JP2010129253A - Focused ion beam device, and device for controlling the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for controlling a focused ion beam device, capable of stably maintaining emission from a liquid metal ion source. <P>SOLUTION: A limiting resistor 28 is provided between an extraction electrode 32 for extracting charged particle beams 2 from a LMIS (liquid metal ion source) 31 and an extraction power source 27 for applying extraction voltage Vext between the LMIS 31 and the extraction electrode 32. A resistance value Ro of the limiting resistor 28, output voltage Vext of the extraction power source 27 and an emission current Ie in the charged particle beams 2 extracted from the LMIS 31, are used for calculating a state discriminating voltage value Vox (=Vext-Ie×Ro) as an evaluated value for the state of the LMIS. When the state discriminating voltage value is out of a predetermined state discriminating voltage range, flashing is performed to restore the state of the LMIS. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a focused ion beam apparatus using a liquid metal ion source (LMIS) and a control apparatus therefor.

  In recent years, processing of samples using a high-energy charged particle beam (for example, a focused ion beam) that has been narrowed down has been actively performed. As such a sample processing technique, a technique is known in which a charged particle beam emitted from a charged particle source (liquid metal ion source) is finely squeezed and irradiated onto the sample to finely process the sample. (Refer to patent document 1 etc.). In addition, a technique for stabilizing ion emission (emission) from a charged particle source (liquid metal ion source) is known for stable supply of charged particle beams (see Patent Document 2, etc.).

Japanese Patent No. 335554846 JP 2005-174604 A

  In the above prior art, charged particles are extracted by applying an extraction voltage between the liquid metal ion source and the extraction electrode, and ion emission is performed depending on whether or not the emission current becomes a predetermined value within the range of the predetermined extraction voltage. The quality of the (emission) state is judged.

  However, since the LMIS state is not considered in the determination of the quality of the emission state using only the relationship between the extraction voltage and the emission current, it cannot be said that the grasp of the emission state is sufficient and the emission current is low. If used, the emission may become unstable.

  The present invention has been made in view of the above, and an object of the present invention is to provide a control apparatus for a focused ion beam apparatus that can stably maintain the emission of a liquid metal ion source.

  A liquid metal ion source, an extraction electrode for extracting a charged particle beam from the liquid metal ion source, an extraction voltage source for applying an extraction voltage between the liquid metal ion source and the extraction electrode, the extraction electrode and the extraction voltage Using the limiting resistor provided between the sources, the resistance value Ro of the limiting resistor, the output voltage Vext of the extraction power supply, and the emission current Ie of the charged particle beam extracted from the liquid metal ion source An arithmetic unit that calculates a state determination voltage value Vox = Vext−Ie × Ro as an evaluation value of the state of the metal ion source, and the liquid metal ion when the state determination voltage value is outside a predetermined state determination voltage range And a recovery processing unit that performs a state recovery process for recovering the state of the source.

  In the present invention, the emission of the liquid metal ion source can be stably maintained.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

In the present embodiment, FIG. 1 is a diagram showing the overall configuration of the focused ion beam apparatus of the present embodiment.

  In FIG. 1, a focused ion beam apparatus according to the present embodiment includes a vacuum vessel 1, an ion gun 3 disposed in the vacuum vessel 1 and irradiating a charged particle beam 2, and a downstream side (sample 6 described later). The objective aperture 4 is a variable objective aperture that can limit the charged particle beam (ion beam) 2 irradiated to the side) and change the aperture, and deflection for controlling the deflection of the ion beam 2 that has passed through the objective aperture 4. The ion beam 2 that has passed through the deflector 5, and the objective lens 7 that irradiates the sample 6 placed on the sample stage (not shown) and the ion beam 2 that is emitted from the sample 6. And a signal detector 9 for detecting the signal 8 to be transmitted.

  The ion gun 3 includes a liquid metal ion source (LMIS) 31, an extraction electrode 32 that extracts an ion beam 2 (ion) from the liquid metal ion source (hereinafter referred to as LMIS) 31, and an extraction electrode 32. A beam limiting aperture 33 for limiting the beam spread of the ion beam 2, and a ground electrode 34 for accelerating the ion beam 2 (ion) that has passed through the beam limiting aperture 33.

  Further, the focused ion beam apparatus according to the present embodiment is connected to the vacuum vessel 1 through the gate valve 11 and is evacuated from the vacuum vessel 1 connected to the vacuum vessel 1 and evacuated from the vacuum vessel 1. An ion pump 12, a high-voltage power supply unit 20 that supplies power to the ion gun 3, and a control unit 40 that controls the operation of the entire focused ion beam apparatus.

  The high-voltage power supply unit 20 includes a grounding electrode 21 that is grounded, an extraction power supply 27 that applies an extraction voltage for extracting the ion beam 2 (ion) from the LMIS 31 between the LMIS 31 and the extraction electrode 32, an extraction power supply 27, and an extraction A limiting resistor 28 provided between the electrodes 32, an acceleration power supply 24 for applying an acceleration voltage for accelerating the ion beam 2 (ion) between the ground electrode 21 and the LMIS 31, and a voltage for heating the LMIS 31. The heating power supply 29 to be applied, the high voltage cables 22 and 25 for connecting the components of the high voltage power supply unit 20, and the high voltage connection units 23 and 25 for connecting the high voltage cables 22 and 25 to the LMIS 31 and the extraction electrode 32, 26. Here, the resistance value Ro of the limiting resistor 28 (hereinafter, appropriately described as the limiting resistor Ro) is, for example, 300 MΩ, and has a resistance value that satisfies Ro >> 1 / m (= 7 MΩ) (described later). ).

  FIG. 2 is a diagram showing the configuration of the LMIS 31 in the ion gun 3.

  In FIG. 2, an LMIS 31, for example, a gallium liquid metal ion source (GaLMIS), energizes and heats a needle-like emitter 35 having a conical tip, a reservoir 36 for storing Ga, and the needle-like emitter 35 and Ga in the reservoir 36. A filament 37, an energizing terminal 38 connected to the filament 37, and an insulator base 39 for fixing the energizing terminal 38 are provided. The acicular emitter 35, the reservoir 36, and the filament 37 are made of, for example, tungsten (W). The reservoir 36 is filled with gallium (Ga).

  FIG. 3 is a block diagram showing an outline of the function of the control unit 40.

  In FIG. 3, the control unit 40 controls the operation of the entire focused ion beam. The resistance value Ro of the limiting resistor, the output voltage Vext of the extraction power source, and the emission current of the charged particle beam extracted from the liquid metal ion source. A calculation unit 41 that calculates a state determination voltage value Vox as an evaluation value of the state of the liquid metal ion source using Ie, a storage unit 42 that stores various data such as the state determination voltage value Vox, and a state And a display unit 43 for displaying various data such as the discrimination voltage value Vox and various warnings.

  Next, the operation principle of the focused ion beam apparatus according to the present embodiment will be described with reference to the drawings.

(1) Radiation angle current density The radiation angle current density Jω of the LMIS 31 is an emission (profile) distribution. Hereinafter, the radiation angle current density Jω will be described.

  FIG. 4 is a diagram showing a measurement result when an emission range (emission range) is measured by changing the emission current Ie when a fluorescent plate is arranged downstream from the LMIS 31 and the charged particle beam 2 is irradiated. .

  In FIG. 4, the emission ranges for the emission currents Ie = 2uA, 5uA, 8uA, and 10uA are shown from the left side, and the emission range becomes wider as the emission current Ie increases.

  FIG. 5 is a diagram showing the relationship between the above-described emission current and the half opening angle (of emission). The vertical axis represents the emission half opening angle α (°) viewed from the LMIS 31, and the horizontal axis represents the emission current Ie (uA). Each is shown.

  As can be seen from FIG. 5, the half aperture angle α of the emission is the half aperture angle α = 11 ° to 15 ° (0.19 rad to 0.26 rad) between the emission currents Ie = 2 uA to 3.2 uA.

