JP2735192B2 - Energy beam processing method and apparatus - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a processing depth control technique for processing an LSI or an exposure mask using a focused ion beam or the like. [Prior Art] In recent years, in the development process of LSI, it has become very important to debug, correct, or analyze a defective portion by cutting or connecting a part of wiring in an LSI chip. . For this purpose, conventionally a focused ion beam
An example of cutting an LSI wiring has been reported. For example, Japanese Patent Application Laid-Open No. 58-106750 (Focus / Ion beam processing method) shows that it is possible to perform processing with different etch depths by changing the irradiation amount irradiation time of the ion beam, the acceleration voltage, and the like. . [Problems to be Solved by the Invention] In the above prior art, how to determine when a desired machining depth is obtained, and how to stop machining,
It merely describes that the irradiation time, the irradiation amount, and the like are made variable. Although an equation representing the etching depth S is shown in P4, there is no specific description as to how the target depth can be processed from this equation. As a specific processing end point detection method, a method of analyzing secondary ions is mentioned. However, the LSI employs multi-layer wiring, and in order to cut the lower layer wiring, as shown in the cross section of the processed part in FIG.
Typically, holes having a high aspect ratio such as a processing area of 5 μm and a processing depth of 10 μm must be formed. When the processing depth is shallow as shown in FIG. 2, a sufficient amount of secondary ions 29 are captured by the secondary ion detector 30, but when the aspect ratio is high as shown in FIG. 29
Is hardly detected. Therefore, it is impossible to detect the processing end point by this method. On the other hand, if the beam current and the acceleration voltage are constant, the processing depth is proportional to the processing time. When the processing depth is shallow, the processing time is short. Therefore, it was assumed that the beam current was constant within this time, and no great error occurred even if the depth was controlled by the processing time. However, in the hole shown in FIG. 3, for example, since a volume of 5 × 5 × 10 = 250 μm ³ is processed at a typical processing speed: 0.3 μm ³ / S, it takes about 14 minutes. As shown in the figure, the drift of the beam current iB cannot be ignored. Drift can exceed 10%. For this reason, as shown in FIG. 5, if the processing time is set based on the initially set current value and the beam current subsequently drifts in the direction of decreasing, the actual depth becomes insufficient, and the cutting of the wiring 31 may fail. Can not. Conversely, if the actual current value drifts in a direction that increases from the initial set current value, it will be processed deeper than the target depth, so it will be processed to the lower layer wiring, and it will be reattached 33 from the lower layer wiring. Problems such as a short circuit with the upper wiring occur. An object of the present invention is to solve the above-mentioned drawbacks of the prior art, and to provide an energy beam processing method capable of controlling a processing depth with high accuracy without being affected by a change in beam current during processing, and an energy beam processing method therefor. It is to provide a device. [Means for Solving the Problems] The object is to measure a beam current at a sufficiently short time interval during processing, and to integrate the measured beam current with time (hereinafter, referred to as a dose amount) to obtain an irradiation ion amount. This is achieved by calculating the machining depth. That is, the above-mentioned object is to irradiate a predetermined area of a workpiece by focusing and deflecting an energy beam and irradiate the energy beam at predetermined time intervals in an energy beam processing method of processing the predetermined area. By irradiating the current measuring unit, the current of the deflected energy beam is measured by the current detecting unit, and a dose amount of the energy beam irradiated to a predetermined area is obtained from the measured current value of the deflected energy beam. Based on the dose amount and the sputter volume per unit incident energy beam amount of the workpiece and the opening area at the start of processing, or based on the relationship between the dose amount and the processing depth obtained in advance, the predetermined area of the workpiece is Calculate the machining depth, compare the calculated machining depth value with a predetermined value, detect the machining end point, and calculate the energy to a predetermined area. It is achieved by employing a method and apparatus for stopping the irradiation of over beam. Further, the above object is to provide an energy beam processing method for focusing and deflecting and scanning an energy beam on a predetermined area of a workpiece and processing the predetermined area, and irradiating the energy beam on a predetermined area of the workpiece. The current of the energy beam flowing into the lens electrode that focuses and deflects the energy beam during the measurement is measured, and the dose amount of the energy beam applied to a predetermined area is determined from the measured current value of the energy beam flowing into the lens electrode. The predetermined amount of the workpiece is determined based on the dose amount, the sputter volume per unit incident energy beam amount of the workpiece, and the opening area at the start of processing, or based on the relationship between the dose amount and the previously determined processing depth. The processing depth of the area is calculated, the calculated processing depth value is compared with a predetermined value to detect the end point of the processing, and It is achieved by employing a method and apparatus for stopping the irradiation of the energy beam to pass. [Operation] In the focused ion beam processing, the atoms 34 sputtered by the ion beam 28 are reattached to the side wall of the processing hole to form a re-adhesion layer 35 as shown in FIG. 6, and the side wall of the processing hole is inclined.
