US 3129327 A
Description (OCR text may contain errors)
A ril 14, 1964 w. M. BRUBAKER 3,129,327
AUXILIARY ELECTRODES FOR QUADRUPOLE MASS FILTERS 4 Sheets-Sheet 1 Filed Dec. 12, 1961 j 6 w Z 42 6 5 a 4.?/ m f \r/ In v A 6 x A m x/ g z AV z w MW A w W). 4 i w v 0 Z A My m April 14, 1964 w. M. BRUBAKER 3,129,327
AUXILIARY ELECTRODES FOR QUADRUPOLE MASS FILTERS Filed Dec. 12, 1961 4 Sheets-Sheet 2 w. M. BRUBAKER 3,129,327 AUXILIARY ELECTRODES FOR QUADRUPOLE MASS FILTERS 4 Sheets-Sheet 5 April 14, 1964 Filed Dec. 12, 1961- April 14, 1964 w. M. BRUBAKER 3,129,327
R QUADRUPOLE MASS FILTERS AUXILIARY ELECTRODES F0 Filed Dec. 12, 1961 4 Sheets-Sheet 4 INVENTOR. VV/L 50/1/ flfimmm? United States Patent Ofiice 3,129,327 Patented Apr. 14, 1964 3,129,327 AUXILIARY ELECTRGDES FOR QUADRUPOLE MASS FHJTERS Wilson M. Brubalier, Arcadia, Califi, assignor, by mesne assignments, to Bell 8; Howell Company, Chicago, Ill.,
a corporation of Illinois Filed Dec. 12, 196i, Ser. No. 158,697 4 Claims. (Cl. 250-413) This invention is concerned with the entrance and exit construction of the quadrupole mass filter, and has particular reference to the disposition and control of auxiliary electrodes adjacent the entrance and/ or exit of the filter in order to improve its operating characteristics and capabilities.
The basic function of any mass filter is to separate or select ions having different ratios of M/ e (mass to charge), a function of primary importance in mass spectrometers and other instruments. The quadrupole filter accomplishes this in a unique way, ithout a magnet, by utilizing the motion of charged particles in a quadrupole electric field having both alternating and static components.
The quadrupole filter consists essentially of four primary electrodes in the form of parallel cylindrical rods arranged symmetrically about a central axis. Electrically, the rods are connected in pairs, opposing rods connected together. If Z denotes the axis of the rod array, then one pair of rods lie with their centers on the X-axis and the other pair on the Y-ayis, according to the convention of a rectangular cartesian coordinate system. The rods are excited by both AC. and DC. voltages. Ions are introduced at one end of the filter, travel generally down the axis of the filter, and emerge at the other end of the filter. In traversing the filter, ions of different M/ e are separated so that only ions of a selected M/ e emerge from the filter. Ion selection may be controlled by varying the voltage levels on the electrodes and by varying the frequency of excitation. An example of a quadrupole filter is disclosed in US. Patent 2,939,952 to W. Paul et al., 1960.
The quadrupole mass filter is potentially very useful for upper atmosphere research, as an analytical device in a satellite vehicle or the like, where the ion source for the filter is the space surrounding the satellite vehicle. For such applications the filter must have high sensitivity, high revolving power and must consume as little electrical power as possible. Under these conditions, the quadrupole filter suffers a serious loss in transmission efficiency; that is, many of those ions entering the filter and having the selected M/ e which should traverse the filter and emerge at the exit, do not do so.
I have discovered a way to greatly enhance the transmission efficiency of the quadrupole filter, Without reducing its sensitivity or revolving power and without increasing the power consumption. According to my discovery, this may be accomplished by controlling the fringe fields adjacent the entrance of the filter and, perhaps to a somewhat lesser extent, by controlling those adjacent the filter exit.
More specifically, my invention is applied to a multipole mass filter of the character described having a plurality of primary electrodes spaced about a central axis, and having means for applying a DC. voltage to the primary electrodes to produce a static multi-pole electric field com ponent between the primary electrodes and means for applying an A.C. voltage to the primary electrodes to produce an alternating multi-pole electric field component between the primary electrodes. I combine with this filter structure a plurality of auxiliary electrodes spaced about the central axis adjacent the end of the primary electrodes, and means for controlling the auxiliary electrodes to produce a decrease in the ratio of the static to the peak alternating component of the multi-pole electric field in the vicinity of said end of the primary electrodes.
