EP0151078B1 - High intensity mass spectrometer with simultaneous multiple detection - Google Patents

Description

The invention relates to a charged clarity particle separator or mass spectrometer for the identification and simultaneous measurement of several elements.

The spectrometer is intended to receive a beam of charged particles or ions, composed by particles of different masses (M - M ₁ , M ₂ , Kt, etc.), animated by slightly different kinetic energies. We denote by V (eV) their average kinetic energy and ± AV the corresponding energy dispersion.

A mass spectrometer generally has an entrance slit, after which the beam passes into an electrostatic sector, then into a magnetic sector. The purpose of this arrangement is to deflect the particles, in a selective manner as to their mass and as much as possible independently of their energy. The deviation occurs in a so-called radial plane which is the plane of symmetry of the instrument and which is perpendicular to the large dimension of the entry slit.

The particle beam therefore has a radial component, and a perpendicular component, in what is called the vertical section.

It is known to use a magnetic sector which has a plane entry face and exit face, and in which the entry face is inclined on the axis of the particle beam, while the plane of the face of exit passes through the intersection of this entry face with the axis of the particle beam. Under these conditions, the angle of deflection of the particles in the magnetic sector does not depend on the mass of the charged particles, which introduces a simplification. The radius of curvature of the trajectories depends on the mass of the particles.

The quality of a mass spectrometer is defined by its separating power M / AM, where AM is the smallest difference in mass that can be distinguished with the instrument. In a spectrometer whose optics would be perfect (the word "optics" is used here in the broad sense), this separating power would only depend on the dimensions of the entrance slit. In reality, the images of the entry slit, or lines, are distorted by the optical defects of the device, called aberrations. These aberrations depend mainly on the energy dispersion AV of the ions, and on the beam opening which is limited by the opening slit which is inserted most of the time before the magnetic sector.

For a given separating power, the best spectrometer is the most sensitive, that is to say the one that accepts the beam with the greatest geometric extent. This ability is called "clarity" of the spectrometer. However, for a given geometry of the spectrometer, clarity can only be increased by reducing the harmful effect of aberrations.

Finally, when one wants to make simultaneous measurements on all the lines of the spectrum (all the masses), the correction or elimination of the aberrations is much more delicate.

The problem is therefore to produce a mass spectrometer with high clarity, capable of simultaneous multiple detection and which has a high separating power.

To this end, a first object of the present invention is to correct the proper aberrations of the spectrometer, in particular of its magnetic sector, and also of its electrostatic sector.

A second object of the present invention is, using a transfer optic placed upstream of the mass spectrometer itself, to improve the adaptation and the transfer of the ion beam on the input of the spectrometer.

These aims are achieved by different aspects of the invention, some of which are interesting in themselves.

In known manner, the device proposed here comprises an inlet slot, followed by an electrostatic sector, then a magnetic sector. An opening slot can be inserted between the electrostatic and magnetic sectors or, in a conventional manner, at the entrance to the electrostatic sector. This assembly makes it possible to deflect a beam of particles in the radial plane perpendicular to the large dimension of the entry slit. As far as it is concerned, the magnetic sector has a planar entry face, inclined on the axis of the particle beam, and an exit face also planar, the plane of which passes through the intersection of the entry face with the axis of the particle beam. This is the arrangement of the actual magnetic faces which differ from the material faces due to the leakage fields.

Furthermore, it is also known, in a particular context, to interpose a quadrupole between the electrostatic and magnetic sectors (publication H. MATSUDA, MASS SPECTRO-METRY REVIEWS, Vol. 2, n ° 2 (1983) John Wi-lev pages 289 to 325). However, this quadrupole operates very differently from that which the present invention can use.

Finally, there are also known mass spectrographs of the Mattauch-Herzog type, in which the aperture on the one hand and the energy bandwidth on the other are delimited respectively using an aperture diaphragm and an "energy diaphragm" whose settings interact. It is then necessary to reduce the ranges of variation of the opening and the width of the energy band to relatively small values, if it is desired that the second order aberrations remain negligible. This results in a low value for the transmission or clarity of the device.

In the French patent published under No. 2,056,163 (US-A-3,622,781), it is proposed to transform a mass spectrograph of the Mattauch-Herzog type, so as to allow independent adjustment of the aperture and the energy bandwidth, while obtaining a higher value for the clarity of the device.

The means recommended by this prior French patent consists of having a first lens electric (18) between the entrance slit (10) and the energy diaphragm (20) and a second electric lens (22) between the energy diaphragm (20) and the magnetic sector (24). The Patent specifies the role of these lenses, in relation to the aperture diaphragm and the adjustable energy diaphragm.

In an exemplary embodiment given by this prior French patent (page 8), the input face of the magnetic sector (24) is inclined by an angle e which happens to be equal to 26.6 °. This use of an inclined entry face, well known elsewhere, makes it possible to define a focal point which is at a distance of twice the radius in the direction perpendicular to the plane of symmetry of the device. The focusing plane (26) is offset behind the magnetic field, by an angle ω which is here equal to 8.1 °.

The value of 26.6 ° for angle e corresponds to a standard deviation of 90 ° from the magnet. Of course, a value different from the deflection angle gives another value for the angle _E of the prior French patent.

It turns out that this value of 26.6 ° is close to that which the present invention will retain, for other reasons, and with a magnetic sector receiving a particle beam which is parallel in the radial plane.