  FIG. 6 is a diagram showing the relationship between the emission current Ie and the radiation angle current density Jω. In FIG. 6, the horizontal axis indicates the emission current Ie (uA), and the vertical axis indicates the radiation angle current density Jω (3.2 uA) when the emission current Ie = 3.2 uA and the current when each emission current Ie. The ratio of density Jω is shown. The diameter of the objective aperture 4 was set to 300 μm, and the beam current Ib that passed through the objective aperture 4 was measured.

  In FIG. 6, for example, when the emission current changes from Ie = 3.2 uA to Ie = 2.2 uA, that is, when the emission current decreases by 1 uA, the radiation angle current density ratio Jω / Jω (3.2 uA) is about 1 to about It changes to 0.92. Since the beam current Ib is proportional to the radiation angle current density Jω, the beam current Ib at the emission current Ie = 2.2 uA is about 0.92 compared with the beam current Ib at the emission current Ie = 3.2 uA. Double.

  FIG. 7 is a diagram showing the relationship between the emission current Ie and the beam current Ib. The horizontal axis indicates the logarithmic display of the emission current Ie (uA), and the vertical axis indicates the beam current Ib (nA).

  As can be seen from FIG. 7, Ib∝ln (Ie), that is, the beam current Ib is substantially proportional to the logarithm of the emission current Ie.

  Further, since the radiation angle current density Jω is obtained by measuring the beam current Ib that has passed through the objective aperture (aperture), the solid angle formed by the aperture is the same. Therefore, the beam current Ib = (radiation angle current density Jω × solid angle formed by the aperture) ∝radiation angle current density Jω, that is, Ib∝Jω. Therefore, approximately, the on-axis radiation current density Jω, the emission current Ie, and the beam current Ib have a relationship of Jω∝Ib∝ln (Ie).

(2) Energy spread of radiation ions (radiation beam) (ΔE)
FIG. 8 is a diagram showing the energy spread ΔE of the radiation beam with respect to the emission current Ie when using GaLMIS. The horizontal axis represents the emission current Ie (uA), and the vertical axis represents the energy spread ΔE of the radiation beam. Logarithmic display.

  FIG. 9 is a diagram showing a change in Jω / ΔE ^ 2 with respect to the emission current Ie. The horizontal axis indicates the emission current Ie (uA), and the vertical axis indicates the logarithm of Jω / ΔE ^ 2. Assuming that Jω / ΔE ^ 2 in FIG. 9 is φ∝Jω / ΔE ^ 2 using the variable φ, it can be seen that φ is near the maximum value in the vicinity of the emission current Ie = 2uA.

(2-1) Physical Meaning of Variable φ Here, the physical meaning of the variable φ will be examined. The beam diameter d of the ion beam 2 is obtained by a relational expression of d ^ 2 = (d0 * M) ^ 2 + (0.5 * Cs * α ^ 3) ^ 2 + (Cc * ΔE / Va * α) ^ 2. Since the energy spread ΔE when using GaLMIS is ΔE> 5 eV, paying attention to the chromatic aberration term (Cc × ΔE / Va × α), αg = (obtained from the relationship of Ib = Jω × π × αg ^ 2 From the relationship of Ib / π / Jω) ^ 0.5 and the relationship of α = αg / Mα, ΔE × α = ΔE × (Ib / π / Jω) ^ 0.5 / Mα = (Ib / π / φ) ^ 0 · 5 / Mα is obtained. From this relational expression, it can be seen that ΔE × α decreases as the variable φ increases. This indicates that the emission current Ie = 2.2 uA at which the variable φ is near the maximum value is a condition under which the influence of chromatic aberration is the minimum value (minimum value).
(3) Ie_V characteristics (examination of operation of liquid metal ion source)
FIG. 10 is a diagram showing the relationship between the emission current change ΔIe and the output voltage Vext output from the extraction power supply. In FIG. 10, the horizontal axis represents the logarithm of ΔV / Vo ^ (3/2), and the vertical axis represents the logarithm of the emission current Ie (uA).

  FIG. 10 is expressed as a logarithm, and since the inclination is 1 with respect to the horizontal axis, the following equation is derived.

  From Equation 1, since ∂Ie / ∂V = m ′ = constant, the following relationship is obtained.

  Here we consider Child-Langmuir Law. This law represents the saturation current density J of parallel flat plates (for example, distance d, potential difference φ), and is expressed by the following equation.

  Equation 2 above shows that the maximum current density is limited by the space charge effect. Therefore, the relationship of Ie∝ (Vo) ^ (3/2) derived from the above equation 1 indicates that the space charge effect (described later) works in the emission of LMIS 31.

  In FIG. 10, when the emission current Ie≈1.1 uA or less, the emission current Ie becomes pulse radiation and becomes unstable. On the other hand, when the emission current Ie> 1.1 uA, the emission current Ie becomes stable continuous radiation. This is because when Ie> 1.1 uA, the emission is limited due to the space charge effect and stable continuous radiation is obtained. Therefore, it can be seen that Ie≈1.1 uA is a critical point where the space charge effect works. In addition, in the state where the space charge effect does not work, it does not become continuous radiation (becomes pulse radiation) and the emission becomes unstable.

(3-1) Field Evaporation and Space Charge Effect The field evaporation and space charge effect will be described.

  First, an outline of the emission operation principle of LMIS (Liquid Metal Ion Source) will be described. Since the tip of the needle-like emitter has a conical shape (see FIG. 12 later), the electric field gradient in the axial direction becomes stronger toward the tip. The liquid metal (for example, gallium (Ga)) near the tip of the needle-like emitter is supplied to the tip of the emitter having a strong electric field due to the electric field stress, and becomes a conical shape with a point apex. That is, the liquid metal forms a cone at the tip, and its apex becomes a strong electric field of about V / Å, so that the ionization potential is lowered and the electrons are lost and easily ionized. Therefore, in a strong electric field of about V / Å, ions jump out into the vacuum and an ion flow is generated. For example, when GaLMIS is used, it is necessary to supply a liquid metal (Ga) corresponding to the amount of Ga ions released along with this ion emission, but since Ga behaves fluidly, a pressure gradient occurs, Surface flow occurs. Since this flow is a Poiseuille flow caused by the surface tension, the flow rate changes due to the change in the surface tension.

  Further, pure gallium (Ga) has good wettability with respect to clean tungsten, and when tungsten (W) has a narrow groove, it diffuses through the groove due to a capillary phenomenon due to surface tension. It is known that the needle-like emitter 35 is formed of a tungsten (W) material and is provided with a thin vertical groove in the axial direction. This allows Ga in the reservoir to pass through the groove by capillary action of surface tension. It can be diffused and supplied to the vicinity of the tip of the emitter.

  If Ga supply corresponding to the Ga ion release amount is performed, the emission becomes stable, but when the surface tension of the liquid metal (Ga) changes, the supply amount changes and the emission changes.

  FIG. 11 is a diagram schematically showing a physical phenomenon at the LMIS tip, and FIG. 12 is a diagram showing a SIM (Scanning Ion Microscope) image at the LMIS tip.

  As the observation result of the size and shape of the ion emission point at the time of emission at the tip of the conical LMIS (tip of the needle-like emitter 35) as shown in FIG. 12, for example, as shown in FIG. 11, the emission current Ie = 10 uA, a radius of 15 mm, that is, a diameter of 30 mm (= 3 nm) is known. From this observation result, the ion emission surface size Ao = 0.7 × π × ro ^ 2 and current density j = i / Ao, and the current density j is determined as j = 10 ^ 12 (A / m ^ 2). Therefore, a large space charge effect works.

  On the other hand, when the electrolytic strength Fo is obtained by electrostatically defining the curvature of the cone tip as ro, Fo = 2.32 Vo / ((h ^ 0.5) (ro ^ 0.5)) and Vo = 10 kV. Assuming h = 1 mm and ro = 20 mm, Fo = 1.6 × 10 ^ 10 (V / m). However, this calculation is not affected by space charge and does not match the actual phenomenon.

  The electric field required for ion field evaporation is about 15 V / nm. From the balance between electric field stress and surface tension, the electric field Eo at the tip is Eo = 2 (γ / (εo × ro)) ^ 0.5, and when ro = 1.5 nm, Eo = 15 V / nm, which is an actual phenomenon. Matches well. Further, when the electric field Eo at the tip of the LMIS is calculated in consideration of the electric field relaxation due to the space charge effect of the emitted ions, Eo / Fo≈1 / 5 is obtained with the emission threshold voltage Voc = 6 kV and the emission current Ie = 1 uA. Since the electric field necessary for field evaporation is Eo = 15 V / nm, Fo before ion emission is about 5 times Eo, and in the LMIS where the emission threshold voltage Voc = 6 kV, the emission current Ie = 1 uA. Must be relaxed to about 1/5 of Fo and Eo≈15 V / nm.

  Here, if the electric field or the ion emission rate does not change even if the emission current increases, the ion emission surface increases, and F = V / (β × ro) can be written (β is a form factor). Even if the emission current increases, the internal pressure is the same (that is, the electric field is the same), so β × ro = constant. Therefore, it can be said that when the curvature ro increases, β decreases, that is, the LMIS tip extends. In general, it is known that the space charge effect works strongly at the emission point, and the emission point increases as the emission increases.

(4) Relationship between the Ie_V characteristic and the resistance value Ro of the limiting resistor The change in the Ie_V characteristic and the emission change are extrapolated to the emission current Ie = 0 in the relationship between the emission current Ie and the output voltage Vext of the extraction power supply 27 when experimentally obtained. The following relationship is established with the voltage at the time being the state determination voltage value Vox.

  The variation of the above equation 4 is obtained by the following equation.

  Therefore, the emission changes when the slope of the Ie_V characteristic changes from Equation 7 above.

(4-1) Limiting resistance (feedback resistance)
Here, the limiting resistor 28 (feedback resistor) used by inserting between the extraction electrode 32 and the extraction power supply 27 for the purpose of stabilizing the emission will be considered.

  When the resistance value of the limiting resistor 28 is Ro and the emission current is Ie, the extraction voltage Ve actually applied to the LMIS 31 (hereinafter referred to as extraction voltage Ve) and the output voltage (display value) Vext (extraction voltage 27). Hereinafter, the relationship of the output voltage Vext) is expressed by the following equation.

  In the above equation 8, the rate of voltage change when the emission current Ie changes is obtained by the following equation.

  For example, in the Ie_V characteristic in GaLMIS, when the extraction voltage Ve changes (increases) by 10 V, the emission current Ie changes (increases) by 1.5 uA. Therefore, the following equation holds.

  In the above equation 10, when the limiting resistance Ro = 300 MΩ (>> 7 MΩ), the following equation is established.

  As can be seen from Equation 11 above, the Ie_V characteristic of the measurement system using the limiting resistance of the limiting resistance Ro substantially matches the output voltage Vext with a slope 1 / Ro.

  Further, when the slope when the Ie_V characteristic is expressed by the extraction voltage Ve is m and the slope when the output voltage Vext is expressed by M, the following equation is established.

  In Equation 12, 1 / m is called flow impedance. In Expression 12, m >> M, so that M≈1 / Ro. That is, the slope M in the Ie_V characteristic of the measurement system including the limiting resistor 28 having the resistance value Ro is M≈1 / Ro.

  Here, in the above equation 8, when the variation is obtained with the output voltage Vext fixed, it is expressed as follows.

  In the above equation 14, it can be seen that in order to suppress (suppress) the fluctuation ΔIe / Ie of the emission current to be small, it is sufficient to increase the slope m or increase the limiting resistance Ro. Although the slope m varies depending on individual differences in LMIS, LMIS produced so that the value of slope m is generally constant is used.

  The resistance value Ro of the limiting resistor 28 in the present embodiment is, for example, 300 MΩ, and has a resistance value that satisfies Ro >> 1 / m (= 7 MΩ).

(4-2) Examination of Ie_V Characteristic The measurement result of the Ie_V characteristic by the measurement system using the limiting resistance (resistance value Ro) will be described with reference to FIG.

  FIG. 13 is a diagram showing the relationship between the output voltage Vext and the emission current Ie when the resistance value Ro is changed. In FIG. 13, the horizontal axis indicates the output voltage Vext (kV), and the vertical axis indicates the emission current Ie (uA). FIG. 13 shows the Ie_V characteristics when the limiting resistance Ro = 0 MΩ, 10 MΩ, and 300 MΩ.

  In FIG. 13, the Ie_V characteristic when the limiting resistance Ro = 0 MΩ is expressed by an approximate expression y = 92.654x−683.18, where Vext = x (uA) and Ie = y (kV). Similarly, an approximate expression when the limiting resistance Ro = 10 MΩ is expressed by y = 62x−461.1, and a recent expression when the limiting resistance Ro = 300 MΩ is expressed by y = 3.243x−23.91. . In addition, when the limiting resistance Ro = 10 MΩ and 300 MΩ, the slopes of the approximate equations are about 1/300 (1 / MΩ) and about 1/17 (1 / MΩ) when calculated in consideration of the unit of each axis. It is.

  In each limiting resistor Ro, when the output voltage Vext is gradually increased from 0 kV (state without emission), the extraction voltage when the emission is first output is set as the threshold voltage Voc. In the Ie_V characteristic when the limiting resistance Ro = 300 MΩ, the threshold voltage Voc = 8.02 kV. When the limiting resistance Ro = 10 MΩ, the threshold voltage Voc≈7.65 kV.

  In addition, when the output voltage Vext is gradually lowered from the state in which emissions are emitted, the extraction voltage when the emission changes from continuous radiation to pulse radiation is defined as the lower limit threshold voltage Voo. In the Ie_V characteristic when the limiting resistance Ro = 300 MΩ, the lower limit threshold voltage V1≈7.75 kV. When the limiting resistance Ro = 10 MΩ, the lower limit threshold voltage V1≈7.55 kV. For example, when the lower limit threshold voltage V1 is reached during imaging with a SIM (Scanning Ion Microscope), black stripes begin to appear on the imaging screen.

  In the Ie_V characteristic when the limiting resistance Ro = 0, when the output voltage Vext (≈extraction voltage Ve) changes by 100 V, the emission current changes by about 15 uA.

  As can be seen from FIG. 13, in the Ie_V characteristic (approximate expression thereof), the extraction voltage (state determination voltage value) Vox extrapolated to the point where the emission current Ie = 0 is equal regardless of the size of the limiting resistor Ro. .

(5) Examination of change over time of Ie_V characteristic The change over time of the Ie_V characteristic will be described with reference to FIGS.

  As described above, when the resistance value of the limiting resistor 28 is Ro and the emission current is Ie, the extraction voltage (application voltage) Ve actually applied to the LMIS 31 and the output voltage (display value) Vext of the extraction power supply 27 are The relationship is expressed by Equation 8 above.

  FIG. 14 is a diagram illustrating the relationship between x and y when Ve−Vox is x and the emission current Ie is y. FIG. 15 is a graph illustrating X and Y when the output voltage Vext is X and the emission current Ie is Y. It is a figure which shows the relationship. 14 and 15 show a case where the slope m of the Ie_V characteristic changes to (m + dm) during the time h.

  14 and 15, the Ie_V characteristics are represented by the following equations, respectively.

(5-1) Examination of relationship between x and y (Ve−Vox = x, Ie = y) When the above formula 15 is fully differentiated, dy = dm × x + m × dx.
When x is constant, the above total differentiation is expressed as dy = dm × x, which corresponds to Δy1 and Δy2 in FIG.
If m is constant, the total differential is expressed as dy = m × dx, which corresponds to (Δx1, Δy1) in FIG.

  When the slope of the Ie_V characteristic changes from m to (m + dm) during time h, y0 = m × x0 changes to y (h) = (m + dm) × x0 at x = x0. Therefore, Δy1 = y (h) −y0 = dm × x0. From this, the following equation is obtained.

  However, in Equation 17, Ve = const.

