US 3213276 A
Description (OCR text may contain errors)
Oct. 19, 1965 H. A. ENGE 3,213,276
MAGNETIC ANALYZING SYSTEM FOR A MASS SPECTROMETER HAVING BI-DIRECTIONAL FOCUSING Filed June 24, 1965 Fig. 2
United States Patent 3,213,276 MAGNETIC ANALYZING SYSTEM FOR A MASS SPECTROMETER HAVING BI-DIREtITIDNAL FOCUSING Harald A. Enge, Winchester, Mass., assignor to High Voltage Engineering Corporation, Burlington, Mass., a corporation of Massachusetts Filed June 24, 1963, Ser. No. 289,892 1 Claim. (Cl. 25041.9)
This invention relates to spectroscopy of the type wherein corpuscular radiation of heterogeneous momentum is analyzed by a momentum separating magnet which serves to spread out in space particles having different momentum. The momentum spectrum may then be measured either by photographic methods or by counting methods. In a magnetic spectrograph, photographic methods are employed and the entire range produced by the magnet is recorded on a photographic plate in one exposure. The photographic image is then examined under a microscope and the number of tracks made by the incident particles is counted by eye.
My invention comprehends a magnet for use in magnetic spectroscopy and, in particular, a homogeneous field deflecting magnet which produces a homogeneous magnetic field in two separated regions through which the charged particles travel successively. In accordance with the invention, the separation between the two regions increases progressively in the direction away from the centers of the radii of curvature of the particle trajectories, and the incident boundary of that region which is first traversed by the particles is concave towards the incoming particles. By the term increases progressively I mean that the rate of increase increases.
In general, the type of momentum separating magnet with which the invention is concerned is used for studies of nuclear physics in the binding energy range. In such studies a thin target whose composition may be known accurately is bombarded with a beam of charged particles, and the energies and intensities of the charged products from the nuclear reactions in the target are measured. In this way a graph of intensity of the charged products of the nuclear reactions as a function of their energy may be obtained. Monoenergetic or nearly monoenergetic particles emitted from the target will appear in such a graph as a peak with a width that depends upon the size of the object (beam spot on the target) and upon the aberration (focusing error) of the instrument. The smaller the widths of such peaks are the more accurate is the information obtained. The mean momentum of the particles in a peak divided by the width of the peak at half maximum intensity is defined as the momentum resolving power or resolution of the instrument.
The better the focusing in the median plane the higher the resolution; the better the focusing in the direction perpendicular thereto (hereinafter called z-direction focusing), the more well defined is the area of the nuclear track plate that must be scanned. Because the separation between the two magnetic field regions increases progressively in the direction away from the center of the radii of curvature of the particle trajectories, the invention provides a z-direction focusing over the whole range of momenta recorded. Because the incident boundary of that region which is first traversed by the particles is concave towards the incoming particles, second-order errors in median plane focusing caused by z-direction motion are made to vanish or are made small.
Especially for the transmutation type experiments, measurements are usually made at various angles with respect to the incident beam of charged particles in a plane which includes the incident beam. Generally it is convenient to select this plane as a horizontal plane with a horizontal incident beam since in this way the magnetic spectrograph or spectrometer may be moved about the target more readily. In the case of the magnetic spectrograph the particles which are emitted into the aperture of the instrument are deflected by a magnetic field which is so constructed that particles having different energies will be focused at different points. A photographic plate is positioned along the locus of these focal points and a picture is obtained showing the momentum spectrum of the particles emitted at that angle. The spectrograph provides a great deal of information in one exposure, and it is not usually of vital importance to have a very large aperture. Accordingly spectrographs have been designed which will record simultaneously a wide range of momenta but have a small solid angle of acceptance. See, for example, the article by Browne and Buechner in The Review of Scientific Instruments, vol. 27, pp. 899-907.
The construction of magnetic spectrographs is well known and need not be gone into in detail herein. As an example, reference is made to the above cited article by Browne and Buechner. In general, a magnetic field of certain precise configuration is provided by a magnet having pole pieces between which the charged particles pass. A target is bombarded by an artificially accelerated beam of charged particles and the entrance aperture of the magnetic spectrograph is positioned such that a small portion of the particles emitted from the target is admitted between the pole pieces. The magnetic field then deflects these charged particlesdifferently for the different momentawith the result that they are spread out into a sheet which is permitted to impinge upon a photographic plate positioned along the focal surface.
The invention may best be understood from the following detailed description thereof having reference to the accompanying drawing, in which:
FIG. 1 is a view along the median plane of a magnetic spectrograph constructed in accordance with the invention; and
FIG. 2 is a view along the line 22 of FIG. 1.
Referring to the drawing, the magnet has two pairs of pole faces 1, 2 between each of which there is produced a uniform magnetic field, the strength of which is the same between each pair of pole faces. While it would be possible to utilize two separate magnet yokes for the two pairs of pole faces without departing from the scope of the invention, the fact that the magnetic fields must be of the same strength means that it is preferable to construct a single yoke 3 for both pairs of pole faces. The magnet, of course, is energized in the conventional manner by appropriate coils 4 (one on each side of the median plane) driven by a DC. electric current produced by a suitable power source. The charged particles which are to be examined travel through the spectrograph successively between the two pairs of pole faces and a vacuum chamber 5 is provided so that the path of the charged particles will not be appreciably interrupted by any gas molecules. From the point of view of the charged particles therefore, in passing through the spectrograph they travel though two distinct areas in which there is a uniform magnetic field. Insofar as the trajectory in the median plane is concerned, it may be assumed that the boundaries of these magnetic fields are well defined. It is diificult and not vitally important to eliminate the magnetic field entirely in the region between the two field areas. In practice, it may be reduced, for example, by a factor of 8. Thus, ase shown at 6, the trajectory of each charged particle is a straight line outside these areas (slightly curved between these. areas) and is circular within these areas, the radius of curvature of the'circular trajectories being directly proportional to the momentum of the charged particle. .The charged particles will be coming from a source such as a target 7 from which they are produced as a result of bombardment by other particles. The charged particles to be examined travel from this target or other source and enter one of the areas across one of its boundaries. For convenience we may refer to this area as the first area 8 and to this boundary as the incident boundary 9 of the first area. The charged particles then traverse the first area 8, the gap 10 between the two areas, and then the second area 11 from which they emerge across a boundary thereof that may be referred to as the exit boundary 12 of the second area. Since the effect of the magnet is to spread the particles out according to their momentum or energy, the entrance or incident boundary 9 of the first area will be of less extent than the exit boundary 12 of the second area. The variables involved are the contours of the four boundaries 9, 12, 13, 14 and the width of the gap 10 between the two areas, the four boundaries being the entrance boundaries. 9, 14 and exit boundaries 12, 13 of each of the two areas 8, 11. The magnetic field strength is also a variable but is not a critical one since the only effect of increasing the magnetic field strength is to enable the device to handle higher momenta. The entrance boundary 9 should be circular with the concave side facing the exterior of the first area 8 (i.e. the entrance boundary 9 of the first area 8 should be concave) while the exit boundary 13 of the first area 8 should likewise be circular but with the concave side facing the first area 8 (i.e. the exit boundary 13 of the first area 8 should be convex). As will appear hereinafter, deviations may be made from the circular pattern without departing from the spirit and scope of the invention. The entrance boundary 14 of the second area is less critical and may be straight or curved but in any event it is essential to the invention that the width of the gap 10 between the two areas increase progressively from the concave side of the particle trajectories to the convex side. The exit boundary 12 of the second area is preferably straight.
It is readily apparent that the effect of the device will be to separate particles having different momenta. As a spectrograph however the device must also focus particles having the same momentum but which diverge from the target. It can be shown geometrically that in the device just described particles emitted from the target and having the same energy will be focused at a point and the loci of these focal points form a focal surface which approximates a straight line and which may be rendered a straight line by adjustment of the shape of exit boundary of the first area or the entrance boundary of the second area. A photographic plate 15 (or other detecting device) is positioned on this focal surface. The precise shape of the boundaries may be computed to a first approximation by geometric tracing of the particle trajectories and further refinements may be made by the use of electronic computers for solving the equations of motion of the particles from source to focal surface. In these calculations semiempirical formulas are used for describing the field distribution in the neighborhood of the boundaries of the pole areas.
A point source of monoenergetic particles produces on the photographic plate an image which because of the aberration of the spectrograph has a width as measured along the plate in the median plane:
When the source is not a point source an additional term O M should be included in Eq. 1. O is the object size in the median plane as measured perpendicular to the central ray and M is a magnification factor which varies with the image position on the plate. The coefficients a and b in the power expansion depend in general upon the radius of curvature of the particle orbits, or, equivalently stated, upon the image position on the photographic plate. The angle a is the half-angle of divergence of the beam in the median plane as it enters the magnet system. The angle 6 is the half-angle of divergence in the z-direction. Because of symmetry about the median plane, the coefficients for odd powers of ,8 vanish. In many spectrographs the angle p is so small as to render the terms in p in Eq. 1 negligible. In instruments where z-direction focusing is provided, the objective is usually to increase ,8. The termb fi will then be important. Any magnet or pair of magnets will produce a first order image (real or virtual) in the median plane. First order focusing is that involved for infinite simal small angle 0:, that is to say, only a needs to be zero. The loci of real focal points for varying momentum form a straight or curved line, along which the photographic plate may be placed.
