JP2000017436A - Deposition apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To stably execute prescribed deposition by extracting parameters which are keys and objectively selecting the controlled variable thereof with a deposition apparatus which executes prescribed deposition on a substrate by controlling the plural parameters. SOLUTION: This apparatus has a memory section which stores the parameter values at the time of deposition inputted at every sampling time and the characteristic values to be inputted after the deposition, a processing section 10 which searches the characteristic values and parameter values at the time of every deposition from the memory section with respect to the combinations of the selected characteristics and parameters, stores these values, analyzes the correlative relations for the combinations and calculates the parameter value range existing within the press characteristic management values and a monitor means 9 which has a display means 12 for displaying the correlative analysis state.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a film forming apparatus,
In particular, the present invention relates to a film forming apparatus suitable for high-frequency bias sputtering in which two electrodes are provided and high-frequency power is supplied thereto to form a film.

[0002]

2. Description of the Related Art There are various methods for forming a film on a substrate, such as sputtering, ion plating and vacuum deposition. Abbreviated)
Is often determined by a process engineer. High frequency (hereinafter abbreviated as RF) bias sputtering, which can form an insulating film such as SiO ₂ or Al ₂ O _3, has good step coverage (step coverage). It is often used for forming an insulating film for insulation. However, it is difficult to obtain stable film formation quality in combination with the large number of parameters. In Japanese Patent Publication No. 7-47820, in RF bias sputtering, a phase difference between RF power on a target side and RF power on a substrate side is controlled as a film forming apparatus for stabilizing discharge and forming a thin film having a uniform thickness on a substrate. Things are disclosed. This is because the amount of change in the discharge is detected as the amount of change in the potential by a monitor sensor arranged on the electrodes of the target and the substrate. The film forming apparatus includes means for detecting a shift in the relationship and displacing the phase of the output of the RF oscillator so as to reduce the shift.

[0003]

The problem is not serious when a thin film is formed in a short time. However, when a relatively thick film is coated to cover a step for a long time, particularly when the film is formed to cover a step. If the parameters are not properly controlled, problems arise in quality and production efficiency. In general, an RF bias sputtering apparatus gives a constant discharge energy to keep the plasma generation state constant, controls the power on the target side to keep the deposition rate of the film on the substrate constant, and removes the film on the substrate. The bias voltage is controlled to keep the ion collision speed for reverse sputtering constant. However, the impedance of the discharge circuit depends on the change in the state of film adhesion to the substrate holder and the inner wall of the chamber, the degree of consumption of target material,
Also, it changes due to the difference in the number of substrates, and it is difficult to maintain a constant discharge state. For this reason, a method of detecting and controlling phase fluctuations of the two power supplies in the above-mentioned known example has been considered. However, in determining the control amount of the parameter, it is still left to the individual skill of the process engineer, and it is set based on the experience of the process engineer through the inspection after film formation. There is a problem that the standard is not constant and it takes a long time before a film of a predetermined quality can be stably formed. In particular, this problem was remarkable when the film formation target was changed or the apparatus was changed. According to the present invention, it is possible to objectively select a parameter that is a key to stably perform a predetermined film formation and set a control amount thereof by data analysis, and to display a change in a parameter value during a process. It is intended to provide a device.

[0004]

SUMMARY OF THE INVENTION The present invention relates to a film forming apparatus for performing a predetermined film formation on a substrate by controlling a plurality of parameters. A storage unit for storing the input characteristic values, and for the combination of the selected characteristic and parameter, the storage unit searches and accumulates the characteristic value and parameter value for each film formation, and stores the correlation for the combination. Analyze the relationship,
There is provided a processing unit for calculating a parameter value range within a preset characteristic management value, and a monitor unit having a display unit for displaying the correlation analysis state. Further, the monitor means can display a parameter value at the time of film formation inputted every sampling time and a preset control range of the parameter value. Based on this, the parameter value falls within the control range. Parameters can be controlled as follows. This control may be performed manually by an operator, or may be performed automatically based on a preset automatic control logic by automatically comparing an input parameter value with a preset parameter management value.

As a specific embodiment, in a film forming apparatus using high frequency bias sputtering for forming a film on a substrate by supplying a target power and a bias voltage to set values, an input signal is input for each sampling time during film formation. A storage unit for storing the target voltage and bias power to be measured, the step coverage characteristic value measured after the film formation, and searching the stored data for the step coverage characteristic value for each film formation batch, the target voltage and the bias power. And a processing unit for storing an index obtained by calculating a ratio between the target voltage and the bias power, analyzing a correlation between the step coverage characteristic value and the index, and obtaining an index value range within the characteristic value management value. It is a film forming apparatus provided with.

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention relates to a process for manufacturing an MR head by forming an element on a substrate and then coating a protective film (overcoat) of an insulating material such as SiO ₂ or Al ₂ O ₃ by sputtering. Will be described as an example. As the sputtering apparatus, an RF bias sputtering method is used in order to adhere the film well along the uneven steps formed by the elements on the substrate surface and deposit the film without forming a gap or a hole in the film. FIG. 2 schematically shows a film formation process of RF bias sputtering. In a normal sputtering method, a target material hit by the collision of ions generated by glow discharge is projected almost vertically onto a substrate. While the material does not easily adhere to the step side surface, the target material deposited on the substrate is reverse-sputtered by the RF bias voltage applied to the substrate, and a part of the target material goes around the step to form a film.

FIG. 1 shows a configuration example of an RF bias sputtering apparatus for explaining the present invention. Two opposing electrodes are provided in the vacuum chamber 1, and an alumina target 2 and a substrate 3 are attached. RF power sources 6 and 7 are connected to the respective electrodes via matching boxes 4 and 5. Further, in order to maintain an appropriate gas pressure after introducing argon which is an inert gas and oxygen which is a reactive gas, a gas introduction device and an exhaust device are provided in the vacuum chamber 1. The RF power supply 6 (hereinafter, referred to as a target power supply) connected to the target has a function of generating glow discharge between the electrodes to generate plasma and sputter the target 2. RF connected to the board side
The power supply 7 (hereinafter, referred to as a bias power supply) has a function of applying a bias for reverse sputtering of an alumina film (hereinafter, abbreviated as a film) deposited on the substrate 3.

[0008] A monitor means 9 which is a point of the film forming apparatus of the present invention is connected via a control device 8 for performing a sputtering process. The monitor means 9 is a CPU
Central processing means 10 having a processing section having a memory and a storage section having a memory;
1 and display means 12 such as a display or a printer to which the processing content is output. The monitor means 9
An input signal of a parameter value, which is a value obtained by measuring a control amount of parameters such as power, voltage of each of the RF power sources 6 and 7, gas pressure in the vacuum chamber 1, argon gas flow rate and oxygen flow rate with the control device 8; An output signal for instructing an operation amount relating to the parameter can be transmitted and received.

The function and operation of the monitoring means 9 will be described below. The monitor means 9 mainly has the following two functions. Calculation of the control range of parameter values. Parameter values of each parameter during the sputtering process,
The measured values of the film characteristics after sputtering are collected, edited, and subjected to data processing to obtain a parameter value management range in which the predetermined characteristics are within the management value range. Display of parameter values and control of parameters during sputtering. The change of the parameter value during the process and the control range obtained as described above are simultaneously displayed so that the quality of the sputtering state can be determined, and the parameter is controlled manually or automatically as necessary.

First, the calculation of the parameter value management range will be described in detail. In determining the sputtering conditions for the new film specification, it is uncertain what parameters are important for obtaining a predetermined characteristic value among a large number of parameters, and how much the control amount should be. It is important from the viewpoint of production efficiency and quality control how to narrow down the correlation between film characteristics and parameters through sputtering. Therefore, two types of parameter values, which are measured during the sputtering process and the characteristic results obtained by measuring the film on the substrate after the process, are collected as basic data and stored in a parameter file and a characteristic result file described later. To go. In addition, it is preferable to record a predetermined film property management value in a property management value file, and to modify the film property when changing the film property due to a change in a product or the like.

As shown in FIG. 3, parameter values are measured at predetermined sampling times, for example, at one minute intervals, from the start to the end of the sputtering process. As parameters, depending on the sputtering apparatus, for example, traveling wave power on the target side, reflected wave power, voltage, and traveling wave power on the substrate side,
Reflected power, voltage, gas pressure in vacuum chamber,
Argon gas flow rate, oxygen gas flow rate, etc., and parameter values are input to the monitor means 9 via the control device 8 or directly, and are stored in a parameter file set in the storage unit of the central processing means 10 as shown in FIG. Are recorded for each process elapsed time, and are displayed on a display 12 as a display means. Note that this parameter file may be created and managed for each deposition batch.

After the process is completed, characteristic data of the film is measured. Characteristics include film thickness, film formation rate (sputter rate), film stress, step coverage, and the like. By inputting these measurement results to the central processing unit 10 through the keyboard 11 which is an external input unit, the results are recorded in the characteristic result file set in the storage unit for each batch or each substrate. FIG. 6 shows an example of the characteristic result file. As shown in FIG. 4, when a combination is selected from various characteristics and parameters by the keyboard 11 based on the parameter file and the characteristic result file obtained as described above, the processing unit searches and collects the data. , And graph the correlation. Further, by incorporating a characteristic management value determined in advance by the specification of the product into the correlation graph, it is possible to determine the degree of influence of the parameter and find the control amount thereof. In order to obtain a highly reliable analysis result, it is preferable to collect as much basic data as possible, and it is desirable to perform 10 or more batches of sputtering. Hereinafter, the correlation with a parameter will be described using step coverage as an example of the characteristic.

Although it depends on the target film formation process, in the film formation as an overcoat, the unevenness is filled, and the deposited film has no voids such as pits and the surface after the film formation becomes flat. Need to be coated. Therefore, a film having a void in the film is regarded as a defect, and the step coverage is evaluated based on the state of occurrence of the defect. For this purpose, the presence or absence of pits, their positions and sizes may be observed with a microscope or the like, but the film growth angle is used as an index that can be measured more clearly and quantitatively. Hereinafter, the film growth angle will be described.

FIG. 7 is a diagram schematically showing film growth at a step portion. Clear growth interfaces of the alumina film appear at the steps, and defects often occur along the growth interfaces. The portions of the flat films a and b shown in FIG. 7 are films grown from flat bases A and B. This flat film portion is one in which sputter particles from the target are directly deposited. On the other hand, the slope film portion is initially formed not to be sputtered particles directly from the target, but to be mainly filled with reverse sputtered particles from the flat film to fill the step, and the slope film grows. Accordingly, sputter particles from the target are also deposited and formed. If the reverse sputtered particles are effectively wrapped around at the initial stage of film growth that fills the step, the slope film grows in the direction that spreads more horizontally as indicated by the arrow, so that the growth interface and the flat base B The film growth angle θ, which is an angle, shows a small value.
It is considered that the occurrence of defects is smaller when the film is formed in this manner, and the relationship between the film growth angle and the defect occurrence rate was examined.
It was found that there was a correlation that the smaller the film growth angle was, the less the occurrence of defects was.

As described above, the RF bias sputtering apparatus generally controls the power of the target side power supply and controls the voltage of the bias side power supply. First, it was set to a predetermined constant value obtained empirically, and the other degrees of vacuum, the flow rate of argon, etc. were similarly set to constant values. FIG. 5 shows an example of a parameter file in which parameter values during a process in a certain batch are measured. It can be seen that the traveling wave power of the target remains at a constant value, but the voltage fluctuates. Similarly, the bias-side voltage remains at a constant value, but the traveling-wave power varies. further,
Looking at the variation between batches by averaging the variation within a batch, it can be seen that there is a large variation between batches as shown in FIG. This is because the impedance of the discharge circuit is different between batches because there is a change in the state of film adhesion to the substrate holder or the inner wall of the chamber, the degree of consumption of the target material, the number of substrates, and the like.

The results of examining the film growth angle for each batch by measuring the characteristics after the end of sputtering show that the target power, the bias voltage, the degree of vacuum, etc. were controlled to be constant throughout the entire batch as initially set. However, the film growth angle fluctuates as shown in FIG. Therefore, it is necessary to understand which parameter influences the change in the film growth angle, and among the parameters, we will examine the relationship between the target voltage and the bias power whose values are changing. .

The relationship between the film growth angle and the target voltage will be described. First, from the film growth angle data stored in the characteristic performance file, the value of batch 1 is extracted, and the average value of all the sampling values of the target voltage data in the batch 1 parameter file is calculated. Is plotted on a graph with the target voltage as a coordinate axis. This is sequentially performed on all the batches after batch 2. Next, the plotted points are approximated to an approximate straight line, and superimposed on the film growth angle management values stored in the characteristic management value file to create a correlation graph as shown in FIG. In this figure, it can be said that there is a correlation between the film growth angle and the target voltage.
It is possible to know a desired target voltage value. This target voltage may be plotted with the above-described average value, or all data may be plotted and analyzed. In addition, it is also possible to select and display only an arbitrary point in the process elapsed time. In this case, a unique target voltage management range at an important elapsed time in the process determined by the target film formation specification can be grasped.

Similarly, FIG. 11 shows a correlation graph between the film growth angle and the bias power. Similarly, it can be seen that there is a correlation, and an appropriate bias power range can be grasped. Further, since the step coverage is considered to be better as the bias effect is larger, the bias effect is represented by an index obtained by dividing the target voltage by the bias power, and the relationship between the film growth angle and the index is examined. It can be seen that there is a clear correlation as shown in FIG.
This means that when controlling the film growth angle to a predetermined value, it is possible to properly operate both the target power supply and the bias power supply without necessarily operating the target power supply and the bias power supply independently. This indicates that any other parameters that may affect it may be controlled.

In addition, it is possible to confirm whether or not the set value is appropriate not only for the fluctuating parameter but also for a parameter set to a constant value, for example, the target power. In this case, basic data is collected from the sputter under the condition that the target power set value is varied by several types, and a correlation graph between the film growth angle and the target power is displayed in the same manner. FIG. 13 shows an example. In the case of setting at 6000 W, it can be determined that setting to 6000 W is better because there are many that fall within the characteristic management value range as compared to the case of 5000 W or 7000 W. In addition, a similar correlation graph can be created between parameters and other characteristics other than the step coverage, such as the film thickness and the film stress. Similarly, by looking at the correlation state with an arbitrary parameter value at this time, it can be determined whether the parameter is important for the characteristic. The method of setting the control range of the parameter value when starting a new process has been described above.

Next, display of each parameter value during the sputtering process and control of the parameters will be described. FIG.
The process is shown in FIG. 4. After the start of the sputtering process, each parameter value is collected at predetermined time intervals and stored in a parameter file. At the same time, a change in the parameter value accompanying the progress of the process is displayed on the screen. At this time, a plurality of parameters may be displayed on the same screen, or a specific parameter may be selected and displayed. At this time, the parameter value management range set as described above is simultaneously displayed. By looking at such a screen, the operator can empirically determine the quality of the sputter state for each characteristic, and can control predetermined parameters as necessary. It is more preferable to display an alarm or an abnormal message when the measured value is out of the management range. Further, the measured value and the management range can be automatically compared, and the parameters can be automatically controlled based on a preset automatic control logic such as PID control. For example, when the target voltage fluctuates as described above, the target power supply may be appropriately operated, or another parameter may be operated. What to operate depends on the apparatus and process conditions, and may be appropriately selected and operated.

As described above, even for a sputtering process in a batch, a good film formation can be performed by controlling the parameters by displaying the change of each parameter value. In addition, by accumulating the parameter values during running one after another and creating a correlation graph of the characteristics and parameters as described above, the accuracy of the parameter value management range obtained based on the provisional condition data gradually becomes more accurate. A parameter value management range with a high value can be obtained, and a more stable film formation can be performed by appropriately updating and setting. It is not necessary to measure the characteristics for all the batches, but it is preferable to perform the measurement at appropriate times to increase the number of data.

[0022]

As described above, the present invention has the following effects. 1) Since the correlation between the parameter value measured during the process and the characteristic after film formation can be analyzed in a free combination, parameters having a strong correlation with the predetermined characteristic can be objectively determined, and the control range can be rationally determined. You can know. 2) It is possible to quickly establish film forming process conditions with new specifications, which is effective in stabilizing production efficiency and quality. 3) Since the time-dependent changes of a large number of parameter values during the sputtering process can be displayed, the operator can visually grasp the abnormality of the process or the apparatus and can quickly respond. 4) Since the correlation is improved by collecting data during the running, the accuracy of the parameter management value increases as the batch is repeated, and a more stable film formation can be performed.

[Brief description of the drawings]

FIG. 1 is an RF bias sputtering apparatus as an example of a film forming apparatus of the present invention.

FIG. 2 is a diagram for explaining a film forming process of RF bias sputtering.

FIG. 3 is a diagram showing a procedure for collecting basic data.

FIG. 4 is an explanatory diagram when a correlation graph of parameter values and characteristics is created.

FIG. 5 is an example of a parameter file.

FIG. 6 is an example of a characteristic result file.

FIG. 7 is a schematic view showing a film growth state at a step portion.

FIG. 8 is a diagram showing an example of a relationship between a film growth angle and a defect occurrence rate.

FIG. 9 is a diagram showing an example of fluctuations in target voltage and bias power between batches.

FIG. 10 is a diagram showing an example of a relationship between a film growth angle and a target voltage.

FIG. 11 is a diagram illustrating an example of a relationship between a film growth angle and bias power.

FIG. 12 is a diagram showing an example of a relationship between a film growth angle and a target voltage / bias power.

FIG. 13 is a diagram showing an example of a relationship between a film growth angle and target power.

FIG. 14 is a diagram showing a procedure for collecting data and displaying parameter values during a film forming process.

[Explanation of symbols]

1 ... vacuum chamber, 2 ... target, 3 ... substrate,
4, 5 ... matching box, 6, 7 ... RF power supply,
8 control device, 9 monitor means, 10 central processing means, 11 external input means, 12 display means

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 4K029 BA44 BA46 BD01 CA05 DC32 DC35 EA09 5F103 AA08 BB54 BB56 DD27 LL20 NN10 RR01 RR04 RR06 RR10

Claims (4)

[Claims] In a film forming apparatus for performing a predetermined film formation on a substrate by controlling a plurality of parameters, a parameter value at the time of film formation inputted every sampling time and a characteristic value inputted after the film formation are stored. For the combination of the characteristic and the parameter selected in the storage unit, the characteristic value and the parameter value for each film formation are searched and stored from the storage unit, and the correlation for the combination is analyzed and set in advance. A film forming apparatus, comprising: a processing unit that calculates a parameter value range within the set characteristic management value; and a monitor that has a display that displays the correlation analysis state. 2. The film forming apparatus according to claim 1, wherein the monitor means can display a parameter value at the time of film formation inputted every sampling time and a preset control range of the parameter value. 3. The monitor means compares a parameter value at the time of film formation inputted at each sampling time with a preset parameter management value, and based on a preset automatic control logic, the parameter value is kept constant. The film forming apparatus according to claim 1, wherein a signal for controlling a parameter to be in a range is output. 4. In a film forming apparatus by high frequency bias sputtering for forming a film on a substrate by supplying a target power and a bias voltage to set values, a target voltage and a bias input every sampling time during film formation. Power, a storage unit for storing step coverage characteristic values measured after film formation, and a step coverage characteristic value for each film formation batch, a target voltage and a bias power, and a target voltage and bias power. A monitor that has a processing unit that accumulates an index for calculating a ratