JPS63279549A - Surface analysis device - Google Patents

Abstract

PURPOSE:To obtain not only depth direction information but also plane direction information by two-dimentionally scanning proton microbeam small in its beam diameter while moving same perpendicularly, and using a Wiener filter for directing a scattering beam and a incident beam to different paths. CONSTITUTION:Protron ions from an ion source 1 pass through a Winer filter 2 and a slit 4 to reach an acceleration/deceleration tube 3, where they are accelerated. The beam is moved in X and Y directions by an X-direction scan electrode 5 and a Y-direction scan electrode 6 so that it impinges on a sample SIGMA. The protrons which have impinged on the atoms of the sample return via the same path such that they pass through the electrodes 6, 5 reversely and further pass through the tube 3 so that they may be decelerated. In this manner they are decelerated and further they pass the Wiener filter 2 reversely after passing through the slit 4 and turns in the direction toward the electric field. There is a constant relation between this deviation L and energy loss E. It is thus possible to obtain the energy loss from the deviation L and also to the mass M of the scattered atoms.

Description

DETAILED DESCRIPTION OF THE INVENTION (g) Technical Field The present invention relates to a surface analysis device that can provide information not only in the depth direction but also in a two-dimensional plane of a sample, and that occupies a small area.

A surface analysis device irradiates a sample with an accelerated proton beam, decelerates the proton beam scattered by collisions with sample atoms, and measures the energy distribution of this proton beam. A device used to identify existing elements and determine their abundance.

The internal crystal structure of the sample can be determined by X-ray diffraction.

The crystal structure near the surface of a sample can be investigated by electron beam diffraction.

However, with either method, the distribution of elements on the surface of a sample, such as a monoatomic layer or a diatomic layer, cannot be determined.

The present inventors proposed proton energy loss spectroscopy (Proton energy loss spectroscopy) as a method for measuring the amount of elements present on the surface.
on Energy Loss 5pectro
Scopy; PELS) was developed - (a) Principle of PELS Since it is a new measurement technology, the principle will be briefly explained.

In Figure 3, an atom of mass M is at rest.

On the other hand, a proton with mass m approaches with velocity U, and an atom M
, and fly at a speed V in the direction of an angle θ. It is assumed that the atom M is thrown at a velocity W in the direction of the angle Φ. From the law of conservation of momentum, mυ=mvcoSθ+MWc
osΦ (1)0 = mV sin f)
-MWsinΦ (2) holds.

Assuming a perfectly elastic collision, the law of conservation of energy holds true.

When W is deleted from (1) to (3), (/'+1)V -2UCO5eV-(r-1)U =
O(4). From here, it becomes 7'+1.

However, M7'=-(6). In (5)), the negative sign indicates scattering toward (π-θ). It is correct to take only the positive sign for the scattering directed at θ.

Protons lose energy through scattering. If the scattering angle is the same, the lighter the partner atom, the greater the energy loss. This tells us what the other atom is.

Letting the initial kinetic energy of the proton be EO, then.

Let the energy after the collision be El. This is always smaller than EO. This ratio is called the damping coefficient.

E l: K E O (8) It is.

θ is not a variable, but r is. θ is constant depending on the experimental equipment.

The inventors first developed a device with a low scattering angle (θ; 0). For example, JP-A-59-180945 (S, 5
9.10.15) and Japanese Patent Application Laid-Open No. 151958/1983 (published on 5, 61, 7, 10) disclose a PELS device with a low scattering angle.

However, as shown in FIG. 12, those with low scattering angles have the disadvantage that they are easily affected by surface irregularities. - In addition to double scattering, double scattering may also occur.

Furthermore, in the case of a low scattering angle, there is a drawback that the difference in K due to F is small and the resolution is low, as can be seen from equation (9).

The principle of PELS is clearly shown in equation (9), but
This is the attenuation due to -fold scattering. Multiple scattering must not occur.

The reason why a low scattering angle was chosen is that the yield is high. FIG. 13 shows the relationship between the scattering angle θ from Au and Si and the yield.

The relative ratio of yields is determined by a geometric relationship,
It does not depend on the physical properties of atoms.

From this figure, it can be seen that the yield reaches its maximum even when θ=180° (7).

Furthermore, if θ=π, it can be seen from equation (9) that the resolution can be the highest with respect to r.