The change in the shape of the processing hole due to the reattachment layer affects the progress speed of the processing depth. We obtained the relationship shown in FIG. 7 as a result of the experiment. First, the processing width L is sufficiently large with respect to twice the beam diameter d (2d) (4d
In the above case, the processing depth Z is proportional to the dose D while the processing is started and a flat portion remains on the bottom surface of the processing hole.
Further processing is performed, and after the flat portion of the processing bottom surface is lost due to re-adhesion and the processing hole becomes wedge-shaped, the progress speed of the processing depth Z becomes slow as shown in FIG. Next, when the processing width L is narrower than the above case, that is, when L is substantially equal to or less than twice the beam diameter d, Z and D
Does not show a proportional relationship, and the progress speed of Z decreases with the progress of processing. Therefore, a depth control method was invented for each of the case where Z is proportional to D and the other cases. When Z is proportional to D, first, the beam current is measured at sufficiently short time intervals during processing, and this is time-integrated to obtain a dose D, and the depth Z is obtained by multiplying D by a proportional constant. Here, the sputter volume per unit incident ion amount of the material M to be processed is k _M [μm ³ / n _C ] (hereinafter referred to as the processing speed coefficient of the material M), and the opening area at the start of the processing is A [μm ² ] (A =
L ₁ , L ₂ , L ₁ , L ₂ are the vertical and horizontal widths of the machined portion. Machining depth Z
Ignored redeposition and assumed that all sputtered atoms were gone. Since it is obtained by dividing the rectangular processing hole volume V (hereinafter referred to as the sputter volume) by the opening area A, the following equation holds. Therefore, the proportional constant between Z and D is k _M / A. Next, when Z and D are not proportional, obtaining the dose amount D is the same as described above. After obtaining D, the depth Z is obtained using a processing depth function Z = g (D) obtained in advance by a trial processing experiment. [Embodiment] Hereinafter, an embodiment of the present invention will be described with reference to FIGS.
This will be described with reference to FIGS. In FIG. 10, the ion beam extracted by the ion source 1 includes first, second, and third lens currents (2, 2 in the figure, respectively).
The light is focused so as to be focused on the sample 8 according to (3) and (4).
By applying a voltage to the blanking electrode 5 as needed, the beam is bent and applied to the blanking aperture 6,
Irradiation to the sample 8 can be eliminated. By applying a deflection voltage to the deflector electrode 7, the beam can be scanned in the processing area. The sample 8 is fixed to a stage 9, and the stage 9 is driven by a driving device (not shown). During processing, the stage 9 is fixed, and the beam is deflected and scanned for processing. The blanking electrode 5 and the deflector electrode 7 respectively
Blanking controller 11, Deflector controller 1
0 supplies the required voltage. FIG. 1 shows a flowchart for controlling the processing depth by the apparatus shown in FIG. In the case of processing, first, the beam diameter d and the processing widths L ₁ and L ₂ (vertical width and horizontal width) are set as processing parameters. d and the size of L ₁ , L ₂
It is determined whether or not the depth Z is proportional to the dose amount D, and the machining is started in different flows. Where Z and D
Since the relationship between d, L ₁ , and L ₂ for determining the proportional relationship varies depending on the workpiece, it is necessary to obtain the relationship in advance by an experiment. For example, LS consisting of Al wiring and sputtered SiO ₂ film
When processing multi-layer wiring of I, when L ₁ ≧ 4d and L ₂ ≧ 4d, Z∝D, when L ₁ <2d or L ₂ 2d, Z is not proportional to D. Required depth accuracy in the middle area , It is determined by an experiment that whether or not Z と D can be approximated is determined according to. When Z is proportional to D, machining is performed according to the flow shown on the left side in FIG. Processing start at the same time to measure the beam current i _B for every predetermined time t _S exerts a timer. The sampling time t _S is selected when the fluctuation of the beam current within this range is sufficiently small. The measurement of the beam current is performed, for example, in the tenth
As shown in the figure, each scanning processing, when operating the blanking, measuring ion currents i _BL flowing toward the ground flows into the blanking aperture 6. The structure of the blanking aperture 6, by a shape encompassing all the secondary electrons generated, i _BL is equal to the value of the beam current i _B. At this time, the minimum unit of the sampling time t _S is one scanning time, which is a sufficiently short time. The method of measuring _iBL will be described with reference to FIGS. 11, 12, and 13. FIG. When the blanking is activated and the ion beam is blocked by the blanking aperture 6, secondary electrons 36 are generated from the blanking aperture 6 due to collision of ions.