Referring to the accompanying drawings:
FIG. 1 is a schematic sectional elevation of a quadrupole mass filter constructed in accordance with my invention;
FIG. 2 is a fragmentary perspective view of the entrance end of the quadrupole filter of FIG. 1;
FIG. 3 is a schematic diagram indicating the general manner in which voltages are applied to the primary and auxiliary electrodes of a quadrupole filter constructed in accordance with my invention;
FIG. 4 is a schematic diagram of an electrical circuit for controlling the voltages applied to the primary and auxiliary electrodes of the quadrupole filter of FIGS. 1 and 2;
FIG. 5 is a graph illustrating a stable trajectory in the X-Z plane for an ion having a selected M/ e to traverse the filter;
FIG. 6 is a graph illustrating a stable trajectory of the ion in the Y-Z plane; and,
FIG. 7 is a graph, called a stability diagram, illustrating the operating characteristics of a quadrupole mass filter.
Referring now to FIGS. 1 and 2, the quadrupole mass filter includes a cylindrical metallic housing 10 having four primary electrodes 12, 14, 16, 18 mounted therein on electrical insulating supports 20. The primary electrodes are in the form of coextensive cylindrical steel rodsrextending parallel to one another and disposedsymmetrically about the central axis Z of the filter. One pair of diametrically opposed rods 12, 14 lie with their centers in the Y-Z plane and may be called the Y-rods; and, the other pair of opposed rods 16, 18 lie with their centers in the X-Z plane and may be called the X-rods. Ideally the rods should provide a hyperbolic curvature in cross section. However, in practice the cylindrical curvature is an adequate approximation.
A conductive plate 22 is mounted across one end of the housing. The plate 22 has a centrally located circular aperture 24 forming the entrance for the filter and called the ion entrance aperture. Similarly, at the opposite end of the filter is mounted a conductive plate 26 having a central circular aperture 28 therein, serving as the ion exit aperture for the filter.
The housing 10 is closed by a conductive rear wall 30 on which an electrical insulating support 32 is mounted, An ion collector electrode 34 is mounted on the support 32 opposite the ion exit aperture 28. Ions entering the I entrance aperture 24 and traversing the filter and imping ing upon the collector 34 will electrically charge the collector. The ion current may be measured by any conventional measuring circuit 35 connected between the collector and ground. The conductive housing 10, the aperture plates 22, 26, and the rear wall 30, all are at ground potential.
When used in the laboratory, the housing 10 is evacuated, and an ion source (not shown) is mounted over the entrance aperture. However, for upper atmosphere research, the ion entrance aperture is exposed, and the vacuum inside the device is the vacuum of space.
Eight auxiliary electrodes 36 through 50 are spaced about the filter axis adjacent the entrance end of the primary electrodes, and eight auxiliary electrodes 36A through 50A are spaced about the filter axis adjacent the exit end of the primary electrodes. The auxiliary electrode arrangement at the entrance end is identical to that at the exit end of the filter, therefore only the former will be described.
As can be seen at the entrance end of the filter, the auxiliary electrodes 36 through 50 are disposed respectively adjacent to and extend in spaced relationship partially around corresponding ones of the primary elec trodes. Typically, auxiliary electrodes 36, 38, each formed of thin steel sheet, extend about primary electrode 12 in spaced relationship and in a generally U-shaped pattern approximating a hyperbolic curvature, with the legs of the U extending between adjacent rods and outward from the filter axis. Preferably the steel sheet is solid, although it may be in. the form of a screen. The auxiliary electrodes 36, 38 are held in position by electrical insulating supports 52, 5,4. The other auxiliary electrodes at the entrance are similarly constituted and supported. Thus, auxiliary electrodes 40, 42 extend in spaced relationship about primary electrode 14; auxiliary electrodes 44, 46 about primary electrode 16; and, auxiliary electrodes 48, 50 about primary electrode 18.