At this time, the damping coefficient is set as θ=π in (9). r is the ratio of the masses of atoms M and protons m.

Although the atomic mass unit 1u and the proton mass m are slightly different, if they are considered to be almost the same, F can be called the mass number.

If a certain element is specified, r can be determined and K kami can be found.

Give an example.

In the case of Helno r=26.98 K=0.8621
For Ga r=69.72 K=0.9442
For 2As /'=749 K=0.9479
It is 9. It is easy to calculate K for all elements.

In other words, by measuring the final kinetic energy E1 of a proton, K can be determined from the ratio to EO. Then, by knowing r, we can find out the atom to which the proton was scattered.

The amount of atoms present can be determined from the energy spectrum.

In many cases, EO is about 100 keV.

The principle of PELS is thus simple.

For such measurements to be possible, it is necessary that the protons be scattered only once.

Also, instead of directly measuring El, the energy loss Δ
It is also possible to measure E=(1-K)EQ.

To measure the distribution of proton energy loss, PEL
There is a name for S.

(b) Depth analysis using shadow cone As shown in FIG. 4, when protons m collide with heavy atoms M and are scattered, a portion is created in front of the protons that the protons cannot reach. This is called a shadow cone. .

Protons that pass far away from the atom M are hardly scattered, and protons that pass close to the atom M are easily scattered. This creates a shadow cone.

The direction in which the protons travel is defined as the X-axis, and the direction perpendicular to this is defined as the y-axis.

If the repulsive force acting between an atom and a proton is a Kulonka, the formula for the shadow cone is given by. e is the charge element, 2 is the charge of the proton,
Z is the charge of the atom (in units of e).

By skillfully using the shadow cone, you can determine which atoms are made up of the first layer of the sample. I understand that.

In FIG. 5, it is assumed that a proton beam is vertically incident on the surface of a GaAs sample. Since it is scattered by both Ga and As, two peaks occur in the energy loss distribution, as shown in FIG. 6.

Next, as shown in FIG. 7, a proton beam is made obliquely incident 6, and the atoms in the second layer are made to enter the shadow cone formed by the atoms in the first layer.

In this case, only those corresponding to the atoms in the first layer appear in the proton energy loss distribution. As shown in FIG. 8, only the Ga peak appears. From these results,
It can be seen that the first layer of the sample is Ga.

) Depth analysis using electron collisions When a proton beam enters a sample, the protons lose energy due to collisions with electrons. Energy loss due to electron collision lIζ is proportional to the distance traveled within the sample.

This will be explained with reference to FIG.

There is a proton beam making an angle θ/2 with the surface layer. There is a difference in energy loss between the case where this is scattered at five points on the first layer and in the direction of θ/2 and the case where it is scattered at point G on the second layer. The energy loss difference is θ ΔF = 2 d S cosec − (12). d is the interlayer distance, and S is the electron stopping power.

The smaller θ is, the larger ΔF is. Even if θ=π, for example, if d=5.6 people and S=10 eV/person,
ΔF=112eV.

Even at normal incidence (θ=π), the energies of the scattered protons in the first and second layers differ by about 100 eV.

This means that even if the atoms M that are the main cause of scattering are the same, 1
This means that the energy losses in the first and second layers are different.

If θ is made sufficiently small and ΔF is made large, proton scattering from foreign atoms in the first layer and from foreign atoms in the second layer can be identified.

(3) Overview of measuring device FIG. 10 shows an overview of a conventional PELS measuring device when the scattering angle θ is 180°.

The proton ion beam extracted from the ion source A is subjected to mass analysis by a magnet B. Only monovalent proton ions are selected, enter the acceleration tube C, and are accelerated.

It has a kinetic energy EO which is the sum of the extraction energy EeX and the acceleration energy EaCC in the accelerating tube C, and this energy hits the sample Σ.

Protons are scattered by surface atoms of sample Σ.

Only those with a scattering angle of θ=π run in the opposite direction through the accelerator tube.

Those with θ≠πθ collide with the chamber wall, return from ions to neutral molecules H2, and are exhausted.

The one with θ=π runs in the opposite direction through the accelerator tube, so it is decelerated. In other words, it will now work as a speed reducer.

