US2400557A - Analytical system - Google Patents

Description

. or vapor form.

' ion collector.

Patented May 21, 1946 UNITED sTA'rfi-:SCPATENT OFFICE ANALYTICAL SYSTEM Reed C. Lawlor, Alhambra, Calif., assignor to Con. solidated Engineering Corporation, Pasadena, Calif., a corporation of California Application July 31, 1942, serial No. 453,128 .3 ,claimsl (Cl. 11s-1s) My invention relates to mass spectrometry, and more particularly to the analysis of a chemical mixture containing a plurality of components which upon ionization may form ions having the same mass-to-charge ratio. More-specically, my method makes possible the analysis of such a mixture by taking into account the fact that such ions are usually produced in different pro- .portionsv from such components ionization voltages.

This application is acontinuation-in-part of copending patent application, Serial No. 378,636, iiled February 12, 1941.'

When analyzing a mixture (which originally may be gaseous. liquid or solid) with a mass at different spectrometer, the mixture is introduced from a.

sample region into an ionization region in gaseous 'In the ionization region, molecules of the mixture are ionized in characteristic'ways under controlled conditions by subjecting them to the ionizing action of particles such as electrons. Such electrons are commonly directed by the action of `an electric field into the ionization region where they encounter and ionize molecules -of the mixture. The ions formed are then withdrawn into an analysis region where ions of different mass-to-charge ratios are 'segregated into beams which may be detected by successive'focusing thereof upon an As the 'beams are successively detected at the collector, the'beam intensities may either be measured directly with a suitable indicator or else automatically and permanently recorded prior-to measurement. The peak -intensities of the ion beams are representative of either the amounts of the withdrawn ions or the ratesl of formation of the respective ions. Such a record or set oi measured peak intensities forms a mass spectrum.

For simplicity, the ionization of a sample by particles having a given amount .of energy corresponding to the kinetic energy of electrons which have been accelerated by somepredetermined voltage will be referred to simply .as ionization of the sample or component at that voltage.-

Either before'or atterra mass spectrum of a mixture has been "obtained, reference samples containing different proportions of the components that may be present in the mixture or substances corresponding chemically thereto are similarly subjected to analysis in a mass spectrometer under substantially the same ionizing conditions to which the mixture was subjected, and corresponding mass spectra obtained for the reference samples. Usually, though not necessarily, the reference samples are relatively pure samples of the respective components that may occur-.in the mixture.

The composition of the mixture is then determined by comparing the mass spectrum of the mixture with the mass spectra oi the reference samples. Such a comparison is most simply carried out by a mathematical procedure involving the comparison of the peak intensities o! the beams forming the spectra.

' In practice, the analysis of such a mixture is greatly facilitated if the intensity of each peak occurring in the mixture spectrum represents the sum of the intensities of corresponding peaks that would be obtained in mass spectra of the separate components ii present alone. A method for producing this eiect, which is known as linear superposition, is disclosed and claimed in copending patent application, Serial No, 320,802, illed by Harold W. Washburn and Daniel Dwight Taylor. Briefly, this method involves maintaining the sample in a sample chamber homogeneous at al1 times during analysis, owing the components from the sample chamber in to an ionization chamber through a gas inlet at mutually independent rates, ionizing each component in proportion to its partial pressure in the ionization chamber and independently of the amounts of the other components there, and pro# viding such pressure conditions in the mass spectrometer that collisions between ions withdrawn from the ionization chamber with any molecules, either in the ionization chamber or in the analyzing chamber, are relatively infrequent.

When conditions suitable for achieving linear superposition during the analysis of a mixture are obtained, the intensity of any peak occurring in the mass spectrum of the mixture due to ions of mass-to-charge ratio m produced at a voltage V may be represented by the following equation:

e X=the partial pressure of component i in the vso mixture and VP"=a sensitivity coefficient representing the efficiency of the mass spectrometer in producing ions of mass-to-charge ratio m obtained from component i by particles having an ionization energy of V electron volts.

made by obtaining mass spectra of known mixtures approximating the unknown in composition and comparingA the spectrum of the unknown mixture vwith the spectra of the known -mixture by the interpolation method which is more fully explained in copending patent application, Serial No. 324,950, led MarchA 20, 1940, by Daniel Dwight Taylor.

For simplicity, I shall discuss the application of my inventionunder conditionsv such that linear superposition is maintained..

The amount of ions of mass-to-charge ratio m `originating from a given component i is found to vary over a wide range with changes in the ioniz- According to the present invention, I ionize the mixture at voltages at which such ions arev formed in diierent proportions from different components, measure the amounts of ions of such mass-- to-charge ratio formed at each voltage, and. utilize those measurements in determining the com-` position of the mixture.

My invention may be applied, forexample, to the analysis of mixtures containing hydrocarbon isomers, which generally produce ions of the same mass-to-charge ratios, and to mixtures con?v taining other components of the .same molecular weights.

