JPH0675053A - Energy beam diameter measuring method - Google Patents

Abstract

PURPOSE:To calculate beam diameter accurately by irradiating each of a plurality of line patterns, formed at a constant interval on the surface of a film, with a beam at different linear dose, forming a pattern having a maximal surface dose lower than a threshold level and a minimal surface dose higher than the threshold level, and then determining the ratio of linear doses corresponding, respectively, to the maximal and minimal values. CONSTITUTION:A silicon nitride film 2 is formed on a semiconductor substrate 1 and then it is scanned by a convergent ion beam 3 of Ga thus forming a plurality of line patterns 4. The film 2 is then subjected to dry etching. In this regard, relationship between the surface dose of Ga and residual film thickness rate varies in view point of threshold value thus producing two threshold values Fi, Fo. Total surface dose exhibits an irregular distribution and thereby linear dose is varied in the writing to regulate the maximal and minimal values 43, 44 thereof such that they match, respectively, with the threshold surface doses Fi, Fo. A ratio R of linear doses L1, L2 corresponding to Fi, Fo is then determined thus calculating a beam diameter delta=Df(A(R)), where D is beam interval, f is a function representative of relationship between R and delta, and A is a constant.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for measuring the beam diameter of an energy beam.

[0002]

2. Description of the Related Art In recent years, energy beams such as focused ion beams and electron beams have come to be used in various patterning processes because they enable fine patterning without a mask. In the patterning process using these energy beams, the beam diameter of the energy beam has a great influence on the patterning performance, and therefore the beam diameter needs to be accurately grasped. Recently, a method of measuring the ion beam diameter using the difference in the ion dose amount due to the superposition of ion beams has been
L. R. The method proposed by Harriott and described in J. Vac. Sci. Technol. A
8, 899 (1990). This method uses that the maximum ion dose amount when a plurality of lines are drawn by a focused ion beam changes depending on the number of lines. For this purpose, first, a Cr film is formed on a glass substrate, and the Cr film is irradiated with a focused ion beam to carry out sputter etching of Cr. At this time, the time required to sputter-etch a certain amount of Cr is measured by using the difference in secondary ion between Cr and glass. By this, the difference in the maximum ion dose amount of the line patterns having different numbers is obtained and converted to the beam diameter.

However, this method has some problems. First, in order to detect the end point of sputter etching, a very special device configuration such as incorporating a mass spectrometer for secondary ions into a focused ion beam device is required. Further, since the measurement accuracy of the beam diameter is directly related to the accuracy of the time measurement, in order to improve the measurement accuracy, a mechanism for accurately measuring the sputtering time in conjunction with the secondary ion mass spectrometer. Need.

[0004]

A conventional energy beam diameter measuring device requires a secondary ion mass analyzer and a mechanism for comprehensively controlling a focused ion beam blanking and a secondary ion mass analyzer. The device configuration was complicated.

The present invention has been made in view of the above problems, and requires a secondary ion mass analyzer and a mechanism for comprehensively controlling the blanking of a focused ion beam and the secondary ion mass analyzer. The present invention provides an energy beam diameter measuring method capable of measuring the energy beam diameter with a simple device configuration.

[0006]

According to a first aspect of the present invention, an energy beam for measuring a beam diameter is formed on a surface of a thin film formed on a substrate for each pattern consisting of a plurality of lines with a fixed line interval (D). A step of forming a pattern having a maximum value of the surface dose amount smaller than a set threshold value and a minimum value of the surface dose amount larger than the threshold value by irradiating while changing the line dose amount, and the maximum value. And a line dose amount ratio (R) according to the minimum value, and R and D are substituted into the following equation (5) to calculate the beam diameter (δ). An energy beam diameter measuring method comprising: δ = Df (A (R)) (5)

However, the function f is a function that represents the relationship between the beam diameter and the ratio of the maximum value and the minimum value of the surface dose obtained as a superposition of the distributions of the single energy beams, and A is a constant. is there.

According to a second aspect of the present invention, a thin film surface formed on a substrate is irradiated with an energy beam for measuring a beam diameter at a constant line dose amount to form a pattern having a plurality of lines with varying line intervals. A step, a step of obtaining a limit line interval (D) at which the distribution of the maximum value and the minimum value of the surface dose amount can be measured, and the beam diameter (δ) is calculated by substituting the D into the following equation (6). An energy beam diameter measuring method comprising: δ = D · g (R _pv ) ... (6)

However, the function g is a function representing the relationship between the beam diameter and the ratio of the absolute value of the difference between the maximum value and the minimum value with respect to the maximum value obtained by calculation as a superposition of distributions of single energy beams. _Yes , R _pv is the difference between the normalized absolute values.

