spectrometer source in the event that Work must be done on the unit when the mechanical pump is connected. Mechanical pump and pipe 36 may be disconnected by including a

disconnect

38 and 39, thereby freeing the pump for other uses in the laboratory. Annular rings 42 and diaphragm 43 are rigidly attached within cylindrical tube 12 and aid in the elimination of secondary emission entities.

OPERATION The elimination of the diffusion pump and its associated cold trap and the substitution therefor of an ion pump has lead to a great many unpredictable advantages which will be pointed out in further portions of the specification. While it was obvious to one skilled in the art that an ion pump might be incorporated into a mass spectrometer that was used to monitor a simple system, it was equally obvious to those skilled in the art prior to my discovery that an ion pump could not be incorporated into an analytical mass spectrometer and used to analyse a plurality of species simultaneously since the ion pump pumped each species in a selective manner. Both the manufacturer of the analytical mass spectrometer and the manufacturer of the ion pump concurred in the opinion that the two could not be incorporated. I found, however, that by controlling the conductance of the system by the source design to a point where the pump could no longer pump each species selectively that the electronic ion pump could satisfactorily be incorporated in an analytical mass spectrometer.

Referring to FIG. 1, with valve 16 in an open position ion pump 10 evacuates cylindrical tube 12,

cylinder portions

25 and 30, and spectrometer source 11; both

valves

32 and 29 under normal operation are operated in a closed position thereby preventing the entry of air into the evacuated system. Inserted within tube portion 30 is shown a plurality of bafiles or diaphragms 31. These bafiles show one method of increasing the source conductance so that the pump can not pump each species in the analytical spectrometer source at a different rate. Other methods of controlling the source conductance are possible, as for example, a plurality of right-angle bends. FIG. 5 illustrates other forms of restriction such as a slit or a plurality of holes 61 which may be used to impede the flow of gases from the leak to the pump 10. The amount of resistance imposed by the battles must exceed a minimum amount which is sufiicient to prevent the pump from operating on the entities in the system relatively, and by a maximum amount determined by the pressure of the source.

The minimum resistance can be determined by placing, under test, two dilterent compositions such as methane and butane or hydrogen and helium, in a predetermined blend and scanning the source composition over a period of time. A decrease in the ratio of the compound from the original mixture would indicate that the resistance was too low and that additional resistance was required. The method for determining if an excess resistance has been inserted in the line is to monitor a sensitive peak contributed to by more than one compound in a mixture and establishing that this peak does not increase with increased time of flow through the leak. In actual practice it is best to keep the resistance of the line to a level as near the minimum as possible, since the greater the resistance the longer the pump-out time required.

When for some reason the source must be shut down and the system vented, valve 16 is closed. This will permit continuous operation of pump 10, and will maintain cylindrical tube :12 in a highly evacuated state, thereby greatly reducing the time required to pump-down the system when the work on the source is completed. If mechanical pump 35 is connected in the manner shown in the drawing, that is, pipe 36 is connected to valve 29, valve 37 must be opened in order to let air into the system. Once the work is completed, valve 37 is closed and the mechanical pump energized. Operation of the mechanical pump will then partially evacuate the spectrometer source and

cylinder portions

25 and 30, thus clearly reducing the work required by the ion pump, and likewise, greatly increasing the speed that the system can be totally evacuated once the ion pump is connected to a spectrometer source. When vacuum gauge 40 indicates that the vacuum is as well as can be obtained by the mechanical pump, valve 29 is closed and valve 16 opened. The remaining air in

tube portions

25 and 30 and spectrometer source 11 will be brought to operating vacuum. If for some reason the ion pump must be changed or replaced thereby requiring that cylindrical tube 12 be purged, rapid evacuation is obtained by connecting mechanical pump 35 through pipe 33 to valve 32. When valve 32 is opened and mechanical pump 35 actuated, the system is rapidly brought to a near operating rate by the mechanical pump. When this occurs, valve 32 is closed and the ion pump energized and the system brought rapidly to its operating vacuum.