The deceleration energy Edec is the acceleration energy: Jf −E
Eacc = Edec (+3) The great advantage of the θ-π case is that the proton beam can be accelerated and decelerated by the same accelerator tube C1 deceleration tube.

The decelerated proton beam is bent at right angle by magnet E, further bent at right angle by magnet F, warped by 6 degrees, -C
, Yoko, Yoaka, Igar G Garlic. Iwasu.

The reason why the two magnets E1F are required is to converge the proton beam energy, which varies due to scattering loss ΔE, to one point.

The energy of the proton beam focused on one point is detected by the analyzer G.

A voltage VQ is applied between the electrode plates of the analyzer 7. The protons entering through the slits enter one of the channels of the microchannel plate H, drawing a parabola. Depending on which channel it enters, the distance from the slit to the position it descended to can be determined. From this, we can find out the kinetic energy of the proton when it is incident.

The greater the proton's kinetic energy, the greater its range in analyzer G.

The kinetic energy of the protons when they enter the slit of the analyzer G is Ea1 The angle between the proton beam and the parallel electrode at the time of incidence is 1, the parallel electrode distance is h1, the electrostatic voltage between the plates is VO
Then, .・The distribution of proton energy Ea can be found from the distribution of range. In other words, it is possible to measure the energy distribution of protons.

The apparatus of Figure 10 is all in high vacuum. In particular, a certain part of the sample Σ must be under ultra-high vacuum. This is because if there is no vacuum, protons will collide with gas molecules, resulting in energy loss, making it impossible to know the scattering loss ΔE due to the sample Σ.

For the sake of simplicity, the vacuum chamber and vacuum device are not shown in FIG. 10.

(b) Energy change of protons The energy change of protons will be explained with reference to FIG.

The position of the proton moves from left to right. This represents potential energy.

The ion extraction voltage of protons at the ion source is Vex.

If the proton charge is q, the potential energy due to this is qVex. When the ion source leaves the source, this amount of energy has become kinetic energy.

The accelerating voltage at accelerating tube C is Vacc. The kinetic energy just before it leaves the accelerator tube and hits the sample Σ is q(Vex +
Vacc). This is Eo. Approximately 100k
The proton beam hits the sample and is scattered. Let the energy loss due to scattering be ΔE.

It then moves backwards through the accelerator tube, losing kinetic energy equal to qVacc.

The kinetic energy Ea when it enters the analyzer G is Ea==qVO−ΔE (+51, where vO is equal to the extraction voltage Vex.

It can also be said to be the incident energy to the analyzer G assuming that the energy loss is 0.

Ea is about 0.5 keV, and the scattered proton energy E.

E s = q (Vex + Vacc) - ΔE
(1c9 is about 100 keV.

E should originally be a variable. However, the target atom is determined in advance, and E is applied to it.

Vacc is determined so that Q is approximately 5 keV.

(+) Entire P E L S device FIG. 14 is a schematic perspective view of the entire PELS device.

The proton beam emitted from the ion source is focused by an Einzel lens, bent by a magnet, and accelerated by an acceleration/deceleration tube.

A sample is set in an ultra-high vacuum chamber and evacuated. The sample can be manipulated using a manipulator.

The accelerated proton beam is focused by a Q lens and hits a sample in an ultra-high vacuum chamber.

Among the protons scattered on the sample surface, those with θ=180° exit the ultra-high vacuum chamber and are decelerated.

This is bent 1800 degrees by magnet 1.2 and enters the analyzer. Here, energy loss ΔE is measured.

(1) Problems with conventional technology The outline of the PELS device has been described above.

The proton ion beam that hits the sample has several diameters.

The planar position resolution of the sample is limited by the proton beam spot size. Since the beam diameter was several degrees, the position resolution was also several degrees.

However, depending on the purpose, there may be cases where it is desired to observe the atomic arrangement state on the surface of the sample with higher resolution.

Another problem is that the entire device is too large.
That's what it means.

Since it is a 180° scattering, the incident beam and the outgoing beam must be separated. This cannot be done by the action of an electric field. Must use Lorentz force and requires Magbot.

Proton ions emitted from the ion source are held at 90° by a magnet.
Deflect it and hit the sample. The beam that has been scattered by the sample and returned is deflected by 90 degrees, and in some cases further deflected by 900 degrees, and is made incident on the electrostatic analyzer 〇.