Accordingly, the principal object of my invention is to provide an improved method for analyzing chemical mixtures with' a mass spectrometer. .My invention possesses numerous other objects and features of advantage, some fwhich together with the foregoing will be set forth in the following description embodying and utilizing my novel method. It is therefore to be understood that my invention is applicable to analyses of a variety of mixtures and to other mass spectrometers -of various types or which may utilize other kinds of ionizing particles, and that'I do not limit myself, in any way, to the analyses, to the apparatus, or to the ionizing particles, of the present application, as I may adopt various other modications of my invention utilizing the method, within the scope of the appended claims. My invention may be more readily understood by direct reference to the drawings in which:

Fig. 1 shows a general organization of a mass spectrometer to which my method may 4be applied.

Fig. 2 is a schematic drawing partly showing a section taken on a line 2-2 of Fig. 1 of part of the mass spectrometer including the ionization chamber and the ionization energy controls.

Fig. 3 is a graph showing ionization curves of a mixture of two components and of the components alone.

Referring to the drawings:

In Fig. 1, I have shown a sample'chamber I connected to an ionization chamber 3 through a which a sample may be admitted; A pressure gauge, for example a McLeod 'gauge I 3, isconnected to the sample chamber, A valve I5 is provided in the line 5 which connects the sample chamber to the ionization chamber.

A s illustrated in Figs. 1 and 2, electrons emitted from a. heated filament I1 are directed in a beam I8 through aperture I9 in the electron beam intensity control electrode 2|, through aperture 2 3 in electronaccelerating electrode 25 which is formed by -part ofthe wall of said ionization chamber 3, and through aperture 21, in the OPPO- site portion of the wall of said ionization chamber, said beam being directed along the line perpendicular to the face of a magnetic pole 29 by the combined'action of jthe magnetic -eld indicated by arrow 3I in Fig. 2, which `field is directed downward perpendicular to the plane of the trodes 2I and .25.

drawing in Fig.1. and electric fields parallel to the magnetic eld. The electrons passirigthrough said apertures impinge on electron-*catcher 32 electrically connected to electrode 25.' The'electric iields are provided by potentials applied from voltage supply circuit 33 to filament I'I and elec- Said voltage supply circuit includes .means for varying the relative potentials between said iilament and said electrodes.

One way of accomplishing this variation is tp connect the lament to the negative ends of a. pair of parallel connected rheostats 35 and'31 through whichcrrents ilow from a battery `v39,

and to connect said Aelectrodes-2| and 25 to electron beam intensity. 'The amounts of ions produced are dependent on the energy of the electrons in beam I8. t

The positive ions formed areaccelerated toward first slit electrode 4I byaction of a small electric potential which maintains said iirst collimating tube 5 containing a gas inlet I which may be in the form of a capillary leak or restricted orice. A gas sample to be analyzed is introducedwhen the mass spectrometer is in condition for operation into the ionization chamber through. said inlet. A T shaped conduit I0 is connected to the sample chamber I. One branch of the T is con nected to a vacuum pump through a valve II.

slit electrode negative with respect to a pusher electrode 43 on the opposite sidetof said electron beam. Some'of the accelerated ions pass through a narrow slit 45 in said iirst collimating elec'- trode 4 I and are thereupon acceleratedby a large negative potential between said first collimating slit electrode 4I and second collimating slit electrode 41. Some of the accelerated ions then pass through a second-slit 49 in said second 'collimating slit electrode.

The ions thus pass through slit 49 at a high velocity. Thepath they 'take,however, is not straight but curved, owing tov-the action of the justnienaeither'of the magnetic neia-,orrofsthe accelerating voltagaor of both, ions- M-Tan'yle' sired massfto-char'ge ratio' can v'be made'to'f'ollow a predetermined radius of curvature varid *thus be focusedat exit sli t 5 I positioned in* front-'ot an ion collector 53. AWhaha containing ions of' predetermined mass-to-chargeIA ratio The other branch of the T has a valve 9 through 75 impinges on collector 53, a corresponding ion cur- 3,609,557 rent is produced, which current may be measured by a suitable vacuum-tube amplier it and gal; vanometer-B'l, or recorder (not shown).

In order to obtain a mass spectrum of a sample, beams of different mass-to-charge ratios are successively moved past said collector slit 5| by changing voltages supplied to said pusher electrode 43, and said collimating slit electrodes 4I and 41, by electrical connection to suitable points of a potential divider t in ion beam deflection control circuit Si. This may be done most readily by changing the voltage across said divider Il by adjustment of a sliding potentiometer contact 6l connected to one end thereof. When beams comprising' ions of different corresponding mass-to-charge ratios are successively focused at said exit slit i. ions in the different beams are successively discharged at the ion collector 53, the respective intensities. of corresponding ion currents are read on galvanom-v which is a, function of the ionization voltage.

From a knowledge of these cracking patterns and their variation with ionization voltage, to-

gether with a knowledge of isotope ratios and packing fractions of the ions recorded, the quantities ol the variousvgases and vapors present in the unknown sample can be determined. However, it is usually unnecessary to take all these factors into account in a single analysis.

According to the present invention. I take advantage of the fact thatthe relative amounts of a given type of ion produced from diiferent components generally varies with the ionization p0- tential. Thus, for example, I may determine the composition of a normal butane-isobutane mixture by measuring the number of ions of a single mass-to-charge ratio derived from such a mixture at two diierent ionization voltages. By comparing these results with. measurements obtained from pure components underlow pressure conditions such as those already hereinabove described, I may compute the quantities of normalbutane and isobutane from equations similar to equation 1. My method may. of course, be applied to mixtures of gases other than hydrocarbons.

In order to determine how the intensity of any V particular ion beam varies with the energy of the electrons in the electron beam, I adopt one of two procedures. 4-In the rst, the total accelerating voltage to which ions are subjected may be set at a predetermined value by adiustment of the total potential between pusher electrode 43 and collimating slitelectrodes 4| and 41 by suitable positioning of the slide contact 01 on potentiometer 69 in the deflection control circuit Gi, thereby focusing ions of any predetermined mass-to-charge ratio desired at the exit slit 5l. With the mass spectrometer adjusted to detect ions of predetermined mass-tocharge ratio, the energy of the electrons in the electron repeated for each peak o: interest in each component of interest. In the second procedure I l set contact 38 successively at different points of rheostat Il to establish different values of electron accelerating voltages as indicated by a voltxneter I6, and obtain a mass spectrum of the sample under investigation at each of these voltages for each component in which I am interested, by the method hereinabove described.

In both of these methode the voltage between the filament Il and the electron beam intensity control electrode 2i is maintained constant and preferably at a value such that the'total electricai current'borne by electron beam i8 is constant.

In Fig. 3, I have plotted typical ionization curves a and b for ions of the same mass-tocharge ratio obtained from equal amounts of two pure components. The appearance potentials,

that is the potentials below which no ions are formed, differ between the respective components in this case. Curve c is a plot of the ratio of the amounts oi ions formed from the two components at the respective voltages. Curve d is a curve obtained for such ions from a mixture containing 'half of said amount of each component. Though the ionization lcurves of the individual components are relatively smooth, the curve d for the mixture has a sharp bend at point p, givingv an indication of the character 'of the mixture. While the composition of the mixture may be determined by comparison of curves a, b, and d, the analysis of the mixture may be made more simply by solving a pair of simultaneous equabeam may be adjusted to successively different tions, of the type given in Equation 1 above. corersponding to two voltages at which the components produce such ions in different proportions. For example, if thev respective ion beam intensities ot a sample o f the mixture and samples of the components are measured at both, 50 volts and 20 volts, and these are the only two components present in the mixture, the following relations hold:

wPIM: 13.7 (4) and t and the sensitivity coemclents for the second component at the'two voltages are If in a particular instance the intensity of the beam obtained for the mixture at 50 and 20 volts respectively are:

Ila-@215.7 (l0) and Y the amounts of the components present in the mixture will be My invention may, of course, be applied to cases in which more than two components are present in the mixture..` vIt 'may even be applied if more than two components produce ions of the same mass-to-charge ratio. In this case a number of variations'ofmy invention will naturally occur to those skilled 'in the art, in view of the foregoing explanations. For example, an analysis of a three-component mixture may be made readily by measuring the intensity of a single ion beam at three voltages. if theshapes of the ionization curves are suiliciently different for the three components. In the event that the shapes of the ionization energy curves for two' components are the same for a particular io`n beam to which another component having-an ionization curve of different shape makes a contribution,

. additional dataare-necessary in order to perform a complete analysis. Such additional data may be 'in the form of ion beam intensity and sensitivity coeflcients for ions of another massto-charge ratio, applicable to the mixture and its components. In any case it will be clear that an analysis containingn-components may be obtained from computations based on n independent measurements obtained in a mass spectrometer on the mixture, provided, of course, corre-v sponding data for the individual components are f known or determined.

From the foregoing description it is seen that I have provided anew method of mass spectrometry which makes possible the determination of the amounts of a greater number of components presentin a mixture than' there are ions of diierent'mass-to-charge ratios measured.

1. In the analysis of a mixture containing a plurality of components involving the comparison of measurements of the quantities of ions of predetermined mass-to-charge ratios formed in a potential.

mass spectrometer from said mixture and from reference samples including different proportions of the components under. the same ionization conditions, the improvement which comprises separately ionizing said mixture at dierent ionization voltages at which ions of one such mass to-charge ratio are formed in diierent proportions from the respective components,and measuring the amounts of such ions formed at each said voltage from the vmixtureand from the several reference samples.

2. In the analysis of a mixture by mass spec- 'l trometry the improvement which comprises bombarding the mixture with an electron beam having a potential such that ions of a common massto-charge ratio are formed from a plurality of components in the mixture, bombarding the mixture with an electron beam-having a different potential but one such that ions of the common mass-to-charge ratio are formed from the same components, but the proportions of the ions de'- rived from the components being different than the proportions derived from these components at the other potential and measuring the amount oi? the -same mass-to-charge ratio such that the ions of the plurality of components are in a dif. ferent-ratio to each other than .when under they mst-mentioned bombardment and measuring the amount of the ions of said mass-to-charge ratio formed from the mixture at each ionization REED C. LAWLOR.

Images