  From the above equation 17, it can be seen that the emission current Ie changes (ΔIe) when the extraction voltage Ve is constant is due to the change in the slope m (Δm) in the Ie_V characteristic. ΔIe decreases when dm <0 (inclination decreases when Ga supply deteriorates).

  Here, consider a case where the emission current Ie is changed (returned) from y (h) = (m + dm) × x0 to y0 = (m + dm) × x1.

  The change of y is expressed as y0−y (h) = (m + dm) × (x1−x0). As for x, when x0 = V0−Vox and x1 = V1−Vox, x1−x0 = (V1−Vox) − (V0−Vox) = V1−V0. Therefore, the following formula is derived.

  That is, according to the above equation 18, changing (returning) the emission current Ie from y (h) to y0 increases the emission current by ΔIe (the extraction voltage becomes ΔVe) on the Ie_V characteristic of the slope m + dm in FIG. Is equivalent to increasing).

(5-2) Examination of influence of change (dm) in slope m on relationship between X and Y (Vext = X, Ie = Y) X = Vext, Ie = y = Y, Ve−Vox = x is expressed by the above equation 8 Substituting into, the following equation is obtained.

  When the slope of the Ie_V characteristic changes from m to (m + dm) during time h, when X is constant, X before the slope change is expressed as X = (x0 + Vox) + Ro × Y0, and X after the change is X = (x1 + Vox) + Ro × Y (h). Taking these differences, 0 = (x0−x1) + Ro × (Y0−Y (h)) = − ΔVe + Ro × ΔIe. However, Y (h) −Y0 = −ΔIe ′.

  However, in the above equation 20, ΔVext≈0.

  Therefore, when there is the limiting resistance Ro, the slope of the Ie_V characteristic when ΔVext≈0 is 1 / Ro.

  Further, when x is constant, the following equation is obtained from the above equation 17.

  From Equation 18 and Equation 21, ΔVe = ΔIe / (m + dm) = Ie × dm / m / (m + dm) = Ro × ΔIe, m> dm is obtained, and Ve = Ro × ΔIe ′ is obtained from Equation 20 above. It is done. The following formula is obtained from these relationships.

  Further, the following formula is also obtained.

  The above expression 25 is a relational expression representing the emission current change ΔIe ′ in the Ie_V characteristics of the emission current Ie and the output voltage Vext.

  From the above equation 25, it can be seen that increasing Ro approaches dM≈0. That is, by increasing Ro, dM decreases and ΔIe ′ decreases.

  As described above, when the limiting resistance Ro is large, the slope 1 / Ro of the Ie_V characteristic is substantially constant (that is, dM≈0). Therefore, when x is constant (ΔIe≈0), the relationship shown in the above equation 16 is satisfied. The relationship of the following formula including is obtained.

  FIG. 15 shows the relationship of Equation 27 above. The emission current Ie changes (ΔIe) due to the change in the slope m (Δm) in the Ie_V characteristic, but when the limiting resistance Ro is large, dm <0 (the slope becomes smaller when the supply of Ga becomes worse). An expression is obtained.

  From the above equation 27, the following can be understood. That is, when the limiting resistance Ro is small, M approaches m (including Ro = 0), and the flow impedance (1 / m: represents the quality of the Ga supply state) changes with time, and (1 / m) increases. If (bad), the emission current Ie decreases by ΔIe.

  On the other hand, when the limiting resistance Ro is large, the change M of the slope M of the Ie_V characteristic is ΔM≈0. Therefore, the change in the emission current Ie depends on the change in the extraction voltage (state determination voltage value) Vox and the change in the output voltage Vext. When the amount of change in Vox is large, and the flow impedance (1 / m: indicating whether the Ga supply state is good) changes with time and (1 / m) becomes large (bad), the emission current Ie decreases.

  Changes in the slope of the Ie_V characteristic as described above are collectively shown in FIG.

  FIG. 16 is a diagram illustrating a difference in the slope of the Ie_V characteristic due to a difference in the size of the limiting resistor Ro. The vertical axis represents the emission current Ie, and the horizontal axis represents the extraction voltage Ve and the output voltage Vext.

  As shown in FIG. 16, the change in the slope of the Ie_V characteristic varies depending on the magnitude of the limiting resistor Ro. When the flow impedance (1 / m) changes when Ro is small, the slope of the Ie_V characteristic changes (changes from the position of the solid line to the position of the dotted line), and when Ro is large, the value of Vox changes.

(6) Examination of influence by limiting resistance Hereinafter, the relationship with each part of limiting resistance (resistance value Ro) is examined.

(6-1) Relationship between Extraction Power Supply and Limiting Resistance The slope m of the Ie_V characteristic when GaLMIS is used is m≈0.15 (uA / V) when GaLMIS is clean (initial value).

  When there is no limiting resistance (limiting resistance Ro = 0), it is necessary to control the extraction voltage Ve in units of 0.66 V in order to control the emission current Ie in units of 0.1 uA. Since the extraction voltage of GaLMIS is about 7 kV, if the voltage setting range of the extraction voltage Ve is 0 V to 10 kV, the extraction power supply must have a voltage resolution of 0.66 / 10000 = 6.6 × 10 ^ −5. Is done.

  On the other hand, when there is a limiting resistance (the limiting resistance Ro is large), for example, when the limiting resistance Ro = 300 MΩ, the emission current Ie can be controlled according to the relationship of ΔVext = Ro × ΔIe. Can be controlled in units of 0.1 uA by controlling the output voltage Vext in units of 30 V, and the voltage resolution required for the extraction power supply can be kept low. At this time, in order to increase the emission current Ie by 1 uA, the output voltage Vext may be increased by 300V. Further, when the maximum value of the emission current Ie is 10 uA, the required increase in the output voltage is Vext = 3 kV. Therefore, if an extraction power supply with a maximum output voltage of 12 kV is used, the GaLMIS is driven with the maximum emission current of 10 uA. can do.

(6-2) Relationship between Emission Stability and Limiting Resistance The slope m of the Ie_V characteristic of GaLMIS is m≈0.15 (uA / V) in a clean state (initial value). However, adhesion of sputtered particles emitted from the beam limiting aperture 33 by the ion beam to the LMIS 31, oxidation of Ga by adsorption and decomposition of water molecules (H 2 O) and oxygen molecules (O 2) in the vacuum vessel 1, or beam limitation The surface impedance of Ga and the melting point of Ga change due to contamination of carbon (C) by secondary electrons emitted from the diaphragm 33, and the flow impedance (1 / m) changes with time. In addition, when emission is started after the voltage applied to the LMIS 31 once becomes 0 V, such as when the apparatus is stopped, the Ga retreats from the tip of the LMIS 31, and the temperature of the Ga decreases and the surface tension increases, so that the Ie_V characteristics. Changes.

  In the Ie_V characteristic, Vox changes with a change in the radius of curvature of the tip of the LMIS 31, and a slope changes with a change in flow impedance (Ga supply state). If there is no change in the flow impedance, even if the emission current Ie increases or decreases, it is a change on the straight line of the slope m indicated by the Ie_V characteristic. Therefore, when the increase in the emission current Ie is ΔIe = 1.5 uA, the extraction voltage Ve ≒ 10V increase. In addition, if there is no change in the Ie_V characteristic, if the increase in the emission current Ie is ΔIe = 1.5 uA regardless of the presence (magnitude) of the limiting resistor Ro, ΔVe≈10V.

  The emission current Ie changes with the change Δ (1 / m) in flow impedance, and as a result, the extraction voltage Ve changes (see Equation 22). When Ro × dm> 1, ΔVext> ΔVe, and when Ro × dm≈1, ΔVext≈ΔVe.

(6-3) Relationship between ΔIe and ΔVe and Limiting Resistance The change ΔVe in the extraction voltage Ve is determined according to the change in the Ie_V characteristic regardless of the limiting resistance Ro. Further, the change ΔVe in the extraction voltage is related to the change in focus. That is, the change in the Ie_V characteristic is related to the change in focus.