Therefore, the line connecting the ion source, magnet, and electrostatic analyzer is perpendicular to the line connecting the magnet and sample. Furthermore, such devices are co-horizontal, and the proton binim moves on the horizontal plane.

For this reason, the arrangement of the components of the device is complicated and distributed in a horizontal plane, resulting in an excessively large area occupied by the device. The problem was that it took up too much space.

The conventional device required an area of 2 m wide x 3 m deep.

A first object of the present invention is to provide a surface analysis device with high plane position resolution. In other words, it is a surface analysis device using the PELS principle that can provide information not only in the depth direction but also in the planar direction.

A second object of the present invention is to provide a surface analysis device that occupies a small area.

←) Configuration The PELS device of the present invention does not move the proton beam horizontally, but moves it vertically from top to bottom and from bottom to top.

In addition, a microbeam with a small beam diameter is used to increase the plane position resolution.

In order to easily obtain two-dimensional position information, the proton beam is scanned two-dimensionally.

Furthermore, the present invention provides an apparatus that eliminates the need for magnets to bend the beam by 90 DEG or 180 DEG in order for the scattered beam and the incident beam to take different paths.

For this purpose, a Wien filter is used.

FIG. 1 shows an outline of the surface analysis apparatus of the present invention.

The ion source 1 is an ion source that can generate a microbeam. For example, a gas field ionization type ion source that can handle hydrogen beams is used. According to this, the proton beam spot size can be reduced to 1 μmφ or less.

The ion source is installed facing downward and emits proton ions downward.

The proton beam travels downward and passes straight through the Wien filter 2.

In FIG. 1, the X-axis and Y-axis are placed on the horizontal plane, and the two axes are placed vertically upward.

In this coordinate system, the beam can be said to be flying antiparallel to the Z axis. Expressed as the Z direction.

It passes through the Wien filter 2 in two directions.

The Wien filter 2 is a filter in which an electric field E and a magnetic field °B are generated at right angles. This allows only charged particles with a specific velocity 7 to pass straight through, while charged particles with other velocities V are bent in the direction of the electric field.

The degree of bending depends on the deviation of the speed V from the reference value.

A charged particle undergoes Lorenkar due to the electric field and magnetic field, but in the MKS unit system, F=Q(Ex\XB) (17). For a certain speed, V=. It is from.

E+ (voxB)=o (19). Particles with other velocities cannot travel straight through the Wien filter, and their paths are bent.

The remaining force F is ? = q (blind x 100) (2φ however, u = v − vo @l).

The directions of the electric field and magnetic field of the Wien filter are orthogonal.

Assume that the electric field is in the X direction and the magnetic field is in the Y direction.

Let the proton velocity vector component be (u, v%W). The electric field is (E, 0.0) and the magnetic field is (0, B10). Regarding the force components F,, Fy, FZ, (Written Equation 14 specifically, F'x=q(E-wB) HF'
y=o hard F'z=quB (23). It can be seen that the straight-line speed vo is as follows. There is only a Z-direction component, and this magnitude is E/B. This is a condition that must be satisfied in the beam passing through the Wien filter from top to bottom. In this coordinate, the Z axis points upward. The velocity of the downward beam is negative, so the two-component E/B of 241 is a negative number.

Since it is complicated to express it as a vector, the straight-line velocity of the Z component is called wo, and hereinafter it is sometimes used to mean 'o=E/B.

When passing through the Wien filter 2 from top to bottom, it is possible to go straight under this condition.

However, when passing from the bottom to the top, it is not possible to go straight through the Wien filter even if the absolute values are the same. It will bend in the X direction.

how to bend? I will talk about that later.

Since it is bent by a Wien filter, it is possible to separate the incident (downward) beam from the scattered (upward) beam.

The acceleration/deceleration tube 3 has a plurality of electrodes arranged in the direction in which the proton beam travels, and an electrostatic voltage is applied thereto.

When passing through the acceleration/deceleration tube 3 from top to bottom, it is accelerated. Passing from the bottom to the top will slow you down.

The slit 4 is provided in the middle of the proton beam path.

This allows only protons H to pass through. H from the ion source
In addition, ions such as H2H2 may also be released. These have different mass and charge from protons H, so their kinetic energy,
The speed is different from that of protons.

