JPH03138921A - Method of correcting positional deviation - Google Patents

Abstract

PURPOSE:To reduce positional deviation by measuring the distance along a coordinate axis of marks on an object, and determine the inclination of the object with respect to a machining apparatus to correct the position. CONSTITUTION:Registration marks 2 are put at four corners A, B, C and D of an LSI chip 1. The distance between two points on the LSI chip is determined by means of laser gage 7, the reading of beam deflection voltage, and an image 10 on SIM(scanning ion microscope) in order to detect the inclination of the chip. The inclination is corrected by mechanical adjustment of the chip, or else by coordinate transform at the point that is irradiated with a laser beam. In this manner, if a chip is inclined, the positional deviation due to the inclination is corrected accurately and quickly with ease.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention is applicable to detecting the amount of inclination of a workpiece relative to surrounding structures in a device that requires highly precise positioning, such as a focused ion beam processing device. The present invention relates to a method for correcting the slope component of C%.

[Conventional technology]

In focused ion beam machining, such as in electron beam lithography equipment, relief must be able to position the workpiece with high precision.

As an example of processing an LSI by focused ion beam processing/dipping, a positioning technique in the prior art will be described next.

Conventionally, in the processing of this cypress, the stage is generally moved so that the length of the axis of the ion beam system is near the processing position, based on the mark on the LSI, and the beam is deflected to the processing position. Advance the sword loe to the specified depth. The amount of workpieces in actual machining has been increasing recently, and the machining positioning is required to be fully automatic and accurate to 5φ. 1 more
Current rules require a positioning precision of ±0.5 μm, whereas future LSIs will have a precision of ±0.3 μm to ±
Water will be sprinkled on the 0.2μ corner.

It is most desirable to define the height at the top surface of the chip, which holds the LSI chip. However, since wiring patterns are usually formed on the chip surface, there is no room to secure the specified surface. In reality, it is difficult to specify the upper surface. Additionally, the focused ion beam is operated in a vacuum chamber.

For this reason, a vacuum chuck cannot be used. Furthermore, an electrostatic chuck is also undesirable because it bends the trajectory of the ion beam. Therefore, the LSI chip is pressed down from the side with a tapered presser foot. At that time, due to the way of pressing from the side, ``c''
The LSI chip may float or foreign objects may become stuck under the chip. Furthermore, the LS during failure analysis
Since the I-chip has gone through several stages of electrical testing, the condition of the chip surface is not good, and there is a high possibility that the chip will be held tilted.

The positional deviation that occurs when the chip is tilted is shown in the sixth factor. Here, calculations are made for the chip 1 that is in IVil contact with the chip holder on side AB.

Assume that the lake is high and Chip 1 is tilted as shown in Figure 6 (CB). The fine oblique angle θ is expressed as tabθ=−−bit・(11. At this time, from mark 2 by l in the X direction (
i.e. horizontally) 1ic (point P in the diagram)
), depending on the inclination of the chip, a lx
The position shifted in the positive direction will be processed. Here, δ is calculated from δ = l (--1) - (KICOXθ. Let's calculate the deviation in the actual LSI. Figure 7 (a) shows the arrangement. Assuming that the length L is 8500-, the dimension in the Y direction is E■55
A coordinate system is determined using two marks 2 separated by 00 μm IL. Then ℃, 40 in the X direction based on the mark 2 of Shikajo
Process &5' at a distance of 00μ.

At this time, 11 (AE side touches f1, CD
Assuming that the tip is tilted by a certain amount on the side, the relationship between the amount of deviation at the machining position and ru is calculated from equation (1), f2) @Figure 7T
It can calculate like AI. For example, a deviation of 1 or 2 μm occurs when a foreign object with a height of 85 μm is caught in the chip, or when the chip is tilted by that amount due to other reasons.

may fail.

! &8 Figure (α) shows the top view of the tilted chip.
Figure (lIB is a view taken along the arrow 70E-E.

Let us consider the case where the LSI chip 1 is in point contact with the chip holder at point A, as shown in the figure.

When the chip 1 is observed from above, the chip 1 is rotated around the diagonal line BD, so it appears to be a diamond shape with the AC direction contracted while keeping the length BD constant. At this time, for example, if marks 2 are located at the center of side BC and the center of side DA, they appear to move in opposite directions. Then, when the rotation angle of the chip 1 in the XY plane is determined using these as a reference, it becomes a value η shifted from the actual dust. Moreover, when determining the machining position based on mark 2 below, even though the coordinate system is distorted because the tip 1 itself is diamond-shaped when viewed from above, the coordinate system determined by the device is an orthogonal coordinate system, so here K This also causes machining position deviation.