  The lens action of the ion gun 3 is represented by ρ ^ 2 = (Va / Ve). Since the change Δρ ^ 2 is Δρ ^ 2 = Δ (Va / Ve) = − (ρ ^ 2) × (ΔVe / Ve), the position of the virtual light source is changed by the change ΔVe of the extraction voltage Ve. The objective lens voltage (focus value) required for focusing on the sample 6 with the objective lens 7 changes.

  The extraction voltage Ve changes when the flow impedance (1 / m) changes and the emission current Ie changes (see Equation 22). When the limiting resistance Ro is large, even if the flow impedance changes by Δ (1 / m), ΔIe ′ = ΔVe / Ro (see Equation 20), so that the change in the emission current Ie can be suppressed to a small value. However, ΔVe changes depending on the inclination change Δm regardless of the limiting resistance Ro (see Equation 22).

  Immediately after flushing the LMIS 31, the flow impedance (1 / m) is small (that is, the Ga supply state is good) (solid line in FIG. 16), and the flow impedance (1 / m) over time. And Vox becomes large (dotted line in FIG. 16).

  When the Ie_V characteristic changes and the supply of Ga enough to meet the emission becomes impossible (the flow impedance becomes high), the slope m becomes small (Δm <0). (See equation 20). At this time, in order to return the emission current Ie to the original value (increase it by ΔIe), it is necessary to increase the extraction voltage Ve by ΔVe according to the increased slope 1 / (m + dm) of the flow impedance. When the limiting resistance Ro is Ro >> 1 / m, the change ΔIe of the emission current Ie changes according to the limiting resistance Ro, not the flow impedance (1 / m), and therefore ΔIe = ΔVe / Ro (see Equation 20) It changes with.

  Therefore, when the limiting resistance Ro is increased, the change (ΔIe) in the emission current Ie can be reduced. However, as the limiting resistance Ro increases, ΔIe decreases. On the other hand, in order to reverse the decrease in the emission current Ie (increase ΔIe), the necessary change ΔVext = Ro × ΔIe ′ of the output voltage Vext increases.

  As described above, the emission stability is determined by the magnitude of the limiting resistor Ro, and the focus stability is determined by a change in the flow impedance (1 / m). That is, even when the stability of the emission state is improved by the limiting resistance Ro, the condition of the optical system changes due to the change of the extraction voltage Ve (ΔVe) due to the change of the flow impedance (1 / m). The stability of is reduced.

  In the case of GaLMIS, if the flow impedance (Ie_V characteristic) does not change, the emission and focus are stable regardless of the size of the limiting resistor Ro.

  The slope of the Ie_V characteristic changes depending on the magnitude of the limiting resistor Ro (see FIG. 16). The slope is changed when the limiting resistor Ro is small, and the value of Vox is changed when the limiting resistor Ro is large.

  When the limiting resistance Ro = 300 MΩ, if the emission current Ie changes by ΔIe≈1 uA due to the change in flow impedance, the output voltage Vext (to return to the original emission) becomes ΔVext≈300V. At this time, if the slope 1 / Ro of the Ie_V characteristic does not change, ΔVox (Vext when extrapolated to the emission current Ie = 0) ≈300V.

  Further, when the emission current Ie at the emission threshold voltage Voc is 2 uA, the emission current Ie is reduced to Ie <1 uA when the change in Vox is ΔVox> 300V. When the emission current Ie becomes Ie <1 uA, the continuous radiation is changed to pulse radiation due to the characteristics of the LMIS 31, and the emission becomes unstable. An example when the emission becomes unstable is shown in FIG.

  FIG. 17 is a diagram illustrating an example of a result of measuring the emission. In FIG. 17, the horizontal axis represents the measurement time, the right vertical axis represents the acceleration voltage Vacc (kV), and the left vertical axis represents the output voltage Vext (kV) and the emission current Ie (uA). From FIG. 17, it can be seen that the Ie_V characteristic changes with time, the emission current Ie gradually decreases, and the emission becomes unstable and drops when Ie≈1 uA.

  As described above, it is necessary to increase the limiting resistance Ro in order to suppress the emission fluctuation, but in order to suppress the focus fluctuation, it is necessary to stabilize the flow impedance (1 / m) instead of the limiting resistance Ro. is there.

  The change Δm in the slope of the Ie_V characteristic varies with time depending on conditions such as the structure around the ion gun and the degree of vacuum. For example, an ion gun using a Ga / W beam limiting aperture (aperture formed of gallium and tungsten) and an ion gun using a tungsten (W) aperture have completely different temporal changes in Ie_V characteristics. An ion gun using a Ga / W beam limiting aperture has little change over time in the Ie_V characteristic. For this reason, if the radiation is continuous for about 10 hours, the emission current is constant, and therefore it is not necessary to perform flushing periodically.

  When the apparatus is stopped and then started, the LMIS Ga retreats from the tip and the ion emission point disappears. In addition, since the Ga temperature is different during emission operation and when it is stopped, the shape of the LMIS tip that balances ion emission and supply is changed due to a difference in surface tension, etc., so that initialization by flushing is required. When the flushing is performed, the LMIS state is initialized. If the initialized LMIS state returns to the state before stopping the emission, the emission state also returns.

  In an ion gun using a tungsten (W) beam limiting aperture, the Ie_V characteristic gradually changes and the emission current Ie changes logarithmically with time. For this reason, a control electrode is provided to change the extraction voltage Ve so that the emission becomes constant. In this case, flushing is required near the maximum value of voltage control, and the voltage of the control electrode is restored to the original value by flushing.

  For stable emission of LMIS, it is desirable to operate in a state close to the extraction voltage (emission threshold voltage Voc) when the extraction voltage is increased from 0 kV and emission is first generated (Ie≈2 uA). If emission is performed in this state, the Ga of the liquid metal source (LMIS) is solidified, the surface of the LMIS is oxidized and the flow impedance is high, or the Ga is retreated from the tip. Even in this case, if at least the flow impedance (1 / m) can be recovered by flushing, the emission state can be recovered. Even if flushing, there will be no emission.

  When operating near Voc, φ can be used near the maximum value. Since the emission current value Ie≈2 uA, it is necessary to increase the limiting resistance Ro in order to suppress the change over time of the flow impedance and reduce ΔIe. For this reason, the beam limiting aperture immediately below the liquid metal ion source (LMIS) is a Ga / W aperture, and the limiting resistance Ro is set to 300 MΩ, thereby improving the reproducibility of the LMIS state recovery by the flushing function.

(7) Measurement of emission state The measurement result of emission state will be described below.

(7-1) Activation of Emission The measurement result of emission activation will be described with reference to FIGS.

  18 and 19 are diagrams showing an emission state at the time of emission activation (during activation operation). In FIG. 18, the horizontal axis represents the measurement time, the right vertical axis represents the acceleration voltage Vacc (kV), and the left vertical axis represents the output voltage Vext (kV) and the emission current Ie (uA). In FIG. 19, the horizontal axis represents the output voltage Vext (kV), and the vertical axis represents the emission current Ie (uA). The limiting resistor Ro is a large value (for example, Ro = 300 MΩ). Vox was obtained by measuring the emission current Ie and the output voltage Vext in real time and substituting them into the above equation 16 (equation 26).

  In FIG. 18, flushing was performed during the time indicated by time 1 to time 5. Also, time zone A is immediately after emission activation (before the first flushing), and time zone B is after the end of the fifth flushing. Between time zone A and time zone B, there is a time of about 4 hours.

  In FIG. 19, the measurement result of the Ie_V characteristic in the time zone A in FIG. 18 and the measurement result of the Ie_V characteristic in the time zone B are shown.

  Vox obtained by extrapolating to Ie = 0 from the Ie_V characteristic shown in FIG. 19 is Vox (A) = 7.52 kV in time zone A, and Vox (B) = 6.99 kV in time zone B. It was. FIG. 18 shows Vox obtained by substituting the emission current Ie and the output voltage Vext in the above equation 27 in real time, and Vox (A) = 7.5 kV and Vox (B) = 7.0 kV.

  FIG. 19 shows the Ie_V characteristics at the time of emission activation (time zone A) and at the end of flushing (time zone B). At the time of emission activation, points on the Ie_V characteristic (time zone A) or Ie_V characteristics are shown. The characteristic deviated from (time zone A) and the Ie_V characteristic (time zone B). However, when flushing is performed, a point on the Ie_V characteristic (time zone B) or a characteristic in the vicinity thereof is shown. When these measurement results (Ie, Vext) are substituted into the above equation 27 to obtain Vox and shown in FIG. 19, Vox = 7.5 kV, 7.2 kV, 7-7. It is concentrated around 1 kV. Thus, it can be seen that the value of Vox converges to Vox = 7.5 kV and 7.2 kV when emission is started, and to Vox = 7 to 7.1 kV when flushing is completed. Further, the same Vox means that the Ie_V characteristics are the same. When the Vox is at a point (Ie, Vext) on the straight line of the Ie_V characteristics, the beam is hardly affected. Therefore, when Vox is adjusted to a desired value by flushing, the beam adjustment is not affected even if the emission is adjusted to the desired value thereafter.

  Note that the change in the focus of the apparatus is due to a change in the flow impedance (1 / m) of the LMIS 31. Therefore, by using the Ga / W aperture as the beam limiting aperture 33 just below the LMIS, the change over time in the surface tension of the liquid metal of the liquid metal ion source (LMIS) 31 is eliminated. In the Ga / W aperture, for example, the beam limiting aperture is formed of tungsten (W), and a gallium (Ga) reservoir is disposed in at least a part of a region where the ion beam is substantially irradiated in the beam limiting aperture. is there.

(7-2) Emission Drive The emission change will be described with reference to FIGS.

  20 and 21 are diagrams showing the relationship between the measurement time and the emission state. The horizontal axis shows the measurement time, the right vertical axis shows the acceleration voltage Vacc (kV), and the left vertical axis shows the output voltage Vext ( kV) and emission current Ie (uA). FIG. 20 shows a change in emission for 900 minutes (15 hours), and FIG. 21 is an enlarged view showing the measurement results after the measurement time 841 (minutes) in FIG. 20 and 21, Vox is obtained by substituting the emission current Ie and the output voltage Vext into the above equation 16 (equation 26).

  In FIGS. 20 and 21, when the time-dependent change from the measurement start to the measurement time (1) (around 600 minutes) is seen, the emission current Ie gradually decreases. During this time, Vox gradually increases.

  In the measurement time (1), the output voltage Vext was increased by 200V. As a result, the emission current Ie increased by about 0.6 uA, but there was no significant change in Vox. Thereafter, the emission current Ie gradually decreases and Vox gradually increases. This indicates that the change dm in the slope of the Ie_V characteristic changes at a substantially constant rate, and the state does not change before and after the change in the output voltage Vext.

  Flushing was performed at measurement time (2) (around 850 minutes). Immediately before the flushing, the emission current Ie = 2 uA, Vext = 8 kV, and Vox = 7.4 kV. After the flushing, Ie = 3.3 uA, Vext = 8 kV, and Vox = 7 kV.

  Further flushing was performed at the measurement time (3). The numerical values after flushing are the emission current Ie = 3.2 uA, Vext = 8 kV, and Vox = 7 kV.

  In the measurement time (4), the output voltage Vext was decreased to 7.7 kV, and the emission current Ie was 2.2 uA. At this time, Vox = 7 kV and did not change before and after the change of the output voltage.

  At the measurement time (5), the emission was once stopped, and the emission was restarted at the extraction voltage Vext = 7.7 kV. At this time, the emission current Ie = 1.9 uA and Vox = 7.1 kV.

  Flushing was performed at the measurement time (6). The numerical values after the flushing are Ie = 2.2 uA, Vext = 7.7 kV, and Vox = 7 kV.

  Further flushing was performed at the measurement time (7). The numerical values after flushing were Ie = 2.2 uA, Vext = 7.7 kV, and Vox = 7 kV, and did not change before and after flushing.

  Emission control in the focused ion beam apparatus of the present embodiment configured as described above will be described with reference to the flowcharts of FIGS.

  FIG. 22 is a diagram illustrating a flowchart of the activation process of the focused ion beam apparatus.

  First, the control unit 40 reads the output voltage Vext (e) from the storage unit 42 and outputs it to the extraction power supply 27 (step S10). The output voltage Vext (e) is the output voltage Vext when the emission of the focused ion beam device was stopped last time. However, when the focused ion beam apparatus is activated for the first time, an appropriate recommended value (standard value) is stored in advance in the storage unit 42 as Vext (e).

  Next, the control unit 40 detects the output power Vext and the emission current Ie and stores them in the storage unit 42 (step S15), and calculates and stores a state determination voltage value Vox (s) at startup from the Vext and Ie. Store in the unit 42 (step S20). The state determination reference value Vox is an evaluation value of the state of the liquid metal ion source, and is a voltage (Vox = Vext−Ie) when extrapolated to the emission current Ie = 0 in the relationship between the emission current Ie and the output voltage Vext of the extraction power supply. X Ro).

  Next, it is determined whether or not the absolute value of the difference between the state determination voltage values Vox (e) and Vox (s) stored in the storage unit 42 is larger than a predetermined reference value (for example, 20V) (step S25). If NO, the determination variable (flag) T = 1 is set (step S26), and then the activation end process is performed (step S100). Here, Vox (e) is the state determination voltage value Vox when the previous emission is stopped (immediately before). However, at the time when the focused ion beam apparatus is started for the first time, an appropriate recommended value (standard value) is stored in advance in the storage unit 42 as Vox (e) in the same manner as Vext (e). If the determination result in step S25 is YES, flushing is performed as a state recovery process for recovering the state of the LMIS 31 (step S30), and the process waits for a predetermined standby time (for example, 1 minute) (step S35). Thereafter, the output voltage Vext and the emission current Ie are detected and stored in the storage unit 42 (step S40), and the state determination voltage value Vox (f1) is calculated from the Vext and Ie and stored in the storage unit 42 (step S40). Step S45). In the determination in step S25, the state determination voltage values Vox (e) and Vox (s) are compared to see the difference in flow impedance when emission is stopped and when emission is started.

  Next, it is determined whether or not the absolute value of the difference between the state determination voltage values Vox (f1) and Vox (s) is greater than a predetermined reference value (for example, 50V) (step S50). (Flag) T = 2 (step S51), and then an activation end process is performed (step S100). If the determination result in step S50 is YES, flushing is performed as a state recovery process for recovering the state of the LMIS 31 (step S55), and the system waits for a predetermined standby time (for example, 1 minute) (step S60). Thereafter, the output voltage Vext and the emission current Ie are detected and stored in the storage unit 42 (step S65), and the state determination voltage value Vox (f2) is calculated from the Vext and Ie and stored in the storage unit 42 (step S65). S70).

  Next, it is determined whether or not the absolute value of the difference between the state determination voltage values Vox (f1) and Vox (f2) is larger than a predetermined reference value (for example, 50V) (step S75). (Flag) T = 3 (step S76), and then an activation end process is performed (step S100). If the decision result in the step S70 is YES, a warning C is displayed on the display unit 43 as a state recovery process (step S80), a decision constant (flag) T = 0 is set (step S85), and then the movement termination process is performed. Perform (step S100). Here, the warning C notifies the operator that the state determination voltage value Vox does not converge.

  FIG. 23 is a diagram showing a flowchart of start-up end processing of the focused ion beam apparatus.

  The control unit 40 newly stores the Vox (e), Vox (s), Vox (f1), and Vox (f2) stored in the storage unit 42 in the storage unit 42 as a log of activation processing (at the time of emission activation) ( In step S110, it is determined whether or not flag T ≠ 0 (step S120). If the determination result is NO, normal processing is performed (step S200). In step S120, if the determination result is YES, it is determined whether or not flag T ≠ 3 (step S130). If the determination result is NO, the determination reference value Vox0 is updated to Vox (f2) (step S130). S131), the Vox allowable range [a, b] is updated (step S132). Here, the Vox allowable range [a, b] is a voltage range with Vox0 as the midpoint, and for example, a = Vox0−50V and b = Vox + 50V are set. As an initial value of Vox0, an appropriate recommended value (standard value) is set in advance as Vox0 and stored in the storage unit 42.