These must be removed. Vienna filter 2
Since these are bent instead of going straight, they hit the rear surface of the slit 4. The role of the slit 4 is to remove ions other than H.

The scanning electrode 5 in the X direction generates an electric field EX in the X direction, and has the effect of swinging the proton beam in the X direction.

The scanning electrode 6 in the Y direction generates an electric field EV in the Y direction, and has the effect of swinging the proton beam in the Y direction.

There is a stage 7 at the lowest end of the proton beam path.

Place sample Σ on top of this. Stage γ is x, y1z
You can move in the direction. In other words, sample Σ is X, 7%
It can be moved in the Z direction.

The entire device is housed in a vacuum container.

The degree of vacuum in sample Σ is the highest. 10-10~10-
',' This is an ultra-high vacuum of about Torr.

The degree of vacuum in the ion source 1 is low (10-' to 10-5 Torr).
That's about it. The space from the ion source 1 to the sample Σ is also maintained at a high vacuum.

Since the required degree of vacuum differs, the spaces are partitioned by plates with small holes, and each space is equipped with an evacuation device. Then, each space is differentially pumped.

(b) Operation The ion source 1 generates an extremely narrow proton ion beam. The diameter of the beam 7 is 1 μmφ or less. The conventional one is 0.2
Since it is MM~0.5ar*φ, it is extremely thin compared to this.

Proton ions emitted from the ion source travel along a path extending in the Z-axis direction. Therefore, this path is called the Z-axis path.

The Wien filter 2 collects proton ions from the ion source.
- Only the vehicle moves straight ahead. Therefore, even if H2+ and H2++ are present, they are bent by the Wien filter 2, hit the slit 4, and are removed.

Only the proton beam passes through the Wien filter 2,
Pass through slit 4. The kinetic energy up to this point is q V
It is ex. Vex is the ion extraction voltage.

Furthermore, it passes through the acceleration/deceleration tube 3 in the acceleration direction and is accelerated. An accelerating voltage Vacc is obtained, resulting in a speed of about 100 keV. Kinetic energy is q (Vex + Vacc
).

Up to this point, the beam is pointing vertically downward.

Furthermore, there are a scanning electrode 5 in the X direction and a scanning electrode 6 in the Y direction, and when passing through these, the beam is swung in the X direction and the Y direction.

The proton beam further impinges on the sample Σ placed on the stage 7. By swinging the beam in the X%Y direction, the surface of the sample Σ can be scanned over a wide range in the X%Y direction.

The protons that collided with atoms of the sample and were scattered by 180° return along exactly the same path in the opposite direction and pass through the scanning electrode 6.5 in the opposite direction.

Since the electric field has time reversal invariance, it can pass through the electrode in the opposite direction.

Of course, since there is energy loss ΔE, the direction is not completely opposite.

Further, the protons pass through the acceleration/deceleration tube 3 in the deceleration direction and are decelerated.

Deceleration voltage and acceleration voltage are equal.

The decelerated protons pass through the slit 4 and pass through the Wien filter 2 in the opposite direction.

The Wien filter 2 allows only particles with a downward straight speed wo to travel straight. Particles with upward velocity do not satisfy the straight-line condition. Therefore, when passing through the Wien filter, protons are bent in the direction of the electric field.

The degree of curvature depends on the speed W. This is due to kinetic energy, but also due to energy loss ΔE.

Therefore, at point W, the protons are biased towards the position detector 8. There is a certain relationship between the amount of bias and energy loss ΔE.

In other words, if the protons arrive at a position separated by L from the Z axis in the position detector 8, a unique relationship exists between the amount of bias and ΔE. ΔE can be found from the deviation amount,
The mass M of the atom to be scattered is known.

By performing such measurements while swinging the proton beam in the X and Y directions, the local distribution of atoms on the sample surface can be determined.

Since the range of X and Y scanning is limited, only planar information on a portion of the sample surface can be obtained.

In order to broadly obtain planar information of the entire sample, stage 7
, move the sample in the x1Y direction, and perform the same measurement again.

By repeating such measurements, the atomic distribution over the entire surface of the sample can be simulated.

FIG. 18 shows the beam scanning operation of sample t[ having a surface composed of A and B atomic weights with the PELS apparatus of the present invention. Scanning is done from left to right. The incident beam hits the sample and the scattered beam travels upward from the sample.