In conventional focused ion beam processing 'V equipment, etc., the half apex angle of the beam is as small as mrad order.
Since blurring of the beam is not a problem when using a concave with a diameter of ±50μ, there is no particular consideration for the tilt of the chip.
It is simply pressed against a flat surface. However, in reality, the chip is tilted due to various I/C reasons as described above, causing misalignment during positioning.

[Problem to be solved by the invention]

In the above-mentioned conventional focused ion beam devices, etc., no consideration is given to the tilt of the tip.

There is a problem in that when the beam irradiation position is automatically determined using the mark as a reference, a one position shift occurs.

The present invention has an objective lens with a high NA of 0.9 and an optical microscope as an observation means that can accurately detect the tilt of the chip by adjusting the focus.
Due to the low a and oos, it is a national right to detect the tilt of the tip by adjusting the focus.In equipment used with electron beams and ion beams, it is possible to detect the tilt of the tip and correct the tilt to reduce the 1-position deviation. The purpose is to provide a way to obtain.

[Means to solve the problem]

The principle of the method of the present invention for achieving the above object is as follows.

6 As explained in Figure 5, if the chip 1 is tilted with respect to the X axis, the position of the mark will appear to be shifted. The present invention uses this [E inversely to calculate and correct the tip tilt amount.

As a specific configuration based on the above-mentioned principle, the method of the present invention detects the relative position between the mechanical device and the construction target.

In the method of correction, a plurality of marks are provided on the object, and coordinate axes are assumed for the fourth Jokishin machine It.

A distance in the ffi reference axis direction is measured between the plurality of marks on the object.

The invention is characterized in that the amount of inclination of the object to be constructed with respect to the mechanical device is calculated and corrected based on the above-mentioned measured values.

However, the construction using the above-mentioned mechanical equipment includes both additional weighing.

Once the amount of inclination is calculated, the inclination can also be corrected by mechanically moving the object to be constructed (that is, the object to be processed or the object to be inspected).

In addition, when the above mechanical device is used for processing using a focused ion beam processing machine, the coordinate system of the beam irradiation system is converted without moving the position of the workpiece relative to the table of the focused ion beam processing machine. It is also possible to eliminate fiber components.

[Effect]

The operation of the island 9@ will be explained by taking as an example the case where the mechanical device is a focused ion beam processing machine. FIG. 9 is a schematic perspective view showing a focused ion beam processing machine.

An ion beam 5 extracted from an ion source 4 is focused and irradiated onto an LSI chip (hereinafter simply referred to as a chip) 1 using an electrostatic lens 8 to process the LSI chip 1. Processing positioning is performed by monitoring the stage movement with the laser gauge 7, stopping the stage 15 at a predetermined position tIL, and removing the components that could not be mechanically tightened into the beam based on the measured value by the laser gauge. This is corrected by electrical deflection by a deflector 6. In addition, the focused ion beam fLsI travels on the chip IE16L, and the secondary electrons emitted by the ion bombardment at that time,! ! Or secondary ion mc
2 consisting of p (microchannel plate) etc.
Next particle detector 9 detects 17 particles, and according to the detected amount, CRT
By applying brightness modulation to the bright spot being scanned by PlJ91 on the ion beam above, a surface unevenness image of the beam scanning region can be obtained. The raising of this image is SIM image 10 (5 can
nirL! It is called 1IonMicrozcopg).

Laser gauge stage movement 'wJ's foot measurement can be guaranteed to be 0.01- by removing vibrations around the stage and maintaining temperature control. Furthermore, the beam deflection precision can be as low as cO/01 μm by optimally designing the deflection electrode and reducing electrical noise. Furthermore, the resolution of the SIM image 10 is about half the beam diameter of the focused ion beam 5, but since a beam diameter of 0.1μ can be obtained relatively easily, the image resolution is considered to be about 0.05P. It's good. Therefore, by combining the laser gauge 7, the reading of the beam deflection voltage, and the SIN image 10, the distance from one point to another point on the LSI chip 1 can be measured, but the accuracy can be less than 0.1 μ3. be.

The position and coordinates of the marks formed inside the LSI chip have an accuracy of ±0.1μ because there is no misalignment between layers as long as they are in the same layer.
It can be considered to be about m. Therefore, distances between a plurality of marks are measured using the method described above, and the amount of deviation of each distance is determined. From this, the tilt state of the chip can be detected, and accordingly, the tilt component of the chip is corrected either mechanically or by converting the seated image of the beam irradiation position. As a result, the deviation of the beam irradiation position can be easily compensated for. [Embodiment] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

(Embodiment) This embodiment is an example of a method in which the tilt type of the chip is detected and then the tilt of the chip is mechanically corrected.