[0010]

According to the present inventor, the function obtained by theoretical calculation is used by observing the change in the maximum value and the minimum value of the total surface dose distribution obtained as a superposition of the energy beam surface dose distributions. Since the beam diameter can be easily calculated from the above formula, the beam diameter can be measured very easily and accurately without requiring a special configuration such as a focused ion beam device incorporating a secondary ion mass spectrometer. it can.

[0011]

EXAMPLES The details of the present invention will be described below with reference to examples. Example 1

First, as shown in FIG. 1A, a silicon nitride film 2 formed on a semiconductor substrate 1 to a thickness of about 8000 Å (hereinafter abbreviated as A) is introduced into a focused ion beam apparatus. , A focused ion beam 3 of Ga for beam diameter measurement is scanned to draw a plurality of line patterns 4. That is, the line pattern 4 in which Ga is injected is formed. As shown in FIG. 1B, a part of the plurality of line patterns 4 is one in which a plurality of lines drawn at constant intervals are one pattern, and the line dose amount is gradually increased for each pattern. .

Next, as shown in FIG. 2, the semiconductor substrate 1 is set in the sample holder 20 of the dry etching apparatus 21 using the fluorine-based plasma radicals 22 to etch the silicon nitride film 2 described above. The Ga nitride-implanted silicon nitride film 2 exhibits selectivity for dry etching by the fluorine-based plasma radicals 22, and under appropriate dry etching conditions that act as a mask for the silicon nitride film 2, as shown in FIG. The relationship between the surface dose amount of Ga and the residual film thickness rate shows a threshold change. At this time,
As shown in (2), two values Fi and Fo are obtained as a surface dose amount (abbreviated as a threshold surface dose amount) that becomes a threshold value.

If the surface dose is smaller than Fi, the mask for etching is not sufficiently formed, so that the silicon nitride film 2 is completely etched.
When the surface dose amount is higher than Fo, a sufficient mask for etching is formed, and the silicon nitride film 2 is not etched and almost the initial film thickness remains. Further, when the surface dose amount is between Fi and Fo, the residual film thickness ratio changes substantially linearly between 0 and 100% with respect to the change of the surface dose amount.

By the way, since the current density distribution of the focused ion beam can be approximated by a Gaussian distribution, the surface dose amount distribution obtained by a plurality of line patterns drawn with a beam interval similar to the beam diameter is shown in FIG. As shown in a). That is, the surface dose distribution 41 of each line is obtained by superimposing the dose distributions of adjacent lines.
On the other hand, the total surface dose amount distribution 42 (referred to as the total surface dose amount distribution) 42 has unevenness.

FIGS. 4B and 4C show changes in the total surface dose distribution 42 when the line dose is changed. By changing the drawing line dose amount, the maximum value 43 and the minimum value 44 increase or decrease. As a result, the case where the maximum value 43 matches the threshold surface dose amount Fi as shown in FIG. 4B and the case where the minimum value 44 matches the threshold surface dose amount Fo as shown in FIG. 4C are realized. can do. By increasing the line dose amount, the state where the silicon nitride film 2 does not remain at all is changed from the state shown in FIG. Further, similarly, with the state shown in FIG. 4C as a boundary, the state in which the silicon nitride film 2 is uneven is changed to a state in which the silicon nitride film 2 remains almost flat. By the way, there is a relationship as shown in FIG. 5 between the line dose amount (L) irradiated with the focused ion beam and the surface dose amount (F) having the maximum value and the minimum value. Here, p: L = aF and v: L = 2 straight lines expressed by bF shows change in the maximum and minimum values, respectively, L ₁
L ₂ is a line dose amount that realizes the states of FIG. 3B and FIG. 3C, respectively. From FIG. 5, it can be seen that there is a relationship between L ₁ and L ₂ that L ₁ / L ₂ = (a / b) · (Fo / Fi) (1). Here, Fo / Fi can be expressed by using the contrast coefficient γ defined by γ = [ln (Fo / Fi)] ^-1 (2), and the irradiation energy of the focused ion beam and the dry etching condition are constant. If so, it will be a substantially constant value. Also a
/ B as a function of the ratio of the beam diameter δ and the beam spacing D is shown in FIG. However, the beam diameter here means a value twice the distance from the center to the point where 1 / e of the peak value of the Gaussian distribution is reached. From the above, Since the relationship is derived, the beam diameter δ is calculated from the ratio of the line dose amounts L ₁ and L ₂ indicating the two boundary states and the beam distance D.
It is possible to ask. Here, if L ₂ / L ₁ = R and the inverse function of the function h is f, the equation (5) is derived. δ = D · f (A (R)) (5)

However, the function f is a function representing the relationship between the beam diameter and the ratio of the maximum value and the minimum value of the surface dose obtained as a superposition of the distributions of the single energy beams, and A is a constant. is there. Since FIG. 6 is a graph showing the relationship between the ratio of the maximum value and the minimum value of the surface dose amount and the beam diameter, it corresponds to both the function f and its inverse function h. Specifically, the beam diameter was measured as follows.

In the present embodiment, first, in the case of measuring a beam diameter of about 0.5 to 2 μm, five lines having a length of 5 μm are drawn at a beam interval of 1 μm to form one line pattern, and the line dose amount is set. 5.0 × 10 ¹¹ cm ^-1 to 3.0
161 patterns were drawn as shown in FIG. 1 while changing the accuracy up to × 10 ¹³ cm ^-1 with two digits.

Next, the sample irradiated with the ion beam was introduced into the microwave-excited dry etching apparatus shown in FIG. 2, and the carbon tetrafluoride gas pressure was set to 1.5 × 10 ^-1 Torr for about 3 minutes. Dry etching was performed. When observing the sample surface after dry etching with an optical microscope, 1.2
Unevenness began to appear from the pattern with a line dose amount of × 10 ¹² m ⁻¹ , and a flat surface was formed again from the pattern with a line dose amount of 2.9 × 10 ¹² cm ⁻¹ . That is, L ₁ = 1.2 × 10
^{It is 12} cm ⁻¹ and L ₂ = 2.9 × 10 ¹² cm ⁻¹ .

Here, when a Ga focused ion beam is injected into silicon nitride at an acceleration voltage of 40 kV and dry etching is performed under the above conditions, the contrast coefficient γ is required to be 4.02. From these, the ratio a / b of the maximum value and the minimum value is found to be 1.88 by using the equation (4).
From this value, from the graph of FIG. 6, 0.87 is obtained as the ratio δ / D between the beam diameter and the beam interval. Beam spacing is 1
Since it is μm, the beam diameter was finally determined to be 0.87 μm. Since the focused ion beam drawing is performed so that the accuracy of the line dose amount becomes two digits, it can be understood that this embodiment is the measurement of the beam diameter with the measurement accuracy of about 1%.

A plurality of line patterns are shown in FIG. 1 of this embodiment.
As described above, a plurality of lines having the same line dose amount may be formed in parallel, but other than this, for example, a line pattern in which the line dose amount is sequentially changed for each line may be used, or the line dose amount can be known after drawing. If it is set, it may be formed randomly. Example 2 In this example, the same parts as in Example 1 are assigned the same reference numerals and detailed explanations thereof are omitted.

First, as shown in FIG. 7A, a silicon nitride film 2 having a thickness of about 800 nm formed on a semiconductor substrate 1 is introduced as a measurement sample into a focused ion beam apparatus, and the beam diameter is changed. Focused ion beam 3 of Ga to be measured
And a plurality of line patterns 4 are drawn in the same manner as in the first embodiment. As shown in FIG. 7B, a part of the plurality of line patterns 4 is a plurality of lines drawn at a constant line dose amount and a certain interval, and each pattern corresponds to a line lateral direction interval. The beam spacing is gradually reduced. When measuring a beam diameter of about 1 μm, the beam interval is reduced from 1.5 μm to 0.2 μm in steps of 0.02 μm for each pattern, while the length is 10 μm and the line dose amount is 2.5 ×. 10 ¹² cm
^-1 draws 66 patterns. The relationship between the beam interval of the line pattern and the total surface dose distribution of Ga irradiated on the measurement sample is shown in FIGS. 8A and 8B because the ion current density distribution of the focused ion beam can be approximated by a Gaussian distribution.
It changes like (b). By superimposing the surface dose distributions of adjacent lines, the maximum and minimum values (hereinafter referred to as unevenness) in the total surface dose distribution 82 ₁ when the beam interval is wide with respect to the surface dose distribution 81 of each line. (FIG. 8A). Further, when the beam interval is narrow, unevenness does not occur in the total surface dose distribution 82 ₂ (FIG. 8).
(B)). Here, since the depth at which the silicon nitride film 2 is sputter-etched by the focused ion beam 3 of Ga is proportional to the surface dose amount of the irradiated Ga ions,
Unevenness reflecting the surface dose distribution is generated on the silicon nitride film 2. When this sample is taken out from the focused ion beam apparatus and observed from the upper surface of the sample with a differential interference microscope, the unevenness on the silicon nitride film 2 can be obtained as the brightness of the image. In the region where the beam interval is wide, that is, the unevenness is generated in the total surface dose distribution 82 ₁ , it appears as a line-shaped light and dark,
Brightness is not observed in a region where the beam interval is narrow, that is, where the total surface dose distribution 82 ₂ has no unevenness. By counting the number of patterns up to the boundary between the region where bright and dark are observed and the region where it is not observed, the beam interval (D) between the uneven and flat boundaries can be obtained.

The beam shape, that is, the ion current density distribution of the ion beam is approximated to a Gaussian distribution, and the ratio of the unevenness to the ratio of the beam diameter (δ) to the beam interval (D) is obtained as shown in FIG. To be However, the ratio of the size of the unevenness referred to here means the difference between the maximum surface dose and the minimum surface dose when the beam diameter is 0.5 μm when the beam interval is sufficiently large. The standardized maximum surface dose is shown. In addition, here, the beam diameter (δ) is 1 / e of the peak value of the Gaussian distribution.
Two times the distance from the center of the beam to the point having a value of (e: base of natural logarithm) is used. Where D is the following (6)
The beam diameter (δ) can be calculated by substituting it into the formula. δ = D · g (R _pv ) ... (6)

However, the function g is a function representing the relation between the beam diameter and the ratio of the absolute value of the difference between the maximum value and the minimum value with respect to the maximum value obtained by calculation as a superposition of distributions of single energy beams. is there. This equation is derived as follows. Here, R _pv is when the ratio of the maximum value and the minimum value of the surface dose amount distribution is equal to the ratio of the resolution of the means for observing the maximum value and the minimum value of the surface dose amount distribution to the maximum observed value,
The maximum value and the minimum value and the beam interval (D) at the flat boundary are given. The function g corresponds to the graph in FIG.

Here, the acceleration voltage 10 of the focused ion beam
Under the condition of 0 kV, the sputter etching depth of the silicon nitride film 2 was found to be 170 nm from an experiment for a dose amount of 1.0 × 10 ¹⁷ cm ^-2 . The beam diameter is unknown, but when Ga ions are irradiated so that the above-mentioned line dose amount is 2.5 × 10 ¹² cm ⁻¹ with a beam diameter of 0.5 μm, for example, the distance between the lines is sufficient. Even if it is wide, the maximum surface dose of the distribution is 1.13 × 10 ¹⁷ m ^-2 ,
The maximum value of the sputter etching depth is 192 nm.
Since the vertical resolution of a differential interference microscope is about 1 nm,
A change of about 0.52% in the maximum unevenness caused by Ga focused ion beam irradiation can be observed. When the change in the unevenness can be observed with an accuracy of 0.52%, the beam diameter at the boundary between the unevenness and the flat surface is required to be 1.63 times the beam interval. Here, as described above, when the 66-th pattern is observed when the 66-pattern drawing is performed by changing the beam interval from 1.5 μm to 0.2 μm and changing it at a constant reduction rate for each line, the 46th pattern is obtained. It was observed that there was unevenness up to and the pattern was flat from the 47th pattern. At this time, when many line patterns are drawn, it is complicated to associate each pattern with a line interval, but the work is easy if appropriate markers (numbers etc.) are arranged and drawn on the pattern. Will be things. From this, it was found that the interval 0.58 μm was the boundary beam interval, and the beam diameter was 1.63 times that value, so the beam diameter could be determined to be 0.95 μm. By the way, although the case where the beam diameter is 0.5 μm was assumed when obtaining the relationship shown in FIG. 9, the beam diameter in this case indicates the beam diameter of the beam used as a simple reference for determining the ratio of the unevenness. However, the actual beam diameter can be obtained by the same operation even if any value is taken, as long as it matches the beam diameter assumed when the sputter etching amount is estimated previously. Further, in the case of this embodiment, when the beam diameter changes by about 0.02 μm, one pattern serving as the boundary between the unevenness and the flat surface shifts, so the beam diameter is 1 μm.
It can be seen that the measurement accuracy is about 2% in the vicinity of m.

The purpose of the present invention is to obtain the beam diameter by observing the change in the ion dose distribution due to the line pattern drawing of the energy beam.
The following various applications are possible without departing from the spirit of the invention.

(1) The energy beam to be measured is Ga, I emitted from a single liquid metal ion source.
In addition to ion beams of n, Al, etc., Si, Ge, As, B, Be, A emitted from the alloy liquid metal ion source
An ion beam such as u or an ion beam obtained by field ionization of a gas such as H or He is also applicable because the spatial current density distribution is clear. Further, an electron beam generated by using thermionic emission or field electron emission can also be applied because its spatial current density distribution is similarly clarified.

(2) As a means for converting the ion dose distribution with which the sample is irradiated into a measurable physical quantity, Ga ions are implanted into silicon nitride capable of measuring the beam diameter in the above vacuum consistent process. However, in addition to the method by the sputter etching that uses the phenomenon that the etching rate of the Ga-implanted region decreases when dry etching with a fluorine-based etching gas, the method by ion beam exposure to an organic polymer resist and the damage by ion implantation A method utilizing an increase in wet etching rate, a method utilizing a decrease in photoelectrochemical etching rate due to ion implantation damage, and the like can also be applied in the sense of converting into geometrical unevenness. In the case of a light element having a low sputter etching rate, there is also a method of measuring the change in the refractive index when injected into a transparent medium.
In the case of a focused ion beam device capable of ion-excited deposition and etching for repairing a photomask,
It is also possible to convert the ion dose distribution into a deposition thickness or etching depth for evaluation. In a multilayer film having different secondary electron emission efficiencies, such as a metal film formed on a dielectric film, its interface acts as a threshold for sputter etching by an ion beam. That is, the time when the upper metal film is removed by sputter etching can be known by the change in secondary electron intensity during drawing or the change in brightness of the scanning ion microscope image. This method is an effective method that is compatible with the vacuum consistent process because it can measure the sample without leaving the sample for measurement in the focused ion beam device, and it can also measure in-situ during irradiation. is there. However, these conversion means are used in a region where the physical quantity to be measured shows a change of monotonous increase or decrease with respect to the change of the ion dose amount, or a physical quantity to be measured has a threshold value with respect to the change of the surface dose amount. Need to be used in. However, the number of thresholds is any as long as the relationship between the thresholds at which the maximum value and the minimum value match is clear and constant, as in the above-described embodiment, and can be regarded as a change in the phenomenon. I don't mind.

(3) To measure the unknown beam diameter,
It is necessary to change the surface dose amount from a very large value to a very small value (Example 1), and it is necessary to change the line interval from a very large value to an extremely small value (Example 2). However, it takes a lot of drawing time. Therefore, actually, the beam diameter is roughly estimated in advance from the image resolution of the scanning ion microscope or the scanning electron microscope usually incorporated in the apparatus. For example,
When the image resolution is about 0.5 μm, the beam diameter can be determined to be about 1 μm. When 66 patterns each consisting of 10 lines with a beam current of 100 pA and a length of 10 μm are drawn so that the line dose amount is 2.5 × 10 ¹² m ⁻¹ as in the above embodiment, The total drawing time is 44 minutes, and it is possible to perform the measurement in a drawing time that is sufficiently realizable. It is possible to measure the beam diameter by the same operation as in the above embodiment, even if line drawing is performed by changing the beam interval of each one instead of drawing a plurality of patterns in which the beam interval is changed little by little. . In such a case, the drawing time can be suppressed to 1/10, so that the measurement can be performed in a shorter time. Further, when ion beam exposure of polymethylmethacrylate (PMMA), which is known as a high resolution resist, is used, the dose amount showing sensitivity is 10 ¹³
Since this is about cm ^{−2, the} line dose can be measured with a smaller amount than that of the above-mentioned embodiment, and the drawing time can be further shortened. By using these applications, it becomes possible to draw a large number of lines with a small amount of change in the beam interval, and it is possible to further improve the measurement accuracy.