  Next, it is determined whether or not the absolute value of the difference between the determination reference values Vox0 and Vox (e) is equal to or less than a predetermined reference value (for example, 150 V) (step S140). If the determination result in step S130 is YES, the determination in step S140 is performed in the same manner. If the determination result is YES in step S140, the output voltage Vext is adjusted (step S150), and normal processing is performed (step S200). The adjustment of Vext in step S150 is to adjust the value of Vext so that the emission current Ie becomes a desired value (for example, Ie = 2uA), and may be performed manually by the operator, or the desired emission current Ie in advance. May be set and automatic adjustment may be performed.

  If the determination result in step S140 is NO, a warning D is displayed on the display unit 43 as a state recovery process (step S141), and then the process of step S150 is performed. Here, the warning D notifies the operator that focus adjustment is necessary.

  FIG. 24 is a diagram showing a flowchart of normal processing of the focused ion beam apparatus.

  First, the control unit 40 determines whether or not there is an instruction for the emission end process by the operator (step S210). If the determination result is NO, the emission end process is performed (step S300). If the determination result in step S210 is YES, the process waits for a predetermined standby time (for example, 1 minute) (step S220), and then detects the output voltage Vext and the emission current Ie and stores them in the storage unit 42 (step S220). In step S230, the state determination voltage value Vox is calculated from the Vext and Ie, and stored in the storage unit 42 together with the calculation time (step S240).

  Next, it is determined whether or not the state determination voltage value Vox is within the Vox allowable range [a, b] (step S250). If YES, normal processing is performed again (step S200). If NO, Flushing is performed as a state recovery process of the LMIS 31 (step S251), and then a normal process is performed (step S200).

  FIG. 25 is a diagram showing a flowchart of the emission end process of the focused ion beam apparatus.

The control unit 40 detects the output voltage Vext and the emission current Ie,
Vext (e) and Ie (e) are stored in the storage unit 42 (step S310), and then the output voltage Vext is stopped (step S320).

  The operation of the present embodiment configured as described above will be described.

(Emission (automatic) startup)
When the emission is stopped and the operator instructs the control device 40 to start emission using an input device (not shown), the control device 40 outputs the output voltage Vext ( e) is read from the storage unit 42, and the output voltage Vext of the extraction power supply 27 is set to Vext (e) for output (step S10 in FIG. 22).

  Then, by comparing Vox (e) and Vox (s), it is determined whether the difference in flow impedance between emission stop and emission activation is within a predetermined range. The process proceeds to end processing (steps S25 and S100 in FIG. 22).

  If the difference in flow impedance is outside the predetermined range, flushing is performed, and then the state determination voltage value Vox (f1) is calculated and compared with Vox (s), so that the flow impedance of the LMIS 31 is flushed. It is determined whether or not it has been recovered (steps S30 to S50 in FIG. 22).

  When the flow impedance of the LMIS 31 is recovered by the flushing (first time), the process proceeds to the start end process (Steps S50 and S100 in FIG. 22). When the flow impedance is not recovered by the first flushing, the flushing (second time) is performed again. After that, the state determination voltage value Vox (f2) is calculated and compared with Vox (f1), thereby determining whether or not the flow impedance of the LMIS 31 has been recovered by the flushing (second time) (step in FIG. 22). S55-S75).

  When the flow impedance of the LMIS 31 is recovered by the flushing (second time), the process proceeds to the start-end process (steps S75 and S100 in FIG. 22). A warning that the value Vox is not converged (warning C) is displayed to notify the operator, and the process proceeds to the activation end process (steps S75 to S100 in FIG. 22).

  In the activation end process, the control unit 40 stores each numerical value (Vox (e), Vox (s), Vox (f1), Vox (f2)) as a log in the storage unit 43 (step S110 in FIG. 23).

  If the flow impedance of the LMIS 31 is not recovered by two flushings, the process proceeds to normal processing (steps S120 and S200). In the case of the exception, it is determined whether or not the focus adjustment is necessary by comparing the determination reference values Vox0 and Vox (e) (step S140 in FIG. 23). If the focus adjustment is necessary, the display unit 43 A warning (warning D) is displayed to notify the operator (step S141 in FIG. 23). Regardless of whether focus adjustment is necessary, the value of the output voltage Vext is adjusted so that the emission current Ie becomes a desired value (for example, Ie = 2 uA) (step S150 in FIG. 23), and then normal processing is performed. Transition.

(Normal operation)
During normal operation (during normal processing), the control unit 40 waits for every fixed time (for example, every minute) when there is no instruction from the operator for the end of emission, and then detects Vext and Ie to store 43, Vox is calculated and stored in the storage unit 43, and it is determined whether or not the Vox is within the Vox allowable range (steps S210 to S250 in FIG. 24). If Vox is within the Vox allowable range, nothing is performed. If Vox is outside the Vox allowable range, flushing is performed to restore the flow impedance of the LMIS 31 (steps S250 to S251 in FIG. 24). When there is no instruction for the emission end process from the operator, the processes of steps S210 to S251 are repeated, and when there is an instruction for the emission end process, the process proceeds to the emission end process.

(At the end of emission)
When the emission end process is instructed by the operator, the control unit 40 stores the output voltage Vext (e) and the state determination voltage value Vox (e) at that time in the storage unit 43, and stops the output voltage Vext.

  The effect of the present embodiment configured as described above will be described.

  1. In the conventional focused ion beam apparatus, the quality of the emission state is determined based on whether or not the emission current Ie becomes a desired value within the range of the predetermined extraction voltage Vext. The state of the LMIS 31 (Ie_V characteristic) Therefore, it is not necessarily sufficient to grasp the emission state, and there is a possibility that the emission becomes unstable. On the other hand, in the present embodiment, the resistance value Ro of the limiting resistor 28 is configured to be Ro >> 1 / m (= 7 MΩ) using the slope m of the Ie_V characteristic of the LMIS 31, and the flow impedance of the MIS 31 is configured. Since the change in the slope (the state of the Ie_V characteristic of LMIS) can be determined by the change in the state determination voltage value Vox, the emission of the LMIS 31 can be stably maintained and managed.

  2. Since the change of the slope of the flow impedance of the MIS 31 (the state of the Ie_V characteristic of the LMIS 31) can be determined by the change of the state determination voltage value Vox, in order to obtain the inclination m of the Ie_V characteristic and the state determination voltage value Vox, There is no need to measure a plurality of sets of Ie and Vext, Vox can be obtained from one set of Ie and Vext, and the state of the LMIS 31 can be determined in a short time (real time). Thereby, the emission stability and life management of the LMIS 31 can be easily performed.

  3. In the prior art, when the emission current Ie is adjusted by changing the output voltage Vext, Vext and Ie change greatly before and after the adjustment, so that it is determined that the beam adjustment is abnormal, and is a target of warning display. On the other hand, in this embodiment, since the necessity of beam adjustment depends on the amount of change in Vox, the necessity for beam adjustment is determined by observing the amount of change in Vox with respect to Vox immediately after beam adjustment. Since Vox does not change before and after the change of Vext, it is possible to adjust the emission appropriately by adjusting Vext while determining the necessity of beam adjustment by Vox.

  4). Since Vox, which is an index for beam adjustment, is used to judge the quality of the emission state, if emission is stopped with the beam adjustment of the focused ion beam device, Vox is the same value when emission is restarted. If the adjustment is made, the emission can be easily restarted without adjusting the beam.

  5. Since Ga / W is used for the beam limiting aperture, the change in flow impedance with time can be reduced, and the change in focus can be reduced.