FIG. 18 shows the case where the beam and the sample surface are at right angles. This is to investigate the elemental composition of the first layer.

FIG. 18(b) shows the relationship of the A atom to the position of the spectrum. Since the surface of the central part of the sample is covered with the B atomic layer, the A atomic spectrum is approximately 0 in the central part and has large values at both ends.

In this way, the horizontal axis can be used as the position coordinate. It is not energy loss ΔE. The original PELS only measured ΔEf at a certain position, but in the present invention, the beam can be scanned, so local changes can be detected in this way.

FIG. 18(c) shows changes in the B atom spectrum depending on the position. It can be seen that there is a layer of B atoms in the center of the sample.

This uses a beam perpendicular to the sample to examine the first layer.

In FIG. 19, the proton beam is applied obliquely to the sample. This is to investigate the composition of the second layer. The stage is tilted to obtain this angle of incidence. The scattering angle is 180°.

FIGS. 19(b) and 19(C) are spectra of A atoms and B atoms in the second layer. It can be seen from (C) that B atoms are also diffused into the second layer. The decrease in the central part of the A atom spectrum in (b) is due to the diffusion of B atoms.
This means that some percentage of atoms are excluded from the second layer.

(■Energy analysis using a Wien filter The motion of protons in a Wien filter is explained using Figure 2.

The electric field exists in the X direction. Parallel electrodes are provided facing each other so as to be parallel to the yz plane.

The magnetic field exists in the Y direction. The magnets are provided facing each other so that their end faces are parallel to the X2 plane.

Consider a coordinate system whose origin is the center O of the Wien filter 2 space. Take the three points U and OlW from the top on the Z axis.

It is assumed that uniform electric and magnetic fields exist in the space surrounded by the electrodes and magnets.

On the Z-axis, U~0~W is the internal space of the Wien filter.

Protons emitted downward from the ion source 1 travel straight along the Z-axis of the Wien filter in the direction of U, O, and W.

The proton scattered beam passing through the acceleration/deceleration tube 3 from bottom to top enters the Wien filter from the lower end point W of the Z-axis of the Wien filter.

For protons moving from bottom to top, electric field E exerts a force in the X direction. Magnetic field B also exerts a force in the X direction.

The force of the electric field E and the force of the magnetic field B are almost the same.

However, the force due to magnetic field B has energy loss, so
It is slightly smaller than the force due to the electric field E.

In any case, the protons are pushed in the X direction in the Wien filter with a force that is approximately twice as strong as the force due to the electric field E.

Due to this force, the proton path deviates from the Z axis and tilts in the X direction, passing through the Wien filter from a point T on the top surface of the Wien filter.

After this, it goes straight and enters one of the microchannels of the position detector 8. Let L be the distance between this microchannel and the Z axis.

Let S be the height of the Wien filter 2 in the Z-axis direction. Let N be the distance in the Z direction from the center of the Wien filter 2 to the detection surface of the position detector 8.

The motion of the scattered beam within the Wien filter is not simple.

Therefore, the movement of protons passing through the Wien filter from bottom to top will be explained below.

Since the mass of the proton is m, the equation of motion is (2+i~
From (23), m5=q(E-wB) (25) m cloudy=
q u B (26j. Since the electric field E is a constant, this can be erased from 0.

(25) and (26) are initial conditions, t=o, u=o, W=
Solve with the condition Wl.

This is easy to solve, w-wO==(wl-wO) C05(Ωt)
'(27) u = (wl-wO) sin (Ωt
) C28) B Ω=□ Shikuto). Wl is the speed of the scattered protons in the Z direction and is positive. WO is the speed of the incident proton in the Z direction and is equal to E/B, which is a negative value. In Figure 2, E is positive and B is negative. wl is smaller than the absolute value of wO, 1wol, by the amount of energy loss ΔE. However, it should be noted that (wl - wo ) is almost twice as large as 1 wol. (27), it can be seen that (w-wO) + u2-(wl-wO)2 (time) from within. Fig. 15 is a velocity space diagram showing the relationship between velocity vectors (u, w). It's a circle.

If the internal space of the Wien filter 2 is sufficiently wide, the proton velocities u and w will change clockwise around this circle from the point w==wl. This circle is centered at WO and has a radius (wl
−wo ), but wo is negative and wl is (
-WO), so the diagram looks like this.

The value of W decreases and the value of U increases.

Figure 16 shows what kind of trajectory a proton draws when its velocity changes as shown in Figure 15.

This is a virtual trajectory. If the Wien filter extends infinitely and E and B exist, the protons will eventually change from upward motion to downward motion. It continues to rise and fall.

It's not just a rotation, it's descending at the speed of WO. The rotational speed is Ω, which is the cyclotron angular frequency.

Since the Wien filter 2 is actually narrow, the movement of the protons ends in the first part of this trajectory.

In Fig. 15, it jumps out of the Wien filter from wl to point E, and in Fig. 16, it jumps to point E.

FIGS. 15 and 16 show that the trajectory WT of protons within the Wien filter is not a simple parabola or circle.

(integrate 2711(28), t = o, X = O1Z
If we set the initial condition of = -s/2 (point W), it becomes 2 Ω. Since it moves in a straight line after leaving the Wien filter 2, the values of X, w, and u at the exit point are required.

Since Z = s/2 at point T, the time t when reaching point T is s ==-(wl-wO) sin (Ωt) +w
It is determined by Ot (33)Ω.

Since this cannot be solved analytically, we will assume that Ωt < 1 and take only one item, 5In(Ωt). When calculating with a computer, t can be found by solving equation (33) as it is.

Then, with such an assumption becomes. This drops the quadratic or higher order terms of (Ωt).

The calculations below are correct only within this approximation.

Then, the Y coordinate at point T is the same approximation, and the velocity U at point T in the X direction is. From point T, the protons go straight and enter a point on the position detector 8. The distance from the Z axis is L = X + u (N −-) / wl using a similar approximation.
(37) This is the time required to reach the detector 8.

In the end, the incident position of the position detector 8 is That's what it means.

Wl is the velocity of the scattered protons in the Z direction. If there is no energy loss ΔE, wl=1wOl. The upper limit of Wl is determined by this. This corresponds to the limit where the mass M of the other atom is infinite.

The lower limit of wl is 0. Collisions with lighter elements can reduce wl.

FIG. 17 shows the range of the emitted proton beam when it is emitted from the Wien filter at point T.

Curve A shown by a two-dot chain line is the limit of wl=1wol. In other words, this is the limit where the mass number M of the atom is infinite and ΔE is 0.

Curve C shown by a dashed-dotted line is the limit of wl=min(wl). In other words, the partner atom is a light element, and the loss Δ
This is the limit when E is large.

Normally, the signal exits the Wien filter through the curved opening surrounded by curves A and C.

From formula (38), L is smaller than when E and B are alone,
It turns out that it is about twice as large.

Even curve A with the least curvature can be sufficiently far away from the Z axis. Therefore, the Wien filter can separate the incident beam and scattered beam.

(2) Energy resolution WO and wl are related to energy by the following equation.

−m w12= (q Vex−ΔF )
(40) Get something from this. When L is differentiated by (ΔE), it becomes. The change is 1/W1 to the third power. The fact that the resolution is particularly high for atoms with a small mass number M (e) is effective.
Consequences (1) Using a microbeam increases positional resolution.

(11) Since the proton beam is scanned on a two-dimensional plane,
Two-dimensional information about the sample can be obtained.

(Since a 110 Wien filter is used, it is easy to separate the upward and downward beams. There is no need to install a magnet on the input side and output side (1 or 2). The beams exist in the vertical direction. All The devices are also arranged vertically.Also, they are arranged vertically in a linear manner.Therefore, the area occupied by the devices can be reduced.

For example, it will be possible to fit it into an area of 2m x 2m.