(4) The ion dose amount distribution replaced by the surface irregularities by sputter etching can be known by a laser microscope or a stylus type surface shape measuring instrument in addition to the differential interference microscope. In these methods for measuring changes in surface conditions (unevenness, etc.) and changes in internal conditions (refractive index changes, etc.) that reflect the surface dose distribution, the horizontal resolution may be about half the beam diameter to be measured. Further, the vertical resolution (in the case of unevenness) or the resolution of the state change (in the case of a change in the refractive index) influences the beam interval at the boundary from the unevenness to the flatness of the ion dose distribution, as described above.

(5) Although the approximation that the beam shape has a Gaussian distribution is used in the embodiment, the beam shape from the field evaporation type liquid metal ion source is actually a region having a current density of about 1% or less of the peak value. Is known to have an exponential distribution that deviates from the Gaussian distribution. However, the change in the current density amount due to the deviation is 1% or less, and the change amount decreases as the distance from the center of the beam increases. Therefore, there is almost no problem when drawing about 10 line patterns. In the above-described embodiment, the measurement accuracy is only about 1% due to the approximation of the Gaussian distribution, but the actual beam shape can be easily taken into the calculation, and the measurement accuracy can be improved. Further, the beam diameter can be measured not only by the Gaussian distribution but also by various distribution shapes by the same operation.

(6) The distribution of the ion dose irradiated to the measurement sample differs from the distribution of the number of ions injected into the sample and the distribution of the energy amount due to the scattering effect, but these relationships are also calculated. Can be incorporated into. In the case of an electron beam, the current density distribution, which can be approximated with a Gaussian distribution when irradiated to the sample, deviates considerably from the Gaussian distribution due to the contribution of backscattering after incidence on the sample and secondary electron emission.
These effects can also be included in the calculation.

[0033]

As described above, according to the present invention,
It is possible to easily and accurately measure the beam diameter of a wide range of energy beams such as electron beams and focused ion beams. Especially when measuring the beam diameter of a focused ion beam, it is possible to measure the beam diameter very easily and accurately without the need for a special configuration such as a focused ion beam device incorporating a secondary ion mass spectrometer. .

[Brief description of drawings]

FIG. 1 is a perspective view illustrating a first embodiment of the present invention.

FIG. 2 is a sectional view of an etching apparatus used in the first embodiment of the present invention.

FIG. 3 is a diagram illustrating a first embodiment of the present invention.

FIG. 4 is a diagram illustrating a first embodiment of the present invention.

FIG. 5 is a diagram for explaining the first embodiment of the present invention.

FIG. 6 is a diagram for explaining the first embodiment of the present invention.

FIG. 7 is a perspective view illustrating a second embodiment of the present invention.

FIG. 8 is a diagram illustrating a second embodiment of the present invention.

FIG. 9 is a diagram illustrating a second embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Semiconductor substrate 2 Silicon nitride film 3 Focused ion beam 4 Line pattern 20 Sample holder 21 Dry etching device 22 Fluorine-based plasma radical

Claims (2)

[Claims] 1. A surface of a thin film formed on a substrate is irradiated with an energy beam for measuring a beam diameter while varying a line dose amount for each pattern consisting of a plurality of lines with a fixed line interval (D). A step of forming a pattern having a maximum value of the surface dose amount smaller than the set threshold value and a minimum value of the surface dose amount larger than the threshold value, and a line dose amount and the minimum value corresponding to the maximum value. Energy beam diameter, which comprises a step of obtaining a ratio (R) of the line dose amount according to the above, and a step of calculating the beam diameter (δ) by substituting the R and D into the following equation. Measuring method. Note δ = D · f (A (R)) However, the function f is the relationship between the beam diameter and the ratio of the maximum value and the minimum value of the surface dose obtained by calculation as the superposition of the distributions of the single energy beams. It is a function to represent, and A is a constant. 2. A step of irradiating a surface of a thin film formed on a substrate with an energy beam for measuring a beam diameter at a constant line dose amount, and forming a pattern in which a line interval composed of a plurality of lines is changed; A step of obtaining a limit line interval (D) at which the distribution of the maximum value and the minimum value of the dose amount can be measured,
Calculating the beam diameter (δ) by substituting the D into the following equation. Note δ = D · g (R _pv ), where the function g is the ratio of the absolute value of the difference between the maximum value and the minimum value to the normalized maximum value obtained by calculation as a superposition of distributions of single energy beams. And R _pv is a standardized absolute value difference.