JPH0310190A - Calibration of energy for charged particle - Google Patents

Abstract

PURPOSE:To establish a calibration of energy of charged particles compensated for linearity by using a target less than a thickness at which a difference becomes equal between a resonance energy on the side of a high energy and that on the side of a low energy adjacent thereto. CONSTITUTION:When energy of charged particles is calibrated by using a nuclear resonance reaction between charged particles having multiple resonance energy and a target substance, a target is used which is thinner than a thickness at which a difference becomes equal between an amount of energy which is attenuated from the charged particles having resonance energy on the side of a high energy and resonance energy on the side of a low energy adjacent thereto. Then, after the charged particles having the resonance energy on the side of a high energy pass through the target, they hold an energy exceeding the resonance energy lower by one level. Therefore, as no resonance lower by one level will not occur in the target, counts of radiation fall down to a background level thereby enabling measuring separating a peak of a resonance point on the side of a high energy.

Description

[Detailed Description of the Invention] [Summary] This invention relates to a method of calibrating the energy of charged particles and the acceleration energy of an accelerator.

By separating multiple resonance points in nuclear reactions.

The purpose of this study is to obtain an energy calibration method for charged particles that guarantees linearity.

When calibrating the energy of a charged particle using a nuclear resonance reaction between a charged particle with multiple resonance energies and a target material, the amount of energy attenuated in the target by a charged particle with a resonance energy on the high energy side is calculated. ,
A target thinner than the thickness at which the difference between the resonance energy on the high energy side and the resonance energy on the adjacent low energy side is equal is used.

[Industrial application field]

The present invention relates to a method for calibrating the energy of charged particles and the acceleration energy of an accelerator.

In recent years, as the performance of semiconductor devices has improved, high-energy implantation of about 3000 KeV has become required in the ion implantation process.

The present invention can be applied to calibrate the acceleration energy of such a high-energy accelerator.

[Conventional technology]

Acceleration energy is usually calibrated using resistance division for accelerators below 200 KeV, but when this is applied to high energy accelerators above 3000 KeV, the error due to resistance division cannot be ignored, and the energy error becomes It's around a few 10%.

Energy errors directly result in errors in the implanted impurity profile, making process control impossible.

Therefore, a nuclear resonance reaction is used, which has a high accuracy of absolute value (the reaction occurs only in a very narrow energy width range of several KeV at most).

The conventional energy calibration method for charged particles using a prompt nuclear resonance reaction involves colliding a target that causes a nuclear reaction with charged particles accelerated in an accelerator, and using a scintillation counter to detect radiation such as gamma rays or neutron beams that are immediately generated. The number of counts was measured by changing the acceleration energy, and the resonance point (peak value of the number of counts) unique to the nuclear reaction system was investigated and calibrated.

However, in this method, in order to guarantee linearity between the energy of charged particles and acceleration energy, it is necessary to We had to check the resonance points of the two points and adjust the gain and offset of the accelerator so that the straight line connecting these two points had a 45° inclination passing through the origin, but we had no knowledge of the thickness of the target. As shown in Figure 2, the peak position of the resonance point on the high energy side was not known.

FIG. 2 is a diagram showing the measurement results of resonance points of nuclear reactions according to a conventional example.

The figure shows the number of counts of radiation generated by the nuclear reaction of protons (charged particles) and AI (target) with respect to the energy of the charged particles.

[Problem to be solved by the invention]

As can be seen from FIG. 2, for nuclear reactions with two or more resonance points, in the conventional method, only one resonance point on the low energy side could be measured, making it impossible to calibrate the energy of charged particles.

The reason for this is that the second and third peaks on the high energy side are mixed with the first peak on the low energy side and are counted.

That is, to explain the case of a resonance reaction between protons and At as an example, when a proton collides with AI with an energy of 1317.19 KeV or more, the energy of the proton becomes A.
Attenuated while traveling in the I, and just 1317.19KeV
At this point, resonance occurs at W (point A) and γ-rays are emitted. However, if the AI is thick, the protons that have passed through this position continue to decelerate, resonate again at the 991.91 KeV position (point B), and emit γ-rays. Therefore, the γ rays emitted from point A (second peak) and point B (first peak) are counted together.

The present invention separates multiple resonance points in a nuclear reaction,
The purpose of this study is to obtain a method for calibrating the energy of charged particles that guarantees linearity.

[Means to solve the problem]

The solution to the above problem is that when calibrating the energy of charged particles using a nuclear resonance reaction between charged particles with multiple resonance energies and a target material, charged particles with resonance energy on the high energy side are and the amount of energy that is attenuated.

This is achieved by a charged particle energy calibration method characterized by using a target thinner than the thickness at which the difference between the resonance energy on the high energy side and the resonance energy on the adjacent low energy side is equal.

[Function] In the present invention, after a charged particle having resonance energy on the high energy side passes through the target, 1
It holds energy higher than the resonance energy of the one below, and therefore does not cause the next resonance in the target, so the radiation count drops to the background level and the peak of the resonance point on the high energy side can be separated. , so that it can be measured.

That is, in the case of the resonance reaction between protons and Al in the above example.