Valve 16, while operating as a valve, also presents a unique feature not apparent. When, for example, the analytical mass spectrometer has been used to test samples of hydrocarbon and ion pump 10 has been used for a period of time, hydrocarbon molecules will become imbedded in the titanium surface on the inside of ion pump 10. If for some reason, as previously explained, the ion pump is disconnected or cylindrical tube 12 is filled with air, these molecules must then be evacuated when the pump is again re-energized. When the pump is initially started, the elevated temperatures which develop cause a release of entities previously pumped which may migrate up cylindrical tube 12 and back to source 11. This emission of migratory molecules if permitted to reach source 11 would obviously introduce errors during analysis. Further secondary emission entities released by the pump during normal operation would enter the source and contaminate same if the pump and source were in line of sight with each other. This, however, is eliminated in my unique construction by placing cylinder 25 at right angle to cylinder 12. Thus, as secondary emission entities leave ion pump 10, they will be inhibited from entering cylinder 25 by the right-angle bend. Thus, as the molecules travel up cylindrical tube 12, they will strike valve head 19 causing them to become grounded or deflected and will migrate back to pump 10. It is obvious that if valve 16 is not included in the system that a battle or diaphragm or series of annular rings or baffles could be inserted within cylindrical tube 12 and accomplish the same results. Further, cylindrical tube 12 may be bent at right angles and thereby result in the grounding or de fleeting of the charged or uncharged entities, the end result being, of course, to prevent the stray entities from migrating back to the spectrometer source and causing an interference in analysis,

While annular ring 42 and diaphragm .43 are shown in cylindrical portion 12, they may be eliminated if valve 16 provides sufficient isolation for the sample under test. If valve 16 is eliminated, rings 42 and diaphragm 43 would be necessary to eliminate the secondary emission entities.

One of the main laboratory problems with the mass spectrometer using a diffusion pump and cold trap has densable to a volatile liquid under trapped conditions and is in equilibrium with the trap and source region,

while the curvesfor the diifusionpumpduring the early portions of pumping are actually not parallel indicating that the modified analytical spectrometer incorporating the ion pump has actually been improved in its operation and accuracy.

FIG. 4 illustrates the exhaust rate for .two radically different hydrocarbons. The graph clearly illustrates that in both cases the ionpump with the modified source geometry exhausted the methane and butane at a much more rapid rate than did the diffusion pump and associated cold trap. Assuming that when a peak height of 5 divisions was reached that the analytical spectrometer was ready for use, a waiting period of less than lminute was all that was 'requiredto sufiiciently exhaustboth the methane and butane using the ion pump While a period of approximately 1% to 2 minutes was required for the diffusion pump.

Shownin the following chart is a plurality of compounds and their corresponding exhaust rate.

Exhaust Rate for Miscellaneous Organic Compounds Most Orig. Pump Out Time In Minutes Compound Sensitive Number Peak of Divisions 1 3 5 7 10 15 27 2,100 1.7 28 5, 120 10. 0 4. 0 43 6,081 3.8 86 810 1.4 43 3,070 19.0 .1.0 0.7 78 7, 400 14. 9 2. 9 1. 5 91 2, 700 30. 8 5. 0 1. 9 1. 3 56 3, 430 1.8 43 6,000 9.7 1.8 0.3 43 7, 500 16. 0 3. 5 0.8 31 5,000 4.2 0.8 31 2,835 8.9 2.4 0.8 94 4, 000 236. 1 46. 8 18. 7 41 3, 430 84. 4 30. 0 16. 8 Carbon Tet 47 1,167 7.0 2. 3 350 C. Inlet System:

Octadecene-l 41 9, 100 500 159 84 Octadecane i, 1 873.1 2754;

, 7 29 Octacosane 394 11 20.0 10.4 5.2 Octadecylbenzene 43 4, 480 600. 0 184. 0 106. 9 Naphthalene 128 8, 700 80.2 19. 8 15.0 Anthracene 178 55 13. 1 6.0 3. 0 i% 9288 it? 32* 1%? Octadecimol 252 209 4. 4 3. 0 1. 1

the ethane therefore would contrrbute to a variable back- In each case the most sensrtlve peak was selected in ground which could not be eliminated by simple subtraction of the background as measured before recording the sample spectrum.

FIG. 2 shows the problem of exhausting ethane with the usual diffusion pump. After a period of six minutes it can be seen that there were still at least divisions peak height of M/e28 remaining and 9 divisions peak height of M e30 remaining when a diffusion pump and cold trap were used. However, with the electronic or ion pump at the end of 6 minutes, a negligible number of divisions were read for both M/e28 and M e30 which indicated that sufficient ethane had been removed to give satisfactory operation with the ion pump. Since the usual time required to sufficiently evacuate the ethane was 30 minutes, a reduction from 30 minutes to 1 minute resulted in a substantial reduction in operating time.