However, in the previous French patent No 2,056,163, there is a focus "in the other direction": while in a normal mass spectrograph the particles which go up are lost, in the previous patent the lenses bring them back preventing the beam to move away. It therefore appears that the inclination of the entry face of the spectrograph is used in the prior patent to tighten the beam, and not to correct the aberrations, as the present invention will do.

According to a first characteristic of the present invention, the spectrometer comprises means such as a quadrupole having an object focus at the level of the real image that the electrostatic sector of the entry slot gives in the radial plane, to supply the sector magnetic a beam of particles which is parallel at least in the radial plane, and which has, moreover, for each energy of the band ± AV, the appropriate inclination so that the magnetic sector has an achromatic functioning at the level of all the lines of the mass spectrum. In addition, e noting the angle of the entry face with the normal to the axis of the beam which is situated on the deflection side thereof, and 0 noting the angle of deflection of the beam in the magnetic sector , these two angles satisfy the relation: tg θ / 2 tg (θ - ε) = 2 This makes it possible to cancel the second order opening aberrations created by the magnetic sector for the beam trajectories located in the radial plane (aberrations in a ² , where a is the inclination of a trajectory in the radial plane with respect to the central trajectory). Of course, the cancellation of these aberrations also occurs for the radial component of the trajectories which also have a vertical component. The trajectories in the vertical section (or the components, in the vertical section of any trajectories) also produce second order opening aberrations in the radial plane (aberrations in b ² , where b is the inclination of a trajectory by relative to the radial plane).

A second characteristic of the invention involves the transfer optic. This is arranged in cooperation with the spectrometer itself, so that in its vertical section, the particle beam has a neck between the entrance slit and the electrostatic sector; there is then provided, at this necking, a first hexapole, arranged to compensate for the second order opening aberrations created by the electrostatic sector for the trajectories located in the radial plane, aberrations at a ² , and for the radial component other trajectories. The location chosen for the hexapole means that the latter does not introduce ab ² aberrations from the opening in the vertical section.

Second order opening aberrations related to the electrostatic sector can be determined, this sector being for example of the spherical type.

Preferably, the transfer optic is arranged to apply a particle beam which is substantially parallel in vertical section to the entry slit. A converging lens is provided between the entry slit and the first hexapole (this converging lens being capable of providing post-acceleration). The hexapole is centered on the conjugate point of the entry slit by said converging lens, in the vertical section of the beam.

According to another aspect of the invention, the transfer optic comprises two electrostatic lenses, cooperating to produce a constriction of the beam at the level of the entry slot in the radial plane. Between these two electrostatic lenses is provided a lens with slits arranged so that the beam is parallel in vertical section at the level of the entry slit. It is then the converging (post-acceleration) lens, which ensures the convergence of the beam at the abovementioned point of necking, in vertical section.

The above concerns the correction of the opening aberrations for the trajectories located in the radial plane, at the level of the magnetic sector then of the electrostatic sector. For trajectories (or trajectory components) in the vertical section, there is no correction of the opening aberrations in b. However, the transfer optic is arranged so that the angles seen b by the spectrometer are always very small, so that the corresponding aberrations in b ² are negligible.

Yet another aspect of the invention concerns the correction of chromatic aberrations, that is to say in energy dispersion.

The electrostatic sector and the magnetic sector having centers of rotation chro respective virtual matics, the quadrupole is placed so as to combine with an appropriate magnification these two centers of chromatic rotation. The quadrupole combines the center of chromatic rotation of the electrostatic sector with the center of chromatic rotation of the magnetic sector corresponding to a given radius, that is to say, to a given mass. The quadrupole is also arranged so that each energy arrives on the magnetic sector with the appropriate inclination; it follows that the chromatic dispersion is completely canceled for the mass considered after leaving the magnetic sector while for the other masses the chromatic dispersion is canceled at their line. This correction occurs within the limit of the energy band ± ΔV defined by the means disposed upstream of the transfer optic or by a filtering slot placed in P3.

In practice, the quadrupole is arranged so that its focal point coincides with the real image that the electrostatic sector of the entry slot gives in the radial plane, this quadrupole being followed by means of compensating for its divergence in the section vertical, so that the particle beam is then parallel in its two transverse dimensions. The means for compensating for the divergence of the quadrupole in vertical section are advantageously a lens with slits.

There are still mixed opening and chromatic aberrations. To compensate for these, the device comprises a second hexapole, disposed after the electrostatic sector and substantially centered on the real image that the electrostatic sector of the entry slot gives in the radial plane.

This arrangement allows the reduction of mixed aberrations, for the trajectories located in the radial plane, with exact compensation for a chosen mass. For the other masses, the aberration is considerably reduced.

The chromatic, that is to say energy, filtering of the particle beam takes place here upstream of the transfer optic. In a variant, it takes place at the level of this second hexapole. This then comprises two hexapoles framing an energy filtering slot.

• - la figure 1 est une première vue schématique d'un spectromètre selon la présente invention, présentée comme une coupe dans le plan radial, cette figure comportant deux faisceaux parallèles arrivant avec des incidences différentes sur le secteur magnétique, ces deux faisceaux correspondant à des énergies légèrement différentes ± AV;
• - la figure 2A est une vue semblable à la figure 1, mais illustrant la séparation, dans un même faisceau monoénergétique, de particules possédant deux masses différentes;
• - la figure 2B est une vue partielle illustrant, en section verticale, la forme du faisceau en amont du secteur électrostatique SE 23 de la figure 2A;
• - la figure 2C est une vue illustrant, encore en section verticale, la forme du faisceau en aval du secteur électrostatique SE 23; et
• - la figure 2D est une vue en section verticale illustrant la forme du faisceau dans la direction que celui-ci présente en sortie du secteur magnétique SM 30;
• - les figures 3A et 3B sont des vues agrandies correspondant aux figures 2A et 2B;
• - les figures 4, 4A et 4B illustrent plus précisément un 5 exemple particulier de réalisation du spectromètre de l'invention;
• - les figures 5A et 5B illustrent, dans cet exemple, le fonctionnement avec post-accélération; et
• - les figures 6A et 6B illustrent, dans le même exemple, le fonctionnement sans post-accélération.