In an actual device fabricated according to the illustrations in FIGS. 1 and 2, the interior of the housing and its contents are gold plated, including the primary and auxiliary electrodes but not including the insulating supports for these electrodes. Of the pair of auxiliary electrodes associated with each primary electrode, the ones closest to the axis of the filter, called the outer electrodes, 36, 40, 44, 48, have a thickness of 0.010 inch. The auxiliary electrodes disposed farthest from the axis, called the inner electrodes, 3%, 42, 46, 50, have a thickness of 0.025 inch. Otherwise, typical dimensions are as labeled in FIG. 2 and as correspondingly set out in the following table:
A inches 0.130 B do 0.220 C do 0.125 D do 0.312 E d 0.437 F do 0.062 R do 0.527 r d0 0.618 L do 40 0 degrees 30 From the above table, it will be seen that FIGS. 1 and 2 are by no means drawn to scale. It will be noticed that the outer auxiliary electrodes 36, 40, 44, 48 are narrower than the inner auxiliary electrodes 38, 42, 46, 50. The purpose is to avoid obstructing the entrance aperture. This instrument was designed to accept ions incident upon the filter with an angle less than 30 indicated by the angle 0.
The electrical connections are indicated in a general fashion in FIG. 3. Diamet rically opposed rods are connected together in pairs, and both A.C. and D.C. voltages 2(V,, cos wt+V are applied between the two rod pairs, and balanced to ground. The X-rods 16, 18 are D.C. positive, and the Y-rods 12, 14 are D.C. negative. This creates a gradrupole electric field having both A.C. and D.C. components between the rods. Similarly, diametrically opposite auxiliary electrodes are electrically connected together in pairs, and both A.C. and D.C. voltages 2(v cos wt+v are applied between the connected pairs.
Thus referring for example to FIG. 2, diametrically opposite auxiliary electrodes 36, 40 are connected electrically as a pair, and opposed auxiliary electrodes 44, 48 are connected electrically as a pair. The corresponding primary electrode pairs are 12, 14 and 16, 18, respectively. Potentials are applied between the auxiliary electrode pairs in a manner which causes a decrease in the ratio of the static to the peak alternating component of the multipole electric field of the primary electrodes in the vicinity of the endsof the primary electrodes e.g., the filter en trance. This is accomplished by maintaining a D.C. voltage level between the two pairs of auxiliary electrodes 36, 40 and 44, 48 different from that maintained between the corresponding pairs of primary electrodes 12, 14 and 16, 18 respectively, the difference being in thedirection to subtract from the D.C. field component created by the primary electrodes. In fact, it may at times be desirable to maintain a D.C. voltage between the auxiliary electrode pairs which is of opposite polarity to that maintained between the corresponding pairs of primary electrodes.
Also, it is desirable to apply between the pairs of auxiliary electrodes a fraction of the A.C. voltage applied between the corresponding pairs of primary electrodes and in the same phase, in order to further increase the A.C. component of the multi-pole electric field in the vicinity.
Since the mass filter is used primarily for the separation of positive ions, the minimum requirement for auxiliary electrodes is one pair, for example 36, 40 associated with a pair of primary electrodes 12, 14 which are D.C. negative. The reason for this is that, according to the convention chosen, the stability of positive ion trajectories in the Y-Z plane is the determining factor for overall transmission eificiency. Preferably, of course, each primary electrode has at least one auxiliary electrode associated therewith. So long as the entrance aperture is not obstructed, the number and arrangement of auxiliary electrodes associated with each primary electrode is determined by conventional design considerations for electric field control. The more elaborate the arrangement, the more precise the control over the electric field.
A more detailed circuit for applying and controlling the voltages on the primary and auxiliary electrodes, particularly as to the apparatus illustrated in FIGS. 1 and 2, is shown in the circuit diagram of FIG. 4. In this circuit a radio frequency generator 56, the peak output voltage of which and the frequency of which is adjustable, supplies balanced A.C. signals through capacitive coupling across the pairs of primary electrodes 12, 14, and 16, 18. A fraction of the A.C. potential also is supplied through variable capacitors across corresponding pairs of auxiliary electrodes, as indicated.
Half wave rectification of each of the A.C. signals from the radio frequency generator is provided by a pair of diodes 58, 60, and the resulting D.C. voltage is applied across the terminals 62, 64 of a voltage divider network. At terminal 62 the D.C. voltage is maintained positive with respect to ground. At terminal 64 the voltage is maintained negative with respect to ground. A balance adjustment is provided by a resistor 66 having an adjustable grounded tap.
The D.C. potential to be applied across the pairs of primary electrodes is taken over a pair of leads 67, 68 from a first arm of the divider network. This D.C. potential may be adjusted by a variable shunt resistor 70.
The other two arms of the divider network each contain crossed potentiometers with ganged taps 72, 74 and 76, 78 respectively, the taps being coupled to the auxiliary electrodes for maintaining a D.C. voltage across auxiliary electrode pairs.