When the generated secondary electrons 36 are drawn away by, for example, an electric field generated by the blanking electrode 5, a current i _{BL in} which the current of the secondary electrons is increased flows through the ammeter 12. The i _BL no longer coincide with the irradiation beam current i _B during processing. Therefore in order to match the measured current i _BL to the irradiation beam current i _B, it is necessary to capture the secondary electrons 36 generated at all blanking aperture 6. In the example shown in FIG. 11, an eave-shaped portion for blocking between the blanking electrode 5 and the blanking electrode 5 is provided above the blanking aperture 6. As a result, the secondary electrons 36 are not affected by the electric field generated by the blanking electrode 5, and can be entirely captured by the blanking aperture 6. In the example shown in FIG. 12, a Faraday cup 37 is provided in the blanking aperture 6, and during the blanking operation,
The ion beam 28 was made incident on the Faraday cup 37. As a result, the secondary electrons 36 are all Faraday cups 37
And the measured current i _BL from the Faraday cup 37 matches the illumination beam current i _B. In the example shown in FIG. 13, a secondary electron trap electrode 38 is provided between the blank aperture 6 and the blanking electrode 5, and a trap voltage is applied from a secondary electron trap power supply 39. The secondary electron trap electrode 38 is formed, for example, by making a hole in the center of a flat plate electrode and providing a metal mesh there. Since the energy of the secondary electrons 36 is on the order of several eV to several tens eV, the trap voltage may be set to about -100V. As a result, all the secondary electrons 36 are repelled toward the blanking aperture 6 and can be caught by the blanking aperture 6. Note that the influence of the trap voltage on the convergence of the ion beam 28 is slight, and the correction can be easily performed by the lens voltage or the like. Next, a method of indirectly calculating the value of i _B from other current values by calculation will be described with reference to FIGS. 14, 15, 16, and 17. FIG. FIG. 14 shows an example in which the total ion current (hereinafter, referred to as source current i _S ) emitted from the ion source and flowing into the first lens electrode 2 (extraction electrode) is used as the measurement current. Drawer power supply 41
The source current i _S flowing through is measured by the ammeter 42 and the A / D
After being digitized by the converter 43, the measured value is transmitted to the CPU 45 via the optical data link 44. Here, since the measurement system of i _S is floating with respect to the ground by the acceleration voltage, the optical data link 44 is used for electrical insulation. The CPU 45 calculates the value of i _B from the measured value i _S using the relationship i _B = F (i _S ) (3) between the source current i _S and the irradiation beam current i _B obtained in advance by experiments. Result of performing actual experiments, the relationship F of the i _S and i _B
(I _S ) could be expressed with sufficient accuracy by the linear function i _B = α ₁ ′ i _S + β ₁ (4) as shown in FIG. Next, FIG. 16 shows an example in which the remaining current (hereinafter, referred to as aperture current i _A ) whose irradiation ion amount is limited by the third lens electrode 4 (beam limiting aperture) is used as the measurement current. It flows from the third lens electrode 4. The aperture current i _A is measured by the ammeter 46, digitized by the A / D converter 43, and transmitted to the CPU 45. In the CPU 45, as in the case of using the source current i _S , the measurement is performed using the relationship i _B = G (i _A ) (5) between the aperture current i _A obtained in advance by experiment and the irradiation beam current i _B. to calculate the value of i _B from the value i _a. According to experiments, the relationship G (i _A ) between i _A and i _B is also sufficiently accurate by the linear function i _B = α ₂ ′ i _S + β ₂ , as shown in FIG. Could be shown. In these methods of measuring the value of i _S or i _A and calculating the value of i _B by calculation, the function F (i _S ) or G