As shown in Figure 1, the four corners A#B of IIC, LSI chip 1
, C, and D. In this example, the distance between these marks is measured to calculate the amount of tilt of chip 1. Distance, L,
The purpose of measuring the marks is to quickly separate cases where the machine is rotating around the diagonal line AC and cases where it is rotating around the diagonal line BD. D
It can also be determined by what coordinates the mark B takes with respect to the coordinate system set by the marks .In that case, the formation of the mark C becomes unnecessary and the classification pattern can be formed.

After fixing mark 2 of A at the center of house 51M1, the enemy takes the SIN image at a position moved by the design dimension L1 of the distance between the marks in the On the wax in Figure 2.

Suppose you can see it. Here, the intersection of the crosses on the screen indicates the position (referred to as point M) that should be located after movement by the design dimension determined from the reading of the laser gauge. The position of mark 2 is shifted to the lower left compared to point M. In order to reduce the distance between the mark center and point M to 7 km, the following unfortunate method is often used. Here, a mark formed in a rectangular shape on the top layer of aluminum wiring is used. First, in order to examine the deviation in the X direction, an X, iM&c ion beam is scanned across this mark in a line 11. Then, the signal 12 detected by the secondary particle detector becomes a signal that rises sharply at the edge of the mark pattern. This peak position k is averaged over all fif&c to give Xl, XjkX. The distance δX from the midpoint between Xl and Xs to the x-seated image of point M is the deviation X&. Ri is also frozen in the same way. Therefore, the magnitude of the deviation a is expressed as δToge-F:F+a%・-tsh, and the inclination angle θ can be calculated from this equation (51) and equation (1). By determining L, half of the tilt state of the entire chip can be understood.

At this stage, it has not yet been determined in which direction the chip is tilted. So, if you go.

Attach the chip holder 15 and the three piezo probes 14 to the island shown in Figure 5.
The piezo element 14 is placed on a structure supported at five points by
A voltage is applied to tilt the chip 1 slightly in a certain direction. On top of that, KLIIL again as above! I□L
, , L4 distance deviation is measured, and it is determined whether it is larger or smaller than the previous time, and the direction of the chip inclination is grasped. Since the amount and direction of the tip inclination are now known, a predetermined voltage is applied to the piezo element 14 accordingly, and the correction of the tip inclination is completed.

It is also possible to make bamboo roses manually. The example is bulk 4
When the center of the mark is shifted from point M as shown in the figure (α),
While looking at the SIN image 1o, try extending some piezo chord 14. At that time, when moving away from the KM point as shown in Figure 4 (side), the piezo element 14 is conversely contracted, as shown in Figure 4 (side).
? ), move the piezo element until the point 1 and the center of mark 2 finally match. This is Lll LII
LII L4 is carried out, and Ii# (is carried out) where the tip inclination is the minimum.

(Mei 6 example) In the above example, the tilt of the chip was mechanically removed using a piezo probe.However, it is not necessary to move the chip mechanically to change the coordinate system of the mark reference to the chip. The tilt component can also be removed by converting to a tilted coordinate system.

In this embodiment (2), the beam irradiation position in ion beam processing is subjected to software KK conversion to remove the tilt component.

In this embodiment (2), there are S marks M1.
M2. MS, according to the following steps.

α, the rotation of chip 1 on the xY plane is determined using Ml and Ml.

Measure the distance between b, &1, and M2, and let it be 4J.

c, M1 coordinates are (0, Ω) #M20 coordinates are (x, 0).

Determine the coordinates of M5 with g and hl as the origin, and convert it to (La
x*Lsy).

Find the value of -9ψ.

(Here, we will consider the case where ψ is less than or equal to ψ. The calculation procedure is almost the same when it is more than 1/2.) Let us consider the transformation of f and P points (xp, yp).

! , ψ less ψ, go to the diagonal @AC (y engineering
The length of the crystal edge drawn diagonally from the point is shortened in the case of − foot. Calculate this.

The lengths of A and Do can be calculated as follows. Note that the sister character 0 in the coordinates means the seated image at the X@Y@intestinal image thread.

The ratio of 1, , @small can be found from the designed distance L- of Ml and M5 and the actual column distance. In other words, -F, Σ=7
-(6) and the reduction/JS rate R8 is.

j, therefore, D, reaches D1=RID・s=(9).

k, where, D, - Dl... (10) If defined, the coordinates of the nine points after the contraction are Kame, Shitakari,
The distance L8 after shrinking is A-R proverb LO! -(15) is accumulated.

Darkness, here 5η Mi D category - C - (16) When defined, the coordinates of the point R after shrinking are X.

! , which further overflows to 7=J and shrinks. Let the slenderness ratio be the bird. R3 is here.

P, convert this from X@ Y@ seated image system (virtual coordinate system) to
(Y coordinate system actually used) It is necessary to perform K transformation. The rotation angle φ of this coordinate axis can be considered as .

Ko, lowered y=JK from 9 points! & Distance Lo from the intersection of the edge and y = x to the origin M1! It is.

q, then the coordinates of the actual beam irradiation position in the final XY seat spear system are.

becomes.

The coordinates obtained as described above are coordinate pressures that take into account the tilt component of the chip, and can exclude deviations of the beam irradiation position due to the chip tilt.

〔Effect of the invention〕

As explained above, when the detection position correction method of this @Akira is applied, a plurality of marks are provided on the workpiece, a coordinate system is assumed for the construction machine, and the above ffi is calculated for each of the plurality of marks. By measuring the distance in the direction of the reference axis and calculating the amount of inclination of the object to be constructed based on this measurement value, it is possible to correct the inclination of the object to be constructed quickly and easily, and with a high fifa. can.

For this reason, when the method of the present invention is applied to a focused ion beam knife, it is effective in correcting positional deviations due to tilting when the chip is installed in an inclined state, so processing defects due to beam irradiation positional deviations can be avoided. Can be reduced.

In the case of an electronic edge drawing device, distortion of the drawing pattern can be avoided, and various defects caused by positional deviation can be reduced in other devices as well.

[Brief explanation of the drawing]

FIG. 1 is a plan view of an LSI chip for explaining mark positions in Embodiment I of the present invention. Stone 2 is the CR7 screen of the 51M image and detection 11! It is an explanatory diagram with the number. The Wks diagram is a perspective view of a chip holder supported by a piezo element, and the fourth factor is an explanatory diagram showing adjustment of the piezo element and changes in the SIN image. FIG. 5 is a plan view of an LSI chip for explaining calculations in Example 2 of the present invention. FIG. 6(α) is a plan view of the inclined chip, and FIG. 6Cb) is a side view. Figure 7 (α) is a diagram showing the 1000-sided factor of an LSI chip, and Figure ψ ( is a chart showing the relationship between the floating dimension of the chip and the bottom of the positional deviation. Figure 1 is a plan view of the tilted chip, Figure (b) is a view taken along the line E-E. Fig. 9 is a schematic diagram showing a focused ion beam processing device. ...Electrostatic lens 7...
Laser gauge 8 - Beam deflector 9...2
Secondary particle detector 1Ω・-5IN image 11... Ion beam scanning beam 12... Detection gi number 15-- Chip holder 14Ij# Piezo element 15... Stage 16...
・Ion beam scanning Notes 17-Secondary particle detection signal ff11 Figure 11'1 Closed 2 Figure 4 Figure Y#J5 Figure 5 Figure Child 6 Figure Sanskrit 7 Figure

Claims (1)

[Claims] 1. A method for detecting and correcting the relative position between a mechanical device and an object to be constructed, comprising: providing a plurality of marks on the object; and assuming coordinate axes for the mechanical device; A detection position detection method characterized in that distances in the direction of the coordinate axis are measured between a plurality of marks on the object, and an amount of inclination of the object to be constructed with respect to the mechanical device is calculated and corrected based on the measured value. Correction method. 2. The method for correcting a detected position according to claim 1, wherein the inclination of the construction target is mechanically corrected based on the calculated inclination amount. 3. The detection position correction method according to claim 1, wherein the mechanical device is a concentrated ion beam processing device. 4. Transform the coordinate system for determining the beam irradiation position of the focused ion beam processing device without changing the amount of inclination of the workpiece, and calculate the component of the relative tilt between the workpiece and the irradiation beam. 4. The method of correcting a detected position according to claim 3, further comprising erasing the detected position. 5. Mechanically correct the inclination of the workpiece based on the above-mentioned calculated inclination amount, and determine the beam irradiation position of the focused ion beam processing device for the inclination component that cannot be completely corrected by the above mechanical correction. 4. The method of correcting a detected position according to claim 3, further comprising converting a coordinate system to complement the mechanical correction.