Typical values for the circuit components are indicated on the drawing. Typical operating frequencies and voltages are as follows:
By adjusting the crossed potentiometers in the voltage divider network, the D.C. potential between auxiliary electrode pairs may be varied, and the polarity changed. The A.C. potential across the auxiliary electrodes is adjusted by means of the variable capacitors shown adjacent these electrodes in the diagram. It should be noted that some capacitive coupling between the auxiliary electrodes and primary electrodes exists by virtue of their physical disposition adjacent one another.
Employing apparatus constructed and operated in accordance with the above teachings, I have observed 30% increases in transmission efiiciency for ions within the pass band of the filter; i.e., ions having the selected M/e to traverse the filter. It is possible to explain to some extent how this comes about by theoretical con siderations, although I do not intend to be bound by them.
The potential in the space between the rods of a quadrupole mass filter is given approximately by the following equation:
where V and V are the potentials applied to the rods, and R is the distance from the central axis Z to the rods. It follows that the force equations for ions of diiferent M/ e are:
The equations of motion of the ions are obtained by the integration of the above force equations. In this case, the integration cannot be done directly or simply. The formal solutions are given by the Mathieu equation: see N. W. McLachlan, Theory and Application of Mathieu Functions, Oxford Press, 1947. These solutions are expressed as infinite series, and the coefiicients are given as continued fractions. In order to learn much about the trajectories by these means, the use of a large computer is required.
The force equations are brought into the standard form of Mathieu equations through the following substitutions:
Whether the trajectory is stable is determined solely by the values of a and q. For the quadrupole filter the region of interest is limited to values of a and q less than unity. It is possible to construct a graph of a and q values to illustrate stable and unstable trajectories. An example of such a stability diagram is illustrated in FIG. 7, where the region under the triangularly shaped curve contains those values of a and q for which the trajectory is stable in both the X and Y directions, with the region to the left of the curve representing unstable trajectories in the Y direction and the region to the right of the curve representing unstable trajectories in the X direction.
The set of trajectories illustrated in FIGS. 5 and 6 were determined by numerical integration of the force equations with a digital computer. It can be seen that the trajectory in the X direction is vastly different from that in the Y direction.
In order for the ions to reach the collector, they must be stable (have bounded amplitudes of motion) in both the X and the Y directions, i.e., for both plus and minus values of a and q. All of this is taken into consideration in constructing the stability diagram of FIG. 7. Ions whose trajectories carry them too far from the axis, come into contact with interior surfaces of the filter and are captured, thereby removing them from the beam. On the other hand, ions whose movement in response to the electric fields is along paths whose amplitude is bounded within the rod array continue down the filter to the collector.
In the X-Z plane, the DC. potential on the X-rods accelerates positive ions toward the filter axis; whereas, the A.C. potential causes the positive ions to oscillate about the filter axis, analogous to a resonant system. If the amplitude of the oscillations becomes too large, the ion will be lost. It should be noted that while in the usual resonant system the amplitude is bounded for excitation frequencies on either side of resonance, in the quadrupole system, the ions appear to oscillate with a bounded amplitude only when their mass is above the resonant mass. For the resonant mass and all lighter masses, the amplitude appears to increase without limit.
In the Y-Z plane, the positive ion is accelerated outwardly from the axis by the negative D.C. potentials on the Y-rods. If the trajectory of the ion is to be stable, it must result from the A.C. field whereby the motion of the ion in a non-uniform electric field causes the net momentum impulse toward the axis to equal or exceed that directed away from the axis when the ion approaches the outer limits of a stable trajectory. This condition will obtain only when the ion has a sufiicient low mass so that it moves enough. For higher masses, the ion will not move enough responsive to the A.C. field to achieve stability, and will be lost from the beam.
Thus, in the X direction positive ions are stable in the DC. fields and the influence of the A.C. field is to make them unstable. The lighter ions tend to be unstable. In the Y direction, stability results solely from the motion of the ions in the non-uniform A.C. fields, and the heavier ions tend to be unstable.