The accuracy of (i _A ) affects the depth accuracy. F (i _S ) or G
When (i _A ) was approximated by a linear function, a depth control of ± 0.3 μm was possible for a processing depth of 5 μm. However, when processing of 5 μm or more is required, or when higher depth accuracy is required, the accuracy of F (i _S ) or G (i _A ) may become a problem. The structure of the blanking aperture is slightly more complicated,
In order to obtain a higher depth accuracy, the value of i _B is measured directly using an i _BL. The methods described above are suitable. In the product of i _B [nA] and t _S [sec] obtained as described above,
By multiplying by the processing speed coefficient k _M [μm ³ / n _C ], the increment ΔV [μm ³ ] of the sputter volume within the sampling time is obtained. ΔV = k _M ′ i _B ′ t _S (7) ΔV is added to the sputter volume V [μm ³ ] from the start of processing, and the opening area A [μm ² ] (A = L ₁ , L ₂₎ ), The current machining depth Z [μm] can be calculated. V = V + k _M ′ i _B ′ t _S (8) Z = V / A (9) Processing is performed until the obtained Z exceeds the set target depth Z _0.
repeated every t _S, it stops the processing at the time of exceeding the Z _0. Among the above flow, the value of the processing speed coefficient k _M varies with the material M of the workpiece. Therefore, when processing a multilayer sample, it is necessary to determine the material M for each sampling of t _S and determine the value of k _M. For the determination of the material M, for example, the thickness of each layer of the workpiece is measured in advance with an interferometer or the like, the material M is obtained as a function f (Z) of the depth Z, and the determination is made based on the value of Z at the time of sampling. Do. FIG. 8 shows an example of the material function f (Z). When processing the multilayer structure shown in FIG. 8 in order from the upper layer, the processing depth Z
The material changes as shown in the graph as the process progresses, and this becomes a material function f (Z). Next, when Z is not proportional to D, processing is performed according to the flow shown on the right side in FIG. Processing start at the same time to measure the beam current i _B for every predetermined time t _S exerts a timer. i _B
[NA] is multiplied by t _S [sec] to obtain an increment ΔD = i _B · t _S of the dose within the sampling time, and ΔD [n _C ] is added to the dose from the start of machining to accumulate the dose D. [N _C ] is obtained. D = D + i _B · t _S (10) The function of Z and D, Z = g (D), obtained in advance
From the obtained D, the current processing depth Z [μm] is obtained. Z until exceeds the set target depth Z _0, repeated processing for each t _S,
Stop machining when Z ₀ is exceeded. FIG. 9 shows a machining depth function Z = g representing the relationship between Z and D.
The example of (D) is shown. Before performing actual processing, a test processing is performed by using a sample having the same layer structure as that of the processed part while changing D, and the depth Z of the processed hole is actually measured by an electron microscope or the like, and the above functional relationship Z = g (D) is obtained. Z = g (D)
Is a monotonically increasing function. Therefore, the trial processing does not need to be performed in detail over the entire processing depth, and is performed in detail only before and after the target depth Z ₀ , and the remaining area is Z = g (D)
May be approximated by a straight line. For example, when processing the multilayer structure shown in FIG. 9 up to the Al layer, that is, up to the target depth of 4 μm, only the region where the depth Z is about 4 μm is subjected to trial processing in detail, and the remaining region is approximated by a straight line. , A graph relationship is obtained, which is a machining depth function g (D). The configuration example of the control device that realizes the above machining flow
Shown in the figure. In the data memory 27, t _S , d, L ₁ , L ₂ , M = f
(Z), Z = g ( D), enter the required data, such as Z _0.