It is a figure which shows the whole structure of the focused ion beam apparatus of this Embodiment. It is a figure which shows the structure of LMIS in an ion gun. It is a block diagram which shows the outline | summary of the function of a control part. It is a figure which shows the measurement result of the emission range at the time of irradiating a fluorescent particle with a charged particle beam. It is a figure which shows the relationship between an emission current and the half opening angle of an emission. It is a figure which shows the relationship between an emission current and a radiation angle current density. It is a figure which shows the relationship between an emission current and a beam current. It is a figure which shows the spread of the energy of the radiation beam with respect to the emission current at the time of using GaLMIS. It is a figure which shows the change of J (omega) / (DELTA) E ^ 2 with respect to an emission current. It is a figure which shows the relationship between the change of an emission current, and the output voltage output from an extraction power supply. It is a figure which shows typically the physical phenomenon of a LMIS front-end | tip. It is a figure which shows the TEM image of the LMIS front-end | tip. It is a figure which shows the relationship between the output voltage at the time of changing a limiting resistance, and an emission current. It is a figure which shows the relationship between the difference of an extraction voltage, Vox, and an emission current. It is a figure which shows the relationship between an output voltage and an emission current. It is a figure which shows the change of the inclination of Ie_V characteristic with respect to the difference in the magnitude | size of a limiting resistance. It is a figure which shows the measurement result of the time-dependent change of emission current. It is a figure which shows the measurement result of the time-dependent change of emission current. It is a figure which shows the measurement result of the time-dependent change of emission current by the relationship between an output voltage and an emission current. It is a figure which shows the measurement result of the time-dependent change of emission current. It is a figure which shows the measurement result of the time-dependent change of emission current. It is a figure which shows the flowchart of a starting process. It is a figure which shows the flowchart of a start completion process. It is a figure which shows the flowchart of a normal process. It is a figure which shows the flowchart of an emission completion | finish process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Vacuum container 2 Charged particle beam 3 Ion gun 4 Objective diaphragm 5 Deflector 6 Sample 7 Objective lens 8 Signal 9 Signal detector 10 Evacuation device 11 Gate valve 12 Ion pump 20 High voltage power supply part 21 Ground electrodes 22, 25 High voltage Cables 23 and 26 High voltage connection 24 Acceleration power supply 27 Extraction power supply 28 Limiting resistor 29 Heating power supply 31 LMIS
32 Extraction electrode 33 Beam limiting aperture 34 Ground electrode 35 Needle-like emitter 36 Reservoir 37 Filament 38 Current terminal 39 Insulator base 40 Control unit 41 Calculation unit 42 Storage unit 43 Display unit

Claims (13)

A liquid metal ion source;
An extraction electrode for extracting a charged particle beam from the liquid metal ion source;
An extraction voltage source for applying an extraction voltage between the liquid metal ion source and the extraction electrode;
A limiting resistor provided between the extraction electrode and the extraction voltage source;
A state as an evaluation value of the state of the liquid metal ion source by using the resistance value Ro of the limiting resistor, the output voltage Vext of the extraction power source, and the emission current Ie of the charged particle beam extracted from the liquid metal ion source A discriminant value calculating means for calculating a discriminant voltage value Vox = Vext−Ie × Ro;
A focused ion beam comprising recovery processing means for performing a state recovery process for recovering the state of the liquid metal ion source when the state determination voltage value is outside a predetermined state determination voltage range apparatus. A liquid metal ion source;
An extraction electrode for extracting a charged particle beam from the liquid metal ion source;
An extraction voltage source for applying an extraction voltage between the liquid metal ion source and the extraction electrode;
A limiting resistor provided between the extraction electrode and the extraction voltage source;
A state as an evaluation value of the state of the liquid metal ion source by using the resistance value Ro of the limiting resistor, the output voltage Vext of the extraction power source, and the emission current Ie of the charged particle beam extracted from the liquid metal ion source A discriminant value calculating means for calculating a discriminant voltage value Vox = Vext−Ie × Ro;
Difference calculating means for calculating a difference between a predetermined state determination reference voltage value Vox0 and the state determination voltage value Vox;
A recovery processing unit that performs a state recovery process for recovering the state of the liquid metal ion source when the difference calculated by the difference calculation unit is outside the predetermined state determination difference range; A focused ion beam device. The focused ion beam device according to claim 1.
Storage means for storing the state determination voltage value Vox1 calculated before the recovery process and the state determination voltage value Vox2 calculated after the recovery process;
And a front-rear difference calculating means for calculating a difference between the state determination voltage value Vox1 and the state determination voltage value Vox2.
The said recovery process part performs the state recovery process for recovering the state of the said liquid metal ion source, when the said difference is outside the predetermined state determination difference range, The focused ion beam apparatus characterized by the above-mentioned. The focused ion beam device according to claim 1.
The state determination voltage value Vox is a voltage when extrapolated to the emission current Ie = 0 in the relationship between the emission current Ie and the output voltage Vext of the extraction power supply,
The focused ion is characterized in that the resistance value Ro of the limiting resistor is determined so that the relationship between the emission current Ie, the output voltage Vext, and the proportionality constant M≈1 / Ro satisfies Ie = M × (Vext−Vox). Beam device. The focused ion beam device according to claim 2, wherein
The focused ion beam apparatus is characterized in that after the state recovery process is executed at least twice, a state determination voltage value Vox3 is calculated and the state determination reference voltage value Vox0 is updated to the state determination voltage value Vox3. The focused ion beam device according to claim 1.
Heating means for heating the liquid metal ion source;
The focused ion beam is characterized in that the state recovery process is a flushing that temporarily raises the temperature of the liquid metal ion source by the heating means. The focused ion beam device according to claim 1.
The focused ion beam apparatus according to claim 1, wherein the state recovery process is high-current irradiation in which an extraction voltage of the extraction power supply is temporarily increased to temporarily make an emission current a large current. The focused ion beam device according to claim 1.
The focused ion beam apparatus, wherein the state recovery process is an emission warning for notifying an operator of an emission abnormality or a beam adjustment warning for notifying an operator of an abnormality of beam adjustment. The focused ion beam device according to claim 1.
A focused ion beam apparatus comprising a display unit that displays the state determination voltage value calculated by the calculation unit. The focused ion beam device according to claim 1.
The focused metal beam apparatus is characterized in that the liquid metal ion source is a gallium liquid metal ion source. The focused ion beam device according to claim 1.
A beam limiting aperture for limiting the irradiation of the charged particle beam extracted from the liquid metal ion source,
The focused ion beam apparatus, wherein the beam limiting diaphragm is formed using gallium and tungsten. A liquid metal ion source, an extraction electrode for extracting a charged particle beam from the liquid metal ion source, an extraction voltage source for applying an extraction voltage between the liquid metal ion source and the extraction electrode, the extraction electrode and the extraction voltage A method of controlling a focused ion beam device comprising a limiting resistor provided between sources,
A state as an evaluation value of the state of the liquid metal ion source by using the resistance value Ro of the limiting resistor, the output voltage Vext of the extraction power source, and the emission current Ie of the charged particle beam extracted from the liquid metal ion source A determination voltage value Vox = Vext−Ie × Ro is calculated, and when the state determination voltage value is outside a predetermined state determination voltage range, a state recovery process for recovering the state of the liquid metal ion source is performed. A method for controlling a focused ion beam device. A liquid metal ion source, an extraction electrode for extracting a charged particle beam from the liquid metal ion source, an extraction voltage source for applying an extraction voltage between the liquid metal ion source and the extraction electrode, the extraction electrode and the extraction voltage A method of controlling a focused ion beam device comprising a limiting resistor provided between sources,
A state as an evaluation value of the state of the liquid metal ion source by using the resistance value Ro of the limiting resistor, the output voltage Vext of the extraction power source, and the emission current Ie of the charged particle beam extracted from the liquid metal ion source The determination voltage value Vox = Vext−Ie × Ro is calculated, the difference between the predetermined state determination reference voltage value Vox0 and the state determination voltage value Vox is calculated, and the difference is outside the predetermined state determination difference range. In this case, the control method of the focused ion beam apparatus, wherein state recovery processing for recovering the state of the liquid metal ion source is performed.

Images