Referring to the stability diagram in FIG. 7, if the ratio of the DC. voltage to the A.C. voltage is constant, the values of a and q lie on a line which may be called the scan line 80. Notice that the values of a and q depend upon the M/e of the ion although the ratio of a to q does not. Hence, for given values of frequency and A.C. and DC. voltage, the a and q values corresponding to ions of differing M/es will be spread along the scan line. The loci of the a and q values of the heavier ions lie nearer the origin and those of the lighter ions lie more remote. If, as in FIG. 7, the scan line is chosen to pass near the apex of the stability curve, the trajectory of ions of a very limited range of M/e are stable, and all others are unstable. The theoretical resolving power increases as the slope of the scan line increases and approaches the apex of the stability curve. The resolution is adjusted by varying the ratio of V to V With this ratio fixed, mass scanning is accomplished by varying the ratio of the voltages to the square of the frequency.
According to the theory of the operation which I propose for the auxiliary electrodes of my invention, the loss of transmission efliciency for those ions lying within the pass band of the filter is primarily attributable to the passage of these ions through the entrance to the filter, where the fields are less than their full value. Depending upon the geometry, a similar but less critical condition exists at the filter exit. The defocusing forces in the entrance region are many times stronger than the focusing forces in the uniform field region. It is the purpose of the auxiliary electrodes to reduce these strong defocusing forces.
Thus, ions entering the filter may do so with a large radial velocity component which may cause them to be lost from the beam prior to the time when the stabilizing influence of the uniform field region of the filter can turn them toward the axis. Also, ions abruptly entering the filter may acquire a radial velocity kick from the A.C. field. Ions passing through the entrance region where the fields are less than their full value also may acquire radial velocity as a result of the action of these fields, particularly as a result of Y unstability.
Thus, according to my theory, the transition from no fields at the entrance should be gradual, not abrupt; and, the ratio of the static to the peak alternating component to the field in the vicinity of the entrance should be reduced, by enhancing the A.C. field and reducing the D.C. field. While the latter may weaken the X stability, the X stability is usually much greater than the Y stability. And, since in the normal use of the mass filter the ions must be stable in both the X and Y directions to reach the collector, nothing is lost as the X stability is weakened so long as it exceeds the Y stability.
1. In a multi-pole mass filter having a plurality of substantially parallel and co-extensive primary electrodes spaced symmetrically about a central axis, means for applying a D.C. voltage to the primary electrodes to produce a static multi -pole electric field component between the primary electrodes and means for applying an AC. voltage to the primary electrodes to produce an alternating multi-pole electric field component between the primary electrodes, the improvement which comprises a plurality of auxiliary electrodes spaced about the filter axis adjacent an end of the primary electrodes, and means for controlling the auxiliary electrodes to produce a decrease. in the ratio of the static to the peak alternating electric field component in the vicinity of said end of the primary electrodes.
2. In a quadrupole mass filter having a plurality of primary electrodes spaced about a central axis, means for applying a D.C. voltage to the primary electrodes to produce a static quadrupole electric field component between the primary electrodes and means for applying an AC. voltage to the primary electrodes to produce an alternating quadrupole electric field component between the primary electrodes, the improvement which comprises a corresponding plurality of auxiliary electrodes in the form of thin conductive sheets spaced about the filter axis adjacent an end of the primary electrodes, the auxiliary electrodes being disposed respectively adjacent to and extending at least partially around corresponding ones of the primary electrodes, and means for maintaining AC. and D.C. potentials on said auxiliary electrodes so as to control the quadrupole electric field in the vicinity of said end of the primary electrodes.
3. Apparatus of claim 2 wherein each conductive sheet has a generally U-shaped configuration, with the legs of the U extending outwardly from the filter axis and between adjacent primary electrodes.
4. In a multi-pole mass filter having at least four approximately parallel and coextensive primary electrodes spaced diametrically opposite one another in pairs about a central axis, means for applying a DC. voltage of one polarity to one pair or diametrically opposite primary electrodes and a D.C. voltage of opposite polarity to another pair of diametrically opposite primary electrodes so as to produce a static multi-pole electric field compo nent between the primary electrodes, and means for applying an A.C. voltage to the primary electrodes to produce an alternating multi-pole electric field component between the primary electrodes, the improvement which comprises at least two auxiliary electrodes spaced diametrically opposite one another about the filter axis, said two auxiliary electrodes being respectively disposed adjacent corresponding ends of one of said pairs of primary electrodes, and means for maintaining a D.C. voltage level on the auxiliary electrodes which is substantially different from the D.C. voltage on the corresponding pair of primary electrodes, the difference being in the direction toward the D.C. voltage level of the other pair of primary electrodes.
Paul et a1 June 7, 1960 Paul et al. Aug. 23, 1960