FIG. 3 illustrates the complete lack of differential pumping when using the ion pump and the modified source. Two radically difierent molecules were selected, namely, hydrogen and helium. The selection was for the purpose of determining whether the pumping rate would be different for molecules having radically different sizes and different charges. A differential pumping effect would be evident by a change in slope between the plot of the hydrogen and helium molecules for either the diffusion pump or the electronic ion pump. Again using the peak height and a 50/50 mixture of the hydrogen and helium molecules, the graph clearly shows that the curves for the ion pump are substantially parallel order to provide more accurate information. The first portion of the table tabulates the pump-out time for saturates, aromatics, cyclics, ketones, and alcohols which were selected as a representative sample of miscellaneous organic compounds to illustrate the times required to evacuate the system under normal analytical mass spectrometer temperatures of C. The second portion of the table illustrates the pump-out time for a high temperature analytical mass spectrometer (350 C. inlet system). A plurality of heavy olefins, saturates, aromatics, and alcohols were selected. In the second portion of the chart it is obvious that a longer pump-out time is required; however, a portion of the extended time is a result of the fact that the mass spectrometer under use was sample side limited. That is, the method of exhausting the sample side was slower than the method of exhausting the source and thus a certain amount of pressure backed into the source side resulting in a longer pump-out time. The limitations on the sample side were necessitated by the high temperature modifications. Thus, an improvement on the high temperature pump-out system (valve sizes, etc.) of the analytical mass spectrometer would result in a much more rapid pump-out time for the heavy organic compounds. Thus, a system has been described and sufficient data included to establish that an ion pump can be operated with an analytical mass spectrometer if the source geometry is modified sufficiently to make the pump non-species selective and that the source could be modified by incorporating a plurality of '7 diaphragms and annular rings in sufficient number for accomplishing the above requirements.

While the preferred embodiments disclose circular tubing, it is obvious to one skilled in the art that other forms and shapes of tubing may be employed without departing from the spirit and scope of this invention and that the principle of operation of the apparatus regardless of its general physical appearance would represent the true invention.

Although this invention has been described with respect to particular embodiment thereof, it is not to be so limited, as changes and modifications may be made therein which are within the spirit and scope of the invention as defined by the appended claims.

I claim:

1. In an analytical mass spectrometer having a source, a selective pumping ion pump, and hardware connecting said pump to said spectrometer source, an improvement comprising, interposing an obstruction in the proximity of said source whereby said source conductance is increased thereby rendering said pump non-species selective.

2. A device as described in claim 1 wherein said obstruction comprises a plurality of diaphragms having openings therethrough whereby said gases are caused to change direction and velocity.

3. A device as described in claim 1 wherein said ohstruction comprises a diaphragm having a plurality of small openings therethrough.

4. A device as described in claim 1 wherein said obstruction is mounted substantially within said source.

8 5. An analytical mass spectrometer comprising in combination a source having an output, an ion pump having an output, means connecting the output of said ion pump to the output of said source, and obstruction means interposed within said connecting means between said pump and said source, said obstruction means adapted to control said source conductance.

6. An analytical mass spectrometer comprising in combination an analytical mass spectrometer source having an output, an ion pump having an output, and a pipe connected between said ion pump output and said spectrometer source output, said source additionally comprising means mounted therein which is adapted to control the conductance of said source by an amount sufficient to make said ion pump substantially non-species selective.

7. In a mass spectrometer having a source, a selective pumping ion pump, and hardware connecting said pump to said spectrometer source, an improvement comprising, interposing an obstruction in the proximity of said source whereby said source conductance is controlled thereby rendering said pump non-species selective.

References Cited in the tile of this patent UNITED STATES PATENTS 2,769,912 Lupfer et al. Nov. 6, 1956 2,880,323 Reinecke et al. Mar. 31, 1959 2,967,012 Connor Jan. 3, 1961 UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Patent No 3 057096 October 9 1962 Ernest Eh Boyer appears in the above numbered pat- It is hereby certified that error (1 Letters Patent should read as ent requiring correction and that the sai corrected below.

Column 2,, line 41., for hiurlclenr read versus line 42 for "hydrocarbon" read hydrocarbons same line 41:2 for 'best" read test 0 Signed and sealed this 21st day of May 1963.

(SEAL) ERNEST W. SWIDER DAVID LADD Attesting Officer Commissioner of Patents Attest: