JP2004228605A - Etching method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an etching method capable of achieving an optimum etching based on an opening area to be etched in such a way that the opening area is reflected in etching treatment conditions. <P>SOLUTION: In this method, etching is made after etching parameters or an etching time is determined based on measured values for dimensions of a resist mask. As the etching parameters or the etching time is determined based on the measured values for the dimensions of a resist mask, the measured values for the dimensions of the resist mask can be reflected in the etching treatment conditions, thereby achieving optimum etching basing on the dimensions of the resist mask. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

  The present invention relates to an etching method according to a dry etching technique used for manufacturing a high-density semiconductor integrated circuit.

  2. Description of the Related Art In recent years, semiconductor integrated circuits have been steadily miniaturized, and dry etching, which is a processing technique, has been required to be processed with even higher precision.

  Hereinafter, an example of a conventional dry etching method will be described with reference to the drawings.

FIG. 10 is a schematic diagram showing a processing flow of an example of a conventional dry etching method for etching a silicon oxide film (SiO ₂ film) on a silicon substrate (Si substrate). Steps 1) to 5) will be described with reference to FIGS. 2A to 2C.

1) Formation of resist pattern: As shown in FIG. 2A, a pattern of a resist 23 serving as an etching mask is formed on the SiO ₂ film 22 existing on the Si substrate 21.

  2) Dimension measurement: 1) The dimensions of the mask formed by forming the resist pattern are measured. More specifically, the dimension measurement (line width) of the resist 23 shown in FIG. 2A is performed to check whether the finish of the resist 23 is as specified.

3) Etching process: As shown in FIG. 2B, the SiO ₂ film 22 on the Si substrate 21 is dry-etched using the resist 23 as a mask. In the etching process, for example, as shown in FIG. 10, the gas flow rate is CHF ₃ : 30 sccm, O ₂ : ₃ sccm, the gas pressure is 300 mTorr, the distance between the electrodes is 30 mm, the substrate temperature is 20 C, the high frequency power is 500 W, and the end point detection. Means that the second derivative of CO emission becomes zero and overetching is performed by 30%.

The point where the second derivative becomes zero means the point at which etching is completed. This is shown in FIG. 11A shows a temporal change in the CO emission intensity when the etching target at the resist opening is the SiO ₂ film 22, and FIG. 11B shows the second derivative of FIG. 11A. And the time when the broken line is drawn is the time when the etching is completed and the second derivative becomes zero.

The term "overetching of 30%" means that when the thickness of the SiO ₂ film to be etched is 100%, additional etching is performed by 30% with respect to the thickness.

4) Stripping of resist: After the etching (main etching and over-etching) of the SiO ₂ film 22 to be etched is completed, as shown in FIG. 2C, the resist 23 on the SiO ₂ film 22 is stripped.

5) Dimension measurement: As shown in FIG. 2C, after the resist 23 is stripped, the dimension x of the groove interval of the etched SiO ₂ film 22 is measured, and the finished dimension of the SiO ₂ film by etching is as specified. Check if it is done.

  According to this dry etching method, the end point detection is performed by detecting the point at which the second derivative of CO emission becomes zero, so that a constant over-etching can be achieved regardless of various film thickness variations and etching speed variations. Can be applied.

This is because, when the end point is detected as described above, even when the film thickness varies from wafer to wafer, and even when the etching rate varies, it is possible to reliably detect the end point of the etching. Over-etching can be started from that point. At this time, the time required from the start of etching to the end point detection when the SiO ₂ film 22 having a predetermined thickness is etched is set to 100%, so that a constant over-etching process is performed. A constant etching process can be performed irrespective of the speed variation. Note that constant over-etching means that the amount of etching that will be performed during the over-etching time is constant.

In FIG. 10, since the over-etching is 30%, the over-etching time is the time required to etch 30% of the thickness of the SiO ₂ film 22, and in the conventional example, the etching rate is constant, that is, the etching is performed. Assuming that the film thickness and the etching time are proportional to each other, 30% overetching is equivalent to 30% of the time required from the start of etching to the end point detection being 100%. The etching is further performed for the time required.

  However, the dry etching method as described above has the following problems.

  First, as for the phenomenon called RIE-lag in a fine pattern, the etching condition is not reflected on the dimension measurement value of the resist 23 as shown in FIG. RIE-lag refers to a phenomenon in which the etching rate decreases as the opening size decreases.

If the contents of the etching conditions are not reflected on the dimension measurement value of the resist 23, the etching is performed under the same condition (for example, based on the etching rate when the front surface of the SiO ₂ film is etched without the resist) regardless of the difference in the dimension measurement value. The condition is created, and the end point is detected by performing the etching end point detection on a wide surface).

  However, if the size of the resist 23 is small at the time of size measurement due to the presence of the RIE-lag, a sufficient etching process may not be performed at the location of the minute opening between the resists 23 and the residue may remain as a residue. Conceivable. This is a problem.

  Second, in dry etching, a phenomenon occurs in which the processed shape changes depending on the size of the etching opening area (the ratio of the area of the portion of the wafer not covered with the resist to the entire surface of the wafer). For the phenomenon, only a method that allows a processing margin can be adopted.

The phenomenon in which the processing shape changes due to the above-described etching opening area means that, for example, when etching an aluminum alloy, the etching shape becomes reverse-tapered as the etching opening area becomes larger, and the etching shape becomes more tapered as the etching opening area becomes smaller. Refers to the phenomenon that occurs. In addition, to allow a processing margin means performing processing such as increasing the time so that predetermined etching can be performed even in a small opening. However, when such processing is performed, processing conditions are constant. it is because that, when the opening size is large, will be subjected to excessive etching, for example, when etching the SiO ₂ in the configuration of SiO ₂ / Si, the fact that the etching time of Si is increased Become. Therefore, the amount of Si abrasion, the amount of lattice defects, and the like also increase.

  SUMMARY OF THE INVENTION An object of the present invention is to provide an etching method in which a measured value of a resist dimension is reflected in an etching process condition and an optimum etching corresponding to the resist dimension can be performed.

  In order to solve the above problems, an etching method according to a first aspect of the present invention is characterized in that etching is performed by determining an etching parameter or an etching time based on a measured value of a dimension of a resist.

  According to this method, the etching parameter or the etching time is determined based on the measured value of the resist size, so that the measured value of the resist size can be reflected in the etching processing conditions, and the optimum etching corresponding to the resist size can be performed. Can be.

  Further, the etching method of the second invention is characterized in that, based on the dimension measurement value d of the resist, the resist having the dimension measurement value d is etched to a predetermined etching depth p according to the following equation.

p = 1 / (1 / a × d + 1 / p ₀ )
Here, a is a constant, and p ₀ is an etching depth in a wide portion.

  According to this method, since the etching is performed to a predetermined etching depth based on the measured value of the resist dimension, the measured value of the resist dimension can be reflected in the etching process conditions, and the optimum etching corresponding to the resist dimension can be performed. It can be carried out.

  Further, in the etching method according to the third invention, the etching is performed by calculating an etching time from a measured value of the dimension of the resist based on a table showing a relationship between the dimension of the resist and an etching depth when the resist is used as a mask for etching. It is characterized by the following.

  According to this method, the etching time is calculated from the measured value of the resist dimension with reference to the table, so that the measured value of the resist dimension can be reflected in the etching processing conditions, and the optimum etching corresponding to the resist dimension can be performed. it can.

  As described above, according to the etching method of the first invention, the etching parameter or the etching time is determined on the basis of the resist dimension measurement value, so that the resist dimension measurement value can be reflected on the etching processing conditions. In addition, it is possible to perform optimal etching corresponding to the resist dimensions.

  According to the etching method of the second invention, the etching is performed to a predetermined etching depth based on the measured dimension of the resist, so that the measured dimension of the resist can be reflected in the etching process conditions, and The corresponding optimal etching can be performed.

  According to the etching method of the third aspect of the present invention, the etching time is calculated from the measured value of the resist dimension with reference to the table. Etching can be performed.

[First Embodiment]
Hereinafter, the dry etching method according to the first embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a schematic diagram showing a process flow in a first embodiment of a dry etching method of the present invention for etching a silicon oxide film (SiO ₂ film) on a silicon substrate (Si substrate). The dry etching method includes the following steps 1) to 5), and each step will be described with reference to FIGS.

1) Formation of resist pattern: As shown in FIG. 2A, a pattern of a resist 23 serving as an etching mask is formed on the SiO ₂ film 22 existing on the Si substrate 21.

  2) Dimension measurement: 1) The dimensions of the mask formed by forming the resist pattern are measured. More specifically, the dimension measurement (line width) of the resist 23 shown in FIG. 2A is performed to check whether the finish of the resist 23 is as specified. At this time, the dimension measurement value of the resist is d.

3) Etching process: As shown in FIG. 2B, the SiO ₂ film 22 on the Si substrate 21 is dry-etched using the resist 23 as a mask. In the etching process, for example, as shown in FIG. 1, the gas flow rate is CHF ₃ : 30 sccm, O ₂ : ₃ sccm, the gas pressure is 300 mTorr, the distance between the electrodes is 30 mm, the substrate temperature is 20 C, the high-frequency power is 500 W, and the end point is detected. Takes the place where the secondary differential coefficient of CO emission becomes zero, and performs over-etching by 30% as in the conventional example. However, the method of determining 30% is different from the conventional example, and the etching time is determined according to the relationship of FIG. Is determined.

FIG. 3 is a graph showing a relationship between the etching depth (Å) and the etching time (sec), and a straight line indicates a case where the etching time is proportional to the etching depth. This linear relationship indicates the relationship between the etching depth and the etching time based on the conventional concept, that is, the relationship between the etching depth and the etching time when the entire surface is etched without a resist. The curve in FIG. 3 shows the relationship between the etching depth and the etching time in the present invention. That is, the relationship between the etching depth and the etching time when the dimension measurement value d of the resist 23 is a predetermined value is shown. Curve in Figure 3, the dimension measurement d as a parameter, and the etching depth is p, the etch depth of a wide portion and p _0, when the constant a, is determined from the following equation (Proceeding of Microprocess Conferene 1988, p138).

p = 1 / (1 / a × d + 1 / p0)
The etching depth represented by this formula well matches the relationship between the actual etching time and the etching depth. The etching time is determined according to the relationship shown in FIG.

The etching depth p ₀ in the wide portion is obtained from the end point detection time and the etching rate (the gradient of the straight line in FIG. 3), and the etching depth p ₀ in the wide portion is obtained by multiplying the etching speed v × time t. When expressed as follows,

p = 1 / {1 / a × d + 1 / (v × t)}
As described above, in the etching with the resist 23 provided, the relationship between the etching depth and the etching time is not a linear relationship as shown in FIG. The depth is shallower than the depth.

  FIG. 4 shows the relationship between the etching depth and the etching time according to the above equation, using the dimension measurement value d of the resist 23 as a parameter (small dimension, large dimension, dimension ∞). FIG. 4 shows that the smaller the dimension measurement value d, the larger the curvature, and the smaller the increase in the etching depth.

Here, a method for determining the over-etching time will be described with reference to FIG. For example, a case will be described in which the thickness of the SiO ₂ film 22 to be etched is 10000 ° and overetching is performed for 30% of the film thickness, that is, 3000 °. Since the end point detection is performed at the portion etched by the characteristics of the straight line in FIG. 3, the time at the point where the straight line intersects the position of the etching depth of 10000 ° is the end point detection time, and in FIG. 3, the end point detection time is 250 seconds. . On the other hand, since the actual etching proceeds according to the curve in FIG. 3, the time at the point where the curve intersects the position of the etching depth of 13000 ° is the end time of over-etching, and it takes 450 seconds including main etching and over-etching. , 30% over-etching is performed for 200 seconds (= 450-250) from the end point detection. Therefore, the over-etching time is determined according to the relationship of the curve in FIG.

4) Stripping of resist: After the etching (main etching and over-etching) of the SiO ₂ film 22 to be etched is completed, as shown in FIG. 2C, the resist 23 on the SiO ₂ film 22 is stripped.

5) Dimension measurement: As shown in FIG. 2C, after the resist 23 is stripped, the dimension x of the groove interval of the etched SiO ₂ film 22 is measured, and the finished dimension of the SiO ₂ film by etching is as specified. Check if it is done.

  According to this dry etching method, an end point is detected by detecting a point at which the second derivative of CO emission becomes zero, and further, an etching time is determined according to FIG. Therefore, constant over-etching can be performed irrespective of various film thickness variations and etching speed variations, and irrespective of the resist dimensions.

  In determining the over-etching time, the etching is performed by calculating the etching time from the measured value of the dimension of the resist based on a table showing the relationship between the dimension of the resist and the etching depth when etching is performed using the resist as a mask. It is also possible. This etching time can be calculated as follows. That is, since the etching depth is obtained by first-order approximation, it is possible to calculate the time required for etching the etching depth of the target resist dimension portion based on the obtained etching depth. In other words, since the etching rate is obtained from the table, it is possible to calculate the time required for etching the predetermined etching depth.

  Table 1 shows an example of a table showing the relationship between the resist size and the etching depth.

また、この時間の計算をエッチング制御装置あるいはＣＩＭ（Computer-Integrated-Manufacturing ）システム（コンピュータ総合生産システム）自体で自動計算してもよい。

Further, the calculation of this time may be automatically calculated by an etching control device or a CIM (Computer-Integrated-Manufacturing) system itself.

[Second embodiment]
Here, a second embodiment for determining an etching parameter based on a measured value of a resist dimension will be described with reference to FIG. For example, when aluminum etching is considered, the relationship between the emission intensity of AlCl ^* radicals and the etching area (the area not covered by the resist) is as shown in FIG. 5, and the etching area can be found from the emission intensity of AlCl ^* radicals. . On the other hand, in aluminum etching, a mixed gas of BCl ₃ + Cl ₂ is used as an etching gas, and etching is performed by RIE. In this case, the relationship between the etching area, the Cl ₂ flow rate, and the etching shape is approximately as shown in FIG. In FIG. 3, (a) shows Al on the SiO ₂ film, and (b) shows the relationship between the angle θ of the etching wall of Al, the flow rate of Cl ₂ , and the etching area. That is, as the Cl ₂ flow rate increases, the etching shape becomes an inversely tapered shape, and as the Cl ₂ flow rate decreases, the etching shape becomes a forward tapered shape. Also, the etching shape becomes reverse tapered as the etching area increases, and the etching shape becomes forward tapered as the etching area decreases. Therefore, by detecting the emission intensity of AlCl ^* radicals, determine the etching area from that value, it determines a Cl ₂ flow rate to approach the by 90 degrees can etch shape of Al.

  According to this embodiment, since the etching parameters are changed according to the etching area, the etching shape can be set to an appropriate state.

[Third Embodiment]
Hereinafter, a dry etching method according to a third embodiment of the present invention will be described with reference to the drawings.

  FIG. 7 is a flowchart showing a processing stage of a dry etching method (aluminum etching) having a multi-stage etching step according to the third embodiment of the present invention. Hereinafter, this flow will be described.

1) Breakthrough etching: Breakthrough etching is performed, and at this time, the emission intensity of a specific wavelength, for example, AlCl ^* radicals is measured.

  2) Main etching: Main etching is performed after adjusting the gas flow ratio as an etching parameter based on the emission intensity measured during the breakthrough etching of 1).

3) Overetching: 2) Overetching is performed for the etching time corresponding to the amount of overetching under the same etching conditions as the main etching. That is, over-etching is performed with parameters set based on the emission intensity of AlCl ^* radicals.

  Here, dry etching including a multi-step etching step will be described in more detail.

  For example, in the case of etching an aluminum alloy, an alumina layer is usually formed on the surface, but the etching does not proceed unless this is removed. For this reason, the etching is performed to some extent under the process conditions (plasma processing conditions) for etching the alumina whose shape has been sacrificed (this is the breakthrough etching of the first step). Next, the aluminum alloy bulk portion is etched. For the etching of this bulk portion, it is necessary to select an etching condition in consideration of the shape. These etching conditions are selected as the second step (main etching).

The etching conditions in each of the above etchings are as follows, except that some parameters such as a chlorine flow rate are changed based on the emission intensity of AlCl ^* radicals in the breakthrough etching. This is the same as the flow of 1.

Etching conditions CH ₃ Cl: 5 sccm
Cl ₂ : 45 sccm
BCl ₃ : 36sccm
N ₂ : 45 sccm
Gas pressure: 250 mTorr
High frequency power: 300W
Here, the relationship between the light emission intensity in the breakthrough etching and the etching area of the wafer is, for example, as shown in FIG. FIG. 7A shows the relationship between the light emission intensity and the wavelength, and FIG. 7B shows the relationship between the light emission intensity and the chlorine flow rate and the etching area. From the figure, it can be seen that the lower the luminous intensity, the smaller the etching area, and the higher the luminous intensity, the larger the etching area. FIG. 8 also shows the relationship between the chlorine flow rate and the etching area when CD-loss is 0. The CD-loss refers to a difference in dimension conversion, which indicates a difference between a resist dimension and an etching dimension, and a unit indicates a horizontal distance.

  FIG. 9 shows the relationship between the etching area, the etching shape (CD-loss), and the gas flow rate. That is, FIG. 9 shows the relationship between the CD-loss and the chlorine flow rate using the etching area as a parameter. From FIG. 9, when the CD-loss is set to zero, the chlorine flow rate must be reduced as the etching area increases.

  Regarding the etching shape, when looking at the angle of the etching side wall, the etching side wall angle is greater than 90 degrees when the CD-loss is on the + side, and the etching side wall angle is less than 90 degrees when the CD-loss is on the negative side. This is because the dimensions are finished as designed.

  A positive value of CD-loss indicates that the dimension after etching is larger than the resist dimension in terms of dimension, and a negative value indicates that the dimension after etching is smaller than the resist dimension. In this case, the post-etching dimension and the resist dimension match, and the product is finished as designed, which is preferable.

  In this embodiment, from the relationship between FIG. 8 and FIG. 9, the main etching conditions are such that the chlorine flow rate in the main etching is changed from the emission intensity in the breakthrough etching and the CD-loss is suppressed to zero.

  According to this embodiment, the etching parameters in the main etching and the over-etching steps are determined from one emission spectrum intensity in the break-through etching step, and the emission spectrum intensity corresponds to the etching opening area. This can be reflected in the processing conditions, and optimal etching corresponding to the etching opening area can be performed.

  Further, according to this embodiment, the over-etching time is determined according to the emission intensity of the plasma etching, and the emission intensity of the plasma etching corresponds to the etching opening area, so that the etching opening area is reflected in the etching processing conditions. Therefore, it is possible to perform optimal etching corresponding to the etching opening area.

  The etching method according to the present invention has an effect that the measured value of the resist dimension is reflected in the etching process conditions, and the optimum etching corresponding to the resist dimension can be performed. Useful as a method.

FIG. 2 is a schematic diagram showing a flow of the first embodiment in the dry etching method of the present invention. FIG. 4 is a schematic cross-sectional view showing a step of etching the SiO ₂ film. FIG. 4 is a characteristic diagram illustrating a relationship between an etching depth and an etching time according to the first embodiment. FIG. 4 is a characteristic diagram showing a relationship between an etching depth and an etching time using a resist dimension as a parameter. FIG. 3 is a characteristic diagram showing a relationship between radical light emission intensity and an etching area. (A) is a schematic diagram showing an etching shape, and (b) is a characteristic diagram showing a relationship between an angle of an etching wall and a chlorine flow rate and an etching area. FIG. 9 is a schematic diagram showing a flow of a third embodiment in the dry etching method of the present invention. (A) is a characteristic diagram showing the relationship between the emission intensity of the plasma and the wavelength, and (b) is a characteristic diagram showing the relationship between the emission intensity and the chlorine flow rate and the etching area. It is a characteristic view which shows the relationship between chlorine flow rate and CD-loss. It is the schematic which shows the flow of the conventional dry etching method. (A) is a characteristic diagram showing a temporal change of the luminous intensity of CO, and (b) is a characteristic diagram showing a temporal change of a second derivative of the luminous intensity.

Explanation of reference numerals

21 Si substrate 22 SiO ₂ film 23 Resist

Claims (3)

  An etching method characterized in that etching is performed by determining an etching parameter or an etching time based on a measured value of a resist dimension. An etching method characterized by etching a resist having a dimension measurement value d to a predetermined etching depth p based on the dimension measurement value d of the resist according to the following equation.
p = 1 / (1 / a × d + 1 / p ₀ )
Here, a is a constant, and p ₀ is an etching depth in a wide portion. An etching method characterized in that etching is performed by calculating an etching time from a measured value of a dimension of a resist based on a table showing a relationship between a dimension of the resist and an etching depth when the resist is etched using the mask as a mask.

Images