Other characteristics and advantages of the invention will appear on examining the detailed description below, and the appended drawings, in which:

• - Figure 1 is a first schematic view of a spectrometer according to the present invention, presented as a section in the radial plane, this figure comprising two parallel beams arriving with different incidences on the magnetic sector, these two beams corresponding to energies slightly different ± AV;
• - Figure 2A is a view similar to Figure 1, but illustrating the separation, in the same monoenergetic beam, of particles having two different masses;
• - Figure 2B is a partial view illustrating, in vertical section, the shape of the beam upstream of the electrostatic sector SE 23 of Figure 2A;
• - Figure 2C is a view illustrating, also in vertical section, the shape of the beam downstream of the electrostatic sector SE 23; and
• - Figure 2D is a vertical section view illustrating the shape of the beam in the direction that it presents at the output of the magnetic sector SM 30;
• - Figures 3A and 3B are enlarged views corresponding to Figures 2A and 2B;
• - Figures 4, 4A and 4B illustrate more specifically a particular embodiment of the spectrometer of the invention;
• - Figures 5A and 5B illustrate, in this example, operation with post-acceleration; and
• - Figures 6A and 6B illustrate, in the same example, operation without post-acceleration.

As previously indicated, the present invention relates to a charged particle separator, or mass spectrometer, with multiple simultaneous detection and high clarity.

In contrast to mass spectrographs, which use a photographic plate as their final detector, mass spectrometers do not necessarily need their detection zone, the focal exit surface of the magnetic sector, to be a plane.

However, this is the case in the spectrometer of the invention, which will be described first of all in general with reference to FIGS. 1 and 2.

The spectrometer has at its input a transfer optic 1. The nature of this can depend on the characteristics of the particle beam applied to the input or "point-source" S. The transfer optic 1 ends at level d an FE 20 entry slot, which constitutes the entrance to the mass spectrometer itself.

In a known manner, the spectrometer comprises, behind the entry slit FE 20, an electrostatic sector SE 23, then a magnetic sector SM 30, upstream of which an opening slit FO 29 is provided. This set of means has the function deflecting the particle beam in a radial plane perpendicular to the large dimension of the FE 20 input slot. The radial plane is that of FIGS. 1 and 2A.

We know that the main organ of a mass spectrometer is its magnetic sector, whose dispersive action depends on both the mass and the energy of each particle; this dispersive action manifests itself by trajectories in an arc of a circle, the radius of which is greater or less depending on mass and energy. It is known to associate with such a magnetic sector an electrostatic sector which precedes it, and which for its part has a dispersive action but only as a function of the energy of the particles. Both sectors are combined so that the dispersive action of the electrostatic sector compensates for the energy dispersive action of the magnetic sector. It then remains in principle, at the exit from the magnetic sector, only the dispersive action as a function of mass.

Furthermore, and although this is not classic, the publication Matsuda already cited shows the use of a quadrupole between the electrostatic sector and the magnetic sector. This is therefore also considered to be known, it being observed that the Matsuda quadrupole does not at all have the same function as that of the present invention.

Finally, it is also known to ensure that the magnetic sector SM 30 has a planar entry face 31, inclined on the axis of the particle beam, and an exit face 32 also planar, the plane of which passes through l intersection 33 of the entry face 31 with the axis of the particle beam. This arrangement has the advantage of providing the same deflection angle regardless of the mass. The deflection angle is equal to twice the angle of the exit face 32 with the axis of the particle beam at the entrance to the magnetic sector SM 30. It also follows that, for a beam parallel to the at the input, the particles at the output of the magnetic sector are focused on a plane PF 35 which also passes through point 33.

To obtain a spectrometer with multiple detection and with great clarity, at the same time as with great separating power (or excellent resolution), it is necessary to compensate for various aberrations which the elements of the spectrometer present, taken separately or in their combinations.

A first aberration is known as a second order aperture aberration of the magnetic sector. In short, this type of aberration lies in the fact that two symmetrical trajectories with respect to the central trajectory at the entrance to the magnetic sector will intersect after the sector at a point situated outside this central trajectory; the offset between the point of intersection and the central trajectory is proportional to the square of the angular inclination a of each of the intersecting trajectories relative to the central trajectory (hence the second order in a ² ).

A first aspect of the invention consists in correcting this type of second order opening aberration, at the level of the magnetic sector itself. We denote by 34 the normal to the axis of the particle beam which is located on the side of the concavity which will be impressed on the beam by the magnetic sector SM 30. We note by e the angle formed by the entry face 31 of the SM 30 magnetic sector with this normal 34. We note by 8 the beam deflection angle in the SM 30 magnetic sector. Quite unexpectedly, it has been observed by the inventors that the second order opening aberrations created by the magnetic sector, for the trajectories located in the radial plane, are canceled when these two angles satisfy the following relation: tg θ / 2 · tg (θ - ε) = 2

The present invention relates to a mass spectrometer with very high clarity, that is to say an apparatus accepting beams with the greatest geometric extent, and with simultaneous detection, which makes the correction of aberrations very delicate.

The invention recommends an appropriate inclination of the input face of the magnetic sector, so that it has an operation devoid of second order aperture aberrations for all masses - all beam trajectories located in the radial plane - provided of course that the aberrations in the opening of the electrostatic sector have been corrected beforehand.

• tg θ/2 . tg (θ - ε) = 2

The inclination e of the input face is, as a function of the angle of deviation 9 of the magnetic sector, given by the above-mentioned relation:

• tg θ / 2. tg (θ - ε) = 2

For a deviation of 90 ° in the magnetic sector SM, we arrive at tg e = 112, ie e - 26.56 °.

This relationship was established from the work of the Applicants, on the basis of the aberration coefficients which occur in the magnetic sectors for many configurations. By combining some of these aberration coefficients, we could find the above relationship, for which the second order aperture aberrations of the magnetic sector are corrected. This is the main point, which makes it possible to correct these aberrations whatever the geometric extent of the beam, which is essential as soon as the present invention aims to collect a beam as wide as possible.

There remains of course the secondary effect that the inclination of the input face releases the focusing plane (PF 35) from the magnetic field of the magnet, something previously known.

It has been observed that the French Patent No. 2,056,163 already cited shows a magnetic sector whose entry face is inclined by 26.6 ° for a deflection angle of 90 °. However, this prior patent uses this provision in a different context, and therefore in no way teaches those skilled in the art that the aforementioned relationship can be used to correct aberrations in the magnetic sector. On the other hand, the use of this same relation according to the present invention produces a correction of the opening aberrations of the magnet for the trajectories located in the radial plane, so that only the aberrations introduced by the electrostatic sector remain. , which can be corrected using a hexapole (HP22) located upstream.

Those skilled in the art will understand that this correction of the second order aperture aberrations linked to the magnetic sector itself is very important, and that it can naturally be used in other spectrometers than that described. in detail here.

The particle beam available at the outlet of the electrostatic sector SE 23 has a necking at a point P3 (FIG. 2A). In order to make the most of the first correction, means are provided downstream of this point P3 to ensure that the magnetic sector SM 30 receives a particle beam which is parallel in the radial plane.

This can be done using one or more quadrupoles such as QP 26, interposed between the electrostatic sector SE 23 and the magnetic sector SM 30. One way of doing this is to arrange the single quadrupole QP 26 so that its focus -object coincides with the point of necking P3. As indicated previously, the position of the quadrupole QP 26 is determined so that the inclination of each of the parallel beams corresponding to the various energies is appropriate to give an achromatic operation at the level of the lines located in the plane PF 35 and this simultaneously for all the masses. This property is illustrated in FIG. 1 which shows that two parallel beams corresponding to the energies V ± ΔV, focus at the same point on the plane PF 35; the ordinate is dilated to make the figure legible. As will be seen later, this point P3 is the real image that the electrostatic sector SE 23 of the entry slit FE 20 gives in the radial plane. Thus, FIG. 2A shows in the radial plane a parallel beam leaving the quadrupole QP 26.

FIG. 2C shows that the QP quadrupole 26, on the contrary, has a diverging action in vertical section. This divergent action is in turn compensated for by an electrostatic lens with slits LF 27. At the output thereof, there is therefore also a parallel beam, which crosses exactly the small dimension of the opening slit FO 29.

If we now examine FIG. 2A again, it appears that the parallel beam (represented single and, to simplify the figure, as corresponding to the average energy of the beam) supplied by the quadrupole QP 26 in the radial plane crosses without modification slot lens LF 27, to pass inside the large dimension of the opening slot FO 29. The comparison with FIG. 1 shows that this large dimension of the opening slot FO 29 allows the passage of the parallel beams various chromatic inclinations from the quadrupole QP 26, taking into account the energy dispersion existing between the electrostatic sector SE 23 and the magnetic sector SM 30.

From the foregoing arrangements, it finally results that the particle beam is parallel in its two transverse dimensions, downstream of the slit lens LF 27, until it is applied to the input face 31 of the magnetic sector SM 30.

It is known that the electrostatic sector SE 23 and the magnetic sector SM 30 each have a respective virtual chromatic rotation center. (The adjective "chromatic" is used here in relation to the energy dispersion). The particles following the central trajectory before entering the electrostatic sector and having an energy slightly different from the nominal energy of the beam will leave the electrostatic sector SE 23 with inclined trajectories. When the energy varies, these inclined trajectories seem to rotate around a point which is called the center of chromatic rotation.

Similarly, the magnetic sector SM 30 has a center of chromatic rotation, towards which particles with similar energies and the same mass must converge with the appropriate angle so that they end up after deviation, at the same point of the focal plane PF 35 and with the same angle (trajectories combined) whatever the energy in the ± AV band. We then obtain a complete compensation (first order) of the energy dispersion of the magnetic prism. It will be noted that there are as many centers of chromatic rotation as there are masses considered.

However, if, at the entrance to sector SM 30, the inclination of the trajectories is simply proportional to the deviation AV with an appropriate factor, the same for all the masses, they will still converge at the same point of the focal plane PF 35, for a given mass, but the trajectories of different energies will no longer have the same inclination.

According to another characteristic of the invention, means are provided for combining the two respective chromatic centers of rotation of the electrostatic sector SE 23 and the magnetic sector SM 30. The quadrupole QP 26 can do this in a very simple manner with magnification appropriate. This allows a complete correction of the chromatic or energetic dispersion of the particle beam for a mass, the quadrupole being moreover arranged to provide for the other masses, trajectories of different energies with the appropriate inclination.

We will now examine the correction of the second order opening aberrations which occur at the level of the electrostatic sector SE 23. This correction essentially involves a first HP 22 hexapole. However, this first HP 22 hexapole also occurs in close combination with the elements. spectrometer 2 proper as well as its transfer optic 1, it is now necessary to describe the whole of the spectrometer.

Reference will first be made to FIGS. 1, 2A, 2B, 3A and 3B for the description of the transfer optics as well as of the input of the spectrometer 2.

The beam of charged particles applied to the input of the transfer optic 1 is presented with a necking at point S. This beam of ions is composed of particles of different masses, which are animated by slightly different kinetic energies. We note as before V their average kinetic energy, which is prime in electron volts, and ± AV the energy dispersion.

The beam has in principle a symmetry of revolution at the point S. Such a beam can consist of secondary ions emitted by a sample subjected to a primary ion beam concentrated on its surface.

A first unipotential electrostatic lens LE 11 gives an image of the source point S at a point 51. Around this point PC plates 12 can be provided allowing the beam to be recentered on the optical axis.

After the first lens LE 11, and optionally the centering plates PC 12, a lens with slots LF 13 is provided.

FIGS. 2A and 3A show that this slotted lens has no effect on the trajectories of ions situated in the radial plane. On the other hand, in vertical section (FIGS. 2B and 3B), the slotted lens LF 13 makes these trajectories converge at a point of necking S2.

A second LE 14 electrostatic lens is placed after the LF 13 slit lens.

In the radial plane (FIGS. 2A and 3A), the lens LE 14 gives points S and 51 an image P situated at the level of the entry slit FE 20 and centered on the axis of the latter.

In the vertical section (Figures 2B and 3B), the lens LE 14 is placed so that its focus is substantially at point S2, this lens therefore providing sensitive rays or trajectories. ment parallel extending along the FE 20 input slot, according to its large dimension (Figure 3B).

In this way, the magnification at the level of the entry slit FE 20 in the radial plane is obtained by playing on the excitation potential of the electrostatic lenses LE 11 and LE 14.

For the trajectories situated in the vertical plane (FIGS. 2B and 38), an independent adjustment is obtained thanks to the lens with slits LF 13.

At the entrance to the spectrometer 2 proper, and behind the entry slit FE 20, there is provided on the one hand a converging electrostatic lens denoted PA 21, which allows a controlled post-acceleration, and then a first HP hexapole 22.

The PA lens 21 acts, in the vertical section of the particle beam, to produce a necking thereof at a point located upstream of the electrostatic sector SE 23. The first hexapole HP 22 is centered at this necking.

This HP hexapole 22 is arranged to compensate for the second order opening aberrations created by the electrostatic sector SE 23 for the trajectories located in the radial plane. At first order, it has no action, and therefore does not modify the trajectories located in vertical section. Furthermore, as already indicated, the hexapole does not introduce b ² type aberrations on the trajectories in the vertical section thanks to the fact that the necking of the beam in this section is located in the center of the hexapole .

The PA 21 post-acceleration lens plays another role. This role consists in modifying the opening angle for the spectrometer 2 proper. Correlatively, seen for the rest of the spectrometer, the necking produced by the transfer optic in P at the level of the entry slit in the radial plane, is transferred to P1 by the post-acceleration lens PA 21. This makes it possible to '' increase the clarity of the spectrometer after removing or correcting the most important aberrations. Post-acceleration brings ions from V energy to Vp energy.

In practice, the main-object plane of the post-acceleration lens PA 21 is located in the plane of the entrance slit FE 20, so that the spectrometer sees an entrance slit located at P1, in the main plane image of the PA 21 lens. For the spectrometer, the size of the Gaussian image has not changed. For a given separating power, only the opening angle available at the entrance increases.

As previously indicated, the spectrometer is arranged so that the image focal point of the post-acceleration lens PA 21 is located in the center of the HP hexapole 22, at point P2.

Post-acceleration also makes it possible to reduce the relative energy dispersion from AVN to AVNp, which leads to a decrease in mixed aberrations and aberrations in (AVNP) 2.

For reasons of convenience, the Applicants have chosen the ratio VNp of the order of a quarter, which implies, for negative ions of incident energy of ± 5 kV, to put all the conductors constituting the spectrometer and located downstream of the PA 21 post-acceleration lens at a voltage of ± 15 kV.

Then, the spectrometer comprises a second hexapole (HP 25), arranged after the electrostatic sector (SE 23), and centered on the real image which the electrostatic sector (SE 23) of the slot (FE 20) gives in the radial plane. . This allows the reduction of mixed opening and chromatic aberrations for the trajectories located in the radial plane, with exact compensation for a chosen mass. The fact that HP 25 is centered on P3 allows correction of mixed aberrations to be carried out without having to touch up the setting of the hexapole HP 22 which corrects opening aberrations (independence of the settings).

In the embodiment described, the energy filtering is carried out upstream of the transfer optic.

In a variant, it is carried out at the level of the second HP 25 hexapole. The latter then comprises two hexapoles framing an energy filtering slot (not shown).

After the second hexapole HP 25, there is the quadrupole QP 26, the lens with slots LF 27, the opening slot FO 29, and finally the magnetic sector SM 30.

FIGS. 1, 2A, 2C, 2D show certain details of the structure of the magnetic sector. This comprises a magnet, not shown, which cooperates with two pole pieces 32A and 32B, the shape of which is given by the views illustrated in the radial plane.

Independently of obtaining the various corrections already described, the invention makes it possible to considerably facilitate these corrections, by effecting them by adjustments which do not require displacement of the components of the spectrometer, and which are made as independent as possible from each other. .

• - Sont implantés tout d'abord le secteur électrostatique SE 23, qui est de type sphérique, ainsi que le secteur magnétique SM 30.
• - Le quadrupôle QP 26 est alors mis en place, pour respecter a priori la condition d'inclinaison appropriée des trajectoires chromatiques et la condition de parallélisme du faisceau entrant dans le secteur magnétique. La lentille à fentes LF 27 est ensuite mise en place, de manière à corriger la divergence du quadrupôle QP 26 dans la section verticale. La fente d'ouverture FO 29 est elle aussi mise en place, de même que le second hexapôle HP 25, lequel est placé au foyer du quadrupôle QP 26. On notera au passage que la structure en section droite des hexapôles HP 22 et HP 25 est illustrée sur la figure 4A, tandis que celle du quadrupôle OP 26 est illustrée sur la figure 4B.
• - En amont du secteur électrostatique SE 23, sont implantés 15 (en remontant le faisceau) l'hexapôle HP 22, centré sur un point choisi à l'avance comme foyer en section verticale, la lentille de post-accélération PA 21, la fente d'entrée FE 20, la seconde lentille électrostatique LE 14, la lentille à fentes LF 13, les plaques de centrage PC 12, et enfin la première lentille électrostatique unipotentielle LE 11.

To this end, the spectrometer is assembled and adjusted as follows:

• - The electrostatic sector SE 23, which is of the spherical type, is first installed, as well as the magnetic sector SM 30.
• - The quadrupole QP 26 is then set up, in order to comply a priori with the condition of appropriate inclination of the chromatic trajectories and the condition of parallelism of the beam entering the magnetic sector. The slotted lens LF 27 is then put in place, so as to correct the divergence of the quadrupole QP 26 in the vertical section. The opening slot FO 29 is also put in place, as is the second hexapole HP 25, which is placed at the focus of the quadrupole QP 26. It will be noted in passing that the structure in cross section of the hexapoles HP 22 and HP 25 is illustrated in FIG. 4A, while that of the quadrupole OP 26 is illustrated in FIG. 4B.
• - Upstream of the electrostatic sector SE 23, there are located 15 (by raising the beam) the hexapole HP 22, centered on a point chosen in advance as focal point in vertical section, the post-acceleration lens PA 21, the slot FE 20 input, the second LE 14 electrostatic lens, the LF 13 slotted lens, the PC 12 centering plates, and finally the first LE 11 unipotential electrostatic lens.

All the elements can thus be placed in predetermined fixed positions, which will not have to be modified subsequently.

• - Le quadrupôle OP 26 est réglé de façon à donner une inclinaison appropriée aux trajectoires dispersées en énergie issues du secteur électrostatique SE 23.
• - La lentille de post-accélération et ses annexes sont réglés de manière à amener le point de striction P3 au foyer du quadrupôle QP 26 et par conséquent au centre de l'hexapôle HP 25; ce réglage assure le parallélisme dans le plan radial du faisceau issu de OP 26 et permet de compenser d'éventuels défauts de position du quadrupôle QP 26.
• - La lentille à fentes LF 13 de l'optique de transfert est réglée pour amener le point de striction P2 au centre du premier hexapôle HP 22.

Additional adjustments are then made as follows:

• - The quadrupole OP 26 is adjusted so as to give an appropriate inclination to the trajectories dispersed in energy coming from the electrostatic sector SE 23.
• - The post-acceleration lens and its annexes are adjusted so as to bring the point of necking P3 to the focal point of the quadrupole QP 26 and therefore to the center of the hexapole HP 25; this adjustment ensures parallelism in the radial plane of the beam from OP 26 and makes it possible to compensate for possible position defects of the QP 26 quadrupole.
• - The LF 13 slot lens of the transfer optic is adjusted to bring the point of necking P2 to the center of the first HP 22 hexapole.

Those skilled in the art that the device is thus entirely solved without that there is no need ⁱ t disassembly for changing the relative positions of the various components.

There is also no need for adjustment at the level of the collection of deflected particles, at the outlet of the magnetic sector SM 30. In fact, the particles of the different masses all arrive on the same plane PF 35.

The particle separator of the invention would make it possible, like a spectrograph, to use a photographic plate for the collection of particles deviated, and having undergone mass analysis.

According to the invention, it is considered preferable to place in the focal plane PF 35 a series of separate collecting devices, such as electronic multipliers whose input surface will be sensitive to the impact of charged particles from the magnetic sector SM 30 .

We will now describe, essentially referring to Figures 4 and following, a particular example of implementation of the invention.

Specific example Input beam

Negative ions of average energy 5 kV, forming a beam of revolution which has a necking at the point S. The half-angle at the top is of the order of 10 ^-2 radians for a separating power M / AM of the order of 4,000.

Transfer optics 1

• - LE 11 electrostatic lens: three circular electrodes, pierced with a central hole 4 mm in diameter. The central electrode is brought to a potential of - 4670 volts. The other two electrodes, located on either side, are at ground potential.
• - PC 12 centering plates: four stainless steel plates, with active surface 18 x 2 mm, mounted to form a square channel. The distance between two facing plates is 3 mm. Their center is placed at a neck of the ion beam.
• - LF 13 slot lens: three electrodes with rectangular openings, the major axis of which is located in the radial plane. Central electrode: 6 x 64 mm; the other two electrodes: 4 x 30 mm. The central electrode, which receives a voltage of - 5 kV, is 66.5 mm away from the axis of the LE 11 lens.
• - LE 13 electrostatic lens: identical to LE 11, except central electrode raised to - 4310 volts. This electrode is located 36 mm from the center of the slot lens LF 13, and 30 mm from the entry slot FE 20 located downstream. Mass spectrometer 2
• - FE 20 input slot: this is a rectangular opening of 0.024 x 0.8 mm (with active post-acceleration), for mass resolution M / _A M - 4,000. The major axis is perpendicular in the radial plane. The slot is adjustable in x and in y.
• - PA 21 post-acceleration electrode: 8 mm thick disc, drilled with an internal diameter of 14 mm, and isolated by an alumina cylinder. This disc is brought to + 10 kV in post-accelerated operation, the downstream elements then having their voltage reference brought to + 15 kV.
• - Hexapôle HP 22: six cylindrical bars 8 mm in diameter and 36 mm in length, with axes regularly distributed over a cylindrical surface with a diameter of 24 mm. Potential difference between neighboring bars, ± 36 volts alternately. The center of the hexapole is 52 mm downstream from the PC 21 electrode and 43 mm upstream from the entry face of the electrostatic sector.
• - SE 23 electrostatic sector: deflection angle 90 °. Two concentric sphere portions of radii 94 and 106 mm. Potential difference applied between interior and exterior: ± 4819.8 volts. Guard slot on entry and exit to limit leakage of electric field.
• - Hexapole HP 25: like H 22, except length of the bars 72 mm, and potential difference ± 1702.5 volts. Its center is placed 77 mm downstream from the exit face of the electrostatic sector SE 23.
• - Quadrupole QP 26: four cylindrical bars 24 mm in diameter for 120 mm in length, with axes regularly distributed over a cylindrical surface of 46 mm in diameter. Potential difference: ± 46.4 volts, alternating, the bars with negative potential being those located in the radial plane.
• - LF 27 slotted lens: same general structure as LF 13; central electrode drilled to 14 x 70 mm, extreme electrodes drilled to 10 x 60 mm. The central electrode is brought to the potential corresponding to a focal distance of 173 mm, and placed 119 mm downstream from the center of the OP 26 quadrupole.
• - FO 29 opening slot: dimensions 5 x 0.7 mm for mass resolution M / AM - 4100; major axis in the radial plane.
• - SM 30 magnetic sector: magnetic circuit in soft iron with U-shaped section. 8 mm magnet gap. Radius of useful paths from 70 to 350 mm. Magnetic induction adjustable up to 1 Tesla.

• Angle of the input face e - 26.56 ° (tg ε = ½)
• Deviation angle 0 - 90 °
• Angle of the exit face θ / 2 = 45 °
• Focus plane angle PF 35, 53 ° 13

The magnetic circuit is at ground potential, but non-magnetic electrodes placed inside the air gap are brought to a potential of + 15 kV in operation in post-accelerated mode. Finally, a magnetic shunt limits the leakage of the field on the entry face of the magnetic sector. On the upper part of the focusing plane PF 35 is placed a mufticollector assembly consisting of ion-electron converters followed by electron multipliers.

The optical elements (except the magnet) are placed in a structural piece of stainless steel which ensures the mechanical positions of the various devices and serves as a vacuum enclosure. A cryogenic pumping unit makes it possible to obtain the desired ultra-vacuum. The magnet is connected to the previous device by an elastic and sealed system, which includes means for mechanical alignment of the optical axes.

Reference is now made to FIGS. 5A and 5B, and 6A and 6B, for a better illustration of the operations with and without post-acceleration, respectively.

Up to the LF 13 slot lens, the trajectories remain those already described, and are the same in the radial plane and in vertical section.

FIG. 5A shows that, in the radial plane, the trajectories pass without alteration through the slit lens LF 13, to join the second electrostatic lens LE 14, and focus in principle at point P. However, the post-acceleration lens PA 21, brought to 10 kV, produces a focal point apparent at point P1, downstream of the spectrometer. It is therefore from this point P1 that the trajectories applied to the HP hexapole 22 will start. For the numerical values previously described, the distance between points P1 and P, or more exactly between the planes PHi and PHO is 11 mm .

In vertical section (Figure SB), the trajectories have been modified by the lens with slits LF 13. They are therefore parallel to the plane PHi, after which they converge slightly to cross the opening slit FE 20 in the plane PHO, and then take their final orientation towards the point of necking P2 on which the HP 22 hexapole is centered.

• - Pour LE 11, f11 = 15 mm.
• - Pour LF 13, f13 = 9,44 mm.
• - Pour LE 14, f 14 - 19,88 mm.
• - Pour PA 21, f2l - 97 mm.

The focal lengths are as follows:

• - For LE 11, f11 = 15 mm.
• - For LF 13, f13 = 9.44 mm.
• - For LE 14, f 14 - 19.88 mm.
• - For PA 21, f2l - 97 mm.

In operation, without post-acceleration (FIGS. 6A and 6B), only the converging lens effect produced by the lens PA 21 remains. The point P therefore remains associated with a point P1, from which appear the trajectories that the found downstream of the lens PA 21 in the radial plane.

In vertical section (FIG. 6B), the settings of the transfer optics are modified so that the trajectories begin to converge as soon as they exit the second electrostatic lens LE 14. They pass through the entry slit FE 20, to further converge a little more at the level of the lens PA 21, and lead fially to the same point of necking P2 as previously.

• - f11 et f14 restent les mêmes
• - pour LF 13, f13 - 8,30 mm
• - pour PA 21, f21 = 139,09 mm

Alors que l'on obtient un grandissement unité au niveau de la fente d'entrée dans le mode avec post-accélération, le mode sans post-accélération fournit un grandissement de 1,32 entre l'image située en P1 et l'image située en P.The focal length values are a little different:

• - f11 and f14 remain the same
• - for LF 13, f13 - 8.30 mm
• - for PA 21, f21 = 139.09 mm

While a unit magnification is obtained at the entry slit in the post-acceleration mode, the mode without post-acceleration provides a magnification of 1.32 between the image located in P1 and the image located in P.

It thus appears that the post-acceleration, accompanied by the convergence effect linked to the additional electrode PA 21 is compatible with the independent adjustment properties possessed by the device according to the invention: only the electrical voltages are to be modified.

Of course, some of the components of the spectrometer described in detail can be replaced by their equivalents. The single quadrupole QP 26 could thus be replaced by two quadrupoles. Those skilled in the art know other equivalents to the quadrupole, as well as to the hexapoles, electrostatic lenses and slit lenses.

Claims (11)

1. A mass spectrometer comprising an inlet slit (FE 20), followed by an electrostatic sector (SE 23), then a magnetic sector (SM 30) preceded by an opening slit (FO 29), mounted to def lect a particle beam in a radial plane perpendicular to the large dimension of the inlet slit (FE 20), the magnetic sector (SM 30) having a planar intake face (31), inclined towards the axis of the particle beam and an outlet face (32) which is also planar, whose plane passes through the intersection (33) of the inlet face (31) with the axis of the particle beam, characterized in that it comprises optical means (QP 26) having an object focus at the level of the real image provided by the electrostatic sector (SE 23) of the inlet slit in the radial plane, to supply to the magnetic sector (SM 30) a particle beam which is parallel in the radial plane, these means being arranged so that this parallel beam has the appropriate inclination for each energy of the incident particles for the magnetic sector to operate achromatically substantially for all the particle masses, and in that, when e denotes the angle of the inlet face (31) with the normal (34) to the axis of the beam which is situated on the side of the deflection of the latter, and 9 denotes the deflection angle of the beam in the magnetic sector (SM 30), these two angles satisfy the relation,

tg θ/2·tg (θ - ε) = 2

which makes it possible to cancel the opening aberrations of the second order created by the magnetic sector for the trajectories situated in the radial plane.

2. A spectrometer according to claim 1 comprising an optical transfer unit (1) ahead of the inlet slit (FE 20), characterized in that, considered in its vertical section perpendicular to the radial plane, the particle beam comprises a constriction between the inlet slit (FE 20) end the electrostatic sector (SE 23), and in that provision is made at the level of this constriction for a first hexapole (HP 22) arranged to compensate the opening aberrations of the second order created by the electrostatic sector (SE 23) for the trajectories situated in the radial plane, without introducing any aberrations of the same order by its action on the trajectories in the vertical section.

3. A spectrometer according to claim 2, characterized in that the optical transfer unit (1) is arranged to apply a particle beam which is substantially parallel in the vertical section to the inlet slit, in that provision is made for a convergent lens (PA 21) between the inlet slit (FE 20) and the first hexapole (HP 22), and in that this hexapole (HP 22) is centred on the conjugate point of the inlet slit (FE 20) by the said convergent lens (PA 21) in the vertical section of the beam.

4. A spectrometer according to claim 3, characterized in that the convergent lens (PA 21) is at the same time a post-acceleration lens, comprising at least one additional electrode for ensuring the adjustable convergence.

5. A spectrometer according to one of claims 2 to 4, characterized in that the optical transfer unit (1) comprises two electrostatic lenses (LE 11, LE 14) which cooperate to produce a constriction of the beam at the level of the inlet slit (FE 20) in the radial plane, and between these two electrostatic lenses, a slotted lens (LF 13) arranged so that the beam is substantially parallel in its vertical section at the level of the inlet slit.

6. A spectrometer according to one of claims 2 to 5, characterized in that since the electrostatic sector (SE 23) and the magnetic sector (SM 30) have respective virtual chromatic rotation centres, a quadrupole-type optical means (OP 26) is positioned so as to join these two chromatic rotation centres with an appropriate magnification, which permits a complete correction of the chromatic dispersion of the particle beam for a given mass, the correction for the other masses being effected in the focal plane (PF 35) of the magnetic sector (SM 30).

7. A spectrometer according to claim 6, characterized in that the quadrupole (QP 26) is followed by means for compensating its divergence in the vertical section, so that the particle beam is subsequently parallel in its two transverse dimensions.

8. A spectrometer according to claim 7, characterized in that the means for compensating the divergence or the quadrupole in the vertical section comprise a slotted lens (LF 27).

9. A spectrometer according to one of claims 2 to 8, characterized in that it comprises a second hexapole (HP 25), disposed after the electrostatic sector (SE 23) and centred on the real image provided by the electrostatic sector (SE 23) of the inlet slit (FE 20) in the radial plane, which permits a reduction of the mixed opening and chromatic aberrations for the trajectories situated in the radial plans, with an exact compensation for a chosen mass.

10. A spectrometer according to claim 9, characterized in that the second hexapole (HP 25) comprises two hexapoles surrounding an energy filtering slit.

11. A spectrometer according to one of claims 1 to 10, characterized in that θ = 90° and tg e = 1/2.

Images