A formula similar to Eq. 1 can be written also for the image size in the z-direction. For a point object one gets Z= 15+ 3B 11l 21 l Again, if the source is not a point source a term O M should be added. The objective of z-direction focusing is to collect as many particles as possible and to reduce the detector area required. Since the aberrations in the z-direction do not affect resolving power but merely give a fuzzy image in that direction, one need not be so concerned about the second and higher order terms in Eq. 2 as in Eq. 1.
The focusing properties in the z-direction for homogeneous field magnets are determined by the angles between the particle trajectory and the normals to the field boundaries at entrance (e and exit (6 The fringing fields at entrance and exit act as lenses, with focal lengths given approximately by R/f=tan e where R is the trajectory radius in the homogeneous field region.
The exponential horn shape of the inter-area space is required in order to enable one to have z-focusing over the whole momentum range, i.e. in order to have enough focusing of particles at the high momentum and of the spectrum. In the embodiment of the invention shown in FIG. 1, the boundaries of this gap bot-h act as positive (focusing) lenses for most of the momentum range. The exit surface of the first field provides a departure angle varying with momentum (radius of orbit) from 28 to 7 while the entrance surface of the second field might provide an incident angle of -3 to 43. The strongest z-focusing in the gap is experienced for the high momentum particles which are less focused at the entrance boundary of region 1. It is evident that a progressive increase of the inter-area gap is necessary to accomplish first order z-focusing over the whole momentum range (to make B zero over the whole range).
The exponential horn arrangement also introduces an error in z focusing, since it not only focuses more strongly high momentum particles, but also over-focuses particles which diverge so as to arrive at the incident boundary of the first area at a larger angle of incidence than that of the central ray (to the left of the central ray in FIG. 1). This error which in Eq. 2 is expressed by the term C m/3 is compensated for or partly compensated for by making the incident boundary of the first area concave. The error is not fully compensated since suflicient curvature cannot be given to the incident boundary without having it bend back into the path of the beam. A representative entrance angle is 40 or 45. The magnitude of this angle as well as the opening angle of the exponential horn are dictated by the need for strong enough z-focusing (to make B =0).
As previously noted, a is always made equal to zero merely by varying the curvature and position of the focal surface. The focal surface may be made a straight line by making the exit boundary of the first area or the entrance boundary of the second area S-shaped. As
noted, a is sensitive to the position and direction of the exit boundary which is then adjusted so that :1 is zero at two points along the focal surface with the result that a is near zero all along the focal surface. The curvature of the incident surface also strongly influences the coeflicient [7 in Eq. 1 and with a judicious choice of curvature, which also is a practical choice, [2 can be made close to zero over the whole range of momenta covered. Second-order focusing is thereby provided in the median plane also with respect to z-direction motion. Because the invention provides focusing in the z direction, the angle B is large enough to produce appreciable errors in median plane focusing if attention is not paid to the co-efiicient b In spectrographs where z-direction focusing is not provided, the term b 5 produced a slight curvature in the line image of a point object. With zdirection focusing this curved line is compressed to an image which is blurred in the median plane direction. If the object is not a point object, the curved image becomes even more blurred. The angle of deflection for the central rays is for the embodiment of the invention shown in FIG. 1 is 67 to 24, depending upon momentum, in the first area and 42 to 85 in the second area. The total angle of deflection is 115 to 112 including the small inter-area deflection. Such a large angle of deflection is necessary to keep the object and image distances of the instrument reasonably small. The exact choice of total deflection angle and of the exit angle of area 2 is made so as to minimize the coeflicient a in Eq. 1. In FIG. 1, the exit angle is shown as -20.
Apparatus for magnetic spectroscopy in which charged particles originating from a source are deflected so as to form trajectories which define a central plane, comprising in combination a vacuum chamber enclosing said central plane, and means for producing a substantially homogeneous and uniform magnetic field perpendicularly across two separated regions of said central plane, the separation between the two regions increasing progressively along the separation-gap, that boundary of the first of said regions to be traversed by said charged particles which is the most remote from said separation gap being concave, the normal to the field boundary at the entrance to the first region lying on the outside of the trajectory as seen from the center of curvature of the trajectory in the first region, whereby particles originating from said source and having the same momentum will be focused towards a point and whereby the loci of these focal points form a focal surface, the focusing action occurring not only in said central plane but also in the direction perpendicular to said central plane, the focusing action occurring for the high momentum particles as well as for the low momentum particles, and a detecting device at said focal surface.
RALPH G. NILSON, Primary Examiner.