When a proton collides with AI with an energy of 1317.19 KeV or more, the energy of the proton is attenuated while traveling through AI, and resonance occurs at the position (4 points) where the value is exactly 1317.19 KeV, and γ radiate a line. If the thickness of the proton is large, the protons that have passed through this position will continue to decelerate, resonate again at the 991.91 KeV position (point B), and emit γ-rays. However, if the thickness of AI is made thinner than the distance between the resistive surface and point B, 9
91.91 KeV resonance does not occur, and the resonance point on the high energy side can be separated.

Next, Table 1 shows some examples of nuclear reactions used in the present invention.

Table 1 Charged particles Resonance energy (in eV) 340.45±0.04 453.6±0.3 872, 1±0.2 1373.0±1.0 Proton Proton 991.91±0.05 1317.19 ±0.4 5801 ±2 Proton 1423.64±0.43 1843.45±0.56 9201 ±4 9512 ±3 Radiation Reaction formula% formula%)))))))))) Here, the resonance energy The value of the value indicates the range over which the reaction occurs. In the 9 reaction formula, P is a proton and α is an α ray.

T indicates a gamma ray, and n of radiation indicates a neutron ray.

1) 5elected Low Energy Nuc
lear Reaction'p281゜For example, 1 reaction "AI (P+ T)" S+ is 27A1
In this reaction, the proton P causes a resonance reaction and emits gamma rays, changing to 2''Si, which has a mass number one larger.

Also, one reaction 19F (P, αγ) 160 is 19F and proton P
causes a resonance reaction and emits α rays ('He) and γ rays.

This is a reaction that changes to 160. What is the mass calculation in this case?

Because the equivalent mass of gamma rays is very small.

19(F) +1(P) −4(He) =16(0
) becomes.

〔Example〕

As an example of the present invention, "A" in Table 1 is used as a nuclear reaction.
l(P, 7')"Si is used, and protons (
Accelerate H゛).

The accelerator will be a tandem Van de Graaf accelerator.

Next, the target, which is a feature of the present invention, will be explained.

This example is a nuclear reaction in the Hel membrane. first,
According to LSS theory or numerical table 3), the energy attenuation rate Δ of 1000 KeV protons in a 1 μm thick layer is approximately 24
It can be seen that the value is KeV/μm.

2) Lfndhard, 5charff, 5ch
Theory 3) Pro'ected Ran e 5tatis
tics.

Sem1conductor and Re1ated
Materials.

2nd Edition, Halsted Pres.
s.

On the other hand, since the smallest energy difference among the resonance points of 1 reaction 2781(P, γ)28Si is 326 KeV, it can be seen that it is sufficient to form an AI film with a thickness of approximately 13 μm or less on the Si.

In the example, AI was deposited on a St wafer by vapor phase growth (CVD).
3,000 people were coated using the method. Δ in this case is 7 KeV.

Protons accelerated by an accelerator equivalent to 1317 Keν are incident on a target coated with an Al film, causing resonance.

Emit prompt gamma rays.

This gamma ray is detected with a scintillation counter.

When the incident energy is continuously changed from 1317 to 991 KeV, two peaks with an energy width of about 20 eV appear as shown in FIG.

As mentioned above, the energy Δ lost when a proton perpendicularly penetrates an AI film with a thickness of 3,000 people is 7 KeV, so the peak energy width should be 7 KeV. In reality, it is wider than this value and reaches about 20 K due to things that are incident off-axis from the perpendicular direction, scattering of free electrons in solids (Compton scattering), etc.
eV.

FIG. 1 is a diagram showing measurement results of resonance points of nuclear reactions according to an example.

The figure shows the number of counts of radiation generated by nuclear reactions with respect to the energy of charged particles.

In this way, the absolute value of the tandem accelerating voltage can be found extremely accurately (±0.05 KeV) at two points, so calibration can be done by adjusting the gain and offset of the accelerating power supply using this as a reference. Bye.

That is, as mentioned above, in the diagram showing the linear relationship between the acceleration energy (relative value) given by the accelerator and the energy (absolute value) from the nuclear reaction, find the lowest two resonance points in 7, and draw a straight line connecting these two points. Adjust the gain and offset of the accelerator 8 times so that it has a 45° inclination that passes through the origin.

In the example, protons were used as accelerated particles.

Instead of this, deuterons, α rays (He ions), etc. may be used.

Further, as for the accelerator, in addition to the tandem type, an RF acceleration method, an electrostatic type Kotscroft accelerator, etc. may be used.

〔Effect of the invention〕

As explained above, according to the present invention, a plurality of resonance points in a nuclear reaction can be separated.

A method for calibrating the energy of charged particles that guarantees linearity is obtained.

[Brief explanation of drawings]

FIG. 1 is a diagram showing measurement results of resonance points of nuclear reactions according to an example. FIG. 2 is a diagram showing the results of conventional measurement of resonance points of nuclear reactions. , Measurement of 5 examples of power supply energy (KeV) of load tm.

Claims (1)

[Claims] When calibrating the energy of a charged particle using a nuclear resonance reaction between a charged particle with multiple resonance energies and a target material, the amount of energy attenuated in the target by a charged particle with a resonance energy on the high energy side is calculated. ,
A charged particle energy calibration method characterized in that a target is thinner than the thickness at which the difference between the resonance energy on the high energy side and the resonance energy on the adjacent low energy side is equal.