[Brief explanation of drawings]

FIG. 1 is an overall schematic diagram of the surface analysis apparatus of the present invention. FIG. 2 is a perspective view for explaining the movement of a proton beam within the Wien filter. FIG. 3 is an explanatory diagram of the velocity relationship before and after the proton m collides with the atom M. FIG. 4 is an explanatory diagram of a shadow cone that occurs in front when protons are scattered by atoms. FIG. 5 is a GaAs crystal surface diagram for explaining normal incidence scattering in PELS. FIG. 6 is a PELS proton energy loss diagram in the incident direction of FIG. 5. Figure 7 shows G for explaining oblique incidence scattering in PELS.
aAs surface diagram. FIG. 8 is a PELS proton energy loss diagram in the incident direction of FIG. 7. FIG. 9 is a scattering cross-sectional view illustrating that the first layer and the second layer can be distinguished by an increase in proton energy loss due to electron collision. Figure 10 is a diagram showing the principle configuration of PELS. Figure 11 is a proton potential energy configuration diagram. FIG. 12 is an explanatory diagram of a scattered beam at a low scattering angle. Figure 13 shows the difference in yield depending on the scattering angle for Au and Si.
Graph showing about. FIG. 14 is a schematic perspective view showing the actual PELS device. FIG. 15 is a diagram showing temporal changes in the X-direction velocity component u1 and the Z-direction velocity component W, assuming that the space inside the Wien filter extends infinitely. FIG. 16 is a diagram showing the movement of protons assuming that the space inside the Wien filter extends infinitely. FIG. 17 is a diagram showing that the angle at which the scattered protons are obliquely exited from the Wien filter varies depending on the value of the Z-direction velocity W1 of the scattered protons when entering the Wien filter. FIG. 18 is a diagram showing a method of examining the atomic distribution on the surface of a sample in which a B atomic layer exists on an A atomic layer by applying a beam perpendicularly to the sample. FIG. 19 is a diagram showing a method of examining the atomic distribution in the second layer of a sample by applying a beam obliquely to the sample. 1 ・・・・・・・・・・・・ Ion source 2・・・・・・・・・Wien filter 3...
・・・・・・・・・Acceleration/deceleration tube 4 ・・・・・・・・・・・・ Slit 5・・・・・・・・・X scanning electrode 6・・・・・・
...... Y scanning electrode 7 ...... Stage 8 ...... Position detector A
・・・・・・・・・・・・Aeon Source B・・・・・・
・・・・・・Magnet C・・・・・・・・・・・・
Acceleration pipe D・・・・・・・・・Deceleration pipe E%
F・・・・・・Magnet G・・・・・・・・・Analyzer H・・・・
・・・・・・Position detector L・・・・・・・・・・・・
Range of the analyzer (distance from the Z-axis of the detection point of the position detector) M・・・・・・・・・Atomic mass m・・・・・・・・・Proton mass N・・・・・・・・・Distance in the Z direction from the center of the Wien filter to the position detector 0・・・・・・・・・・・・Center of the Wien filter (
Origin) U......Incidence point of protons emitted from the ion source into the Wien filter W......Incidence of the scattered beam into the Wien filter, White Σ・・・・・・・・・Test
Fee W・・・・・・・・・Z-direction velocity component U...
......X-direction velocity component wO...
... Z-direction velocity component wl of protons emitted from the ion source ... Z-direction velocity component Vex of protons after scattering deceleration ... Ion extraction voltage Vacc ... Accelerating voltage Ea... Voltage ΔE when incident on the analyzer
...Energy loss Vo due to scattering...
・Analyzer voltage inventor Masahiko Aoki

Claims (2)

[Claims] (1) An ion source 1 kept in a vacuum that generates a microbeam of protons, and an electric field E and a magnetic field B in a vacuum that are perpendicular to each other so that the proton beam generated from the ion source 1 travels straight. A Wien filter 2 in which an electric field E and a magnetic field B are set, a slit 4 in a vacuum that passes the proton beam emitted from the Wien filter 2, and excludes charged particles other than protons; The proton beam that has passed through is accelerated and applied to the sample Σ and scattered by the sample Σ. Scattering angle θ = 1
An acceleration/deceleration tube 3 that passes an 80° proton beam in the opposite direction to decelerate it, and an ultra-high vacuum chamber that keeps the sample Σ in an ultra-high vacuum.
A stage 7 that movably holds the sample Σ and an acceleration/deceleration tube 3
An X scanning electrode 5 and a Y scanning electrode 6 are provided between the proton beam and the sample Σ to two-dimensionally scan the proton beam on the sample surface. A surface analysis device comprising a position detector 8 provided in a vacuum on an oblique side of a Wien filter 2 in order to detect bent protons incident thereon. (2) The surface analysis apparatus according to claim (1), wherein the ion source 1 is a gas field ionization type ion source, and the proton beam spot size is 1 μm or less.