During the processing, the beam current i _B detected by the blanking aperture 6 is measured by the ammeter 12, and a signal digitized by the A / D converter 13 is sent to the control device. Machining simultaneously with the start, the timer 14 acts, a trigger signal is sent to the A / D converter 13 for each t _S, A / D converter 13 is operated, the value of i _B is sent to the switch circuit 15. Based on the values of the input data d, L ₁ and L ₂ , the determination circuit 16 determines the comparison relationship between Z and D, and the switch circuit 15 distributes the value of i _B to the next circuit. If Z and D are proportional, the value of i _B is sent to the multiplier 17 side. The depth Z and the material function M = f (Z) from the determination circuit 21 determines the material M, sets the value of k _M. The values of i _B , t _S , and K _M are multiplied by the multiplication circuit 17, the value of the sputter volume V is obtained by the addition circuit 18, and the division circuit 19 divides V by the opening area A to obtain the current depth Z. Is calculated. Here, the value of A is obtained from the values of L ₁ and L ₂ by the multiplying circuit 22. Z and target depth Z
₀ compared in comparator circuit 20, Z signals the blanking controller 11 at the time of exceeding the Z _0, it stops the application processing the voltage to the blanking electrode. When Z and D are not proportional, the value of i _B is sent to the multiplication circuit 23 side. The values of i _B and t _S are multiplied by the multiplication circuit 23, the value of the cumulative dose D is obtained by the addition circuit 24, and the value of Z =
The current depth Z is obtained from g (D). The comparison circuit 26 compares Z with the target depth Z ₀ , and when Z exceeds Z ₀ , sends a signal to the blanking controller 11 to stop machining. [Effects of the Invention] As described above, according to the present invention, even when processing is performed for a time that cannot be ignored, the fluctuation of the beam current,
Controls the processing depth based on the beam current value measured at sufficiently short time intervals by beam current measuring means such as a blanking aperture or Faraday cup, or the beam current value calculated from the current value flowing through the lens electrode during processing Therefore, there is an effect that drilling with high depth accuracy can be performed.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing a flowchart of an embodiment of the present invention;
2 and 3 are cross-sectional views for explaining a secondary ion detection method, FIG. 4 is a diagram showing a time change of a beam current, FIG. 5 is a diagram showing a processing state by a conventional method, and FIG. FIG. 7 is a sectional view showing a hole, and FIG.
FIG. 9 is a diagram showing a material function of the embodiment, FIG. 9 is a diagram showing a machining depth function of the embodiment, FIG.
Fig. 12, Fig. 12 and Fig. 13 show the method of measuring the beam current, Fig. 14, Fig. 15, Fig. 16 and Fig. 17 show the method of indirectly calculating the beam current It is. DESCRIPTION OF SYMBOLS 1 ... Ion source, 2 ... 1st lens electrode, 3 ... 2nd lens electrode, 4 ... 3rd lens electrode, 5 ... Blanking electrode, 6 ... Blanking aperture, 7 ... Deflector electrode, 8 sample, 9 stage, 10 deflector controller, 11 blanking controller, 1
2,42,46 …… Ammeter, 13,43 …… A / D converter, 14 ……
Timer, 15: Switch circuit, 16: Judgment circuit, 17:
Multiplication circuit, 18 addition circuit, 19 division circuit, 20 comparison circuit, 21 judgment circuit, 22 multiplication circuit, 23 multiplication circuit, 24 addition circuit, 25 judgment circuit , 26 ... Comparison circuit, 27 ... Data memory, 28 ... Ion beam, 29 ...
Secondary ion, 30: Secondary ion detector, 31: Wiring, 32
…… Interlayer insulating film, 33 …… Reattached substance from lower wiring, 34
... sputtered atoms, 35 ... redeposition layer, i _B ... beam current, i _BL ... blanking aperture current, t _S ... current sampling time, d ... beam diameter, L ₁ , L ₂ ... processing width, M ...... material, f (Z) ...... material function, Z ₅ ...... machining depth, g (D) ...... machining depth function, Z ₀ ...... target depth, k _M ...
... Material speed coefficient of material M, V ... Sputter volume, A ...
Processing hole opening area, D: dose amount. 36 secondary electron, 37 Faraday cup, 38 secondary electron trap electrode, 39 secondary electron trap power supply, 40
… Acceleration power supply, 41… Drawer power supply, 44 …… Optical data link,
45 …… CPU.

   ────────────────────────────────────────────────── ─── Continuation of front page    (72) Inventor Akira Shimase               292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa               Hitachi, Ltd. Production Technology Laboratory (72) Inventor Takahiko Takahashi               2326 Imai, Ome City, Tokyo Japan               Inside the Device Development Center at Ritsumi Works

Claims (1)

(57) [Claims] In an energy beam processing method of converging and scanning an energy beam to irradiate a predetermined region of a workpiece and processing the predetermined region, the energy beam to be irradiated is deflected every predetermined time to a beam current measuring unit. Irradiating, measuring the current of the deflected energy beam by the beam current measuring unit, and calculating the dose amount of the energy beam irradiated to the predetermined region from the measured current value of the deflected energy beam; The predetermined amount of the workpiece is determined based on the dose amount, the sputter volume per unit incident energy beam amount of the workpiece, and the opening area at the start of processing, or based on the relationship between the dose amount and the previously determined processing depth. Calculating the processing depth of the area, comparing the calculated value of the processing depth with a predetermined value to detect the end point of the processing, and detecting the predetermined area. Energy beam machining method characterized by stopping the irradiation of the energy beam. 2. In an energy beam processing method for focusing and deflecting an energy beam to irradiate a predetermined area of a workpiece by processing the energy beam, and irradiating the energy beam to a predetermined area of the workpiece, Measuring the current of the energy beam flowing into the lens electrode that focuses and deflects the energy beam, and calculates the dose amount of the energy beam applied to the predetermined area from the measured current value of the energy beam flowing into the lens electrode. Asked,
Based on the dose and the sputter volume per unit incident energy beam amount of the workpiece and the opening area at the start of processing, or based on the relationship between the dose and the processing depth obtained in advance, the processing of the workpiece is performed. Calculating the processing depth of a predetermined area, comparing the calculated processing depth value with a predetermined value, detecting the end point of the processing, and stopping the irradiation of the energy beam to the predetermined area. An energy beam processing method characterized by the above-mentioned. 3. In an energy beam processing apparatus that focuses, deflects, scans, and irradiates an energy beam onto a predetermined area of a workpiece and processes the predetermined area, the energy beam to be irradiated is deflected at predetermined time intervals to deflect the energy beam. Beam current measuring means for measuring the current of, and a dose calculating means for calculating the dose of the energy beam irradiated to the predetermined area from the current value of the deflected energy beam measured by the beam current measuring means, Based on the dose amount calculated by the dose amount calculation means, the sputter volume per unit incident ion amount of the workpiece and the opening area at the start of processing, or based on the relationship between the dose amount and the processing depth obtained in advance. Calculating a machining depth of a predetermined region of the workpiece, comparing the calculated machining depth value with a predetermined value, and ending the machining. Energy beam machining apparatus being characterized in that a end point determination means for stopping the irradiation of the energy beam is detected and the predetermined region. 4. 4. The energy beam processing apparatus according to claim 3, wherein said beam current measuring means is constituted by a blanking aperture. 5. The blanking aperture is an eaves-shaped antireflection plate for preventing secondary electrons generated by the energy beam applied to the aperture at the time of measuring the current of the energy beam to prevent the influence of an electric field generated by a blanking electrode. 5. The energy beam processing apparatus according to claim 4, wherein the energy beam processing device is provided between the energy beam processing device and the blanking electrode. 6. When the energy beam is focused and deflected to irradiate a predetermined area of the workpiece and irradiates the predetermined area of the workpiece, and in the energy beam processing apparatus for processing the predetermined area, the energy beam is radiating the predetermined area of the workpiece Beam current measuring means for measuring the current of the energy beam flowing into the lens electrode for focusing and deflecting the energy beam; and determining the predetermined value from the current value of the energy beam flowing into the lens electrode measured by the beam current measuring means. Dose amount calculating means for calculating the dose amount of the energy beam applied to the region; and the dose amount calculated by the dose amount calculating means, the sputter volume per unit incident energy beam ion amount of the workpiece, and the opening area at the start of processing. Or the predetermined amount of the workpiece based on the relationship between the dose and the processing depth obtained in advance. End point determination for calculating the processing depth of the area, comparing the calculated processing depth value with a predetermined value, detecting the processing end point, and stopping irradiation of the energy beam to the predetermined area. And an energy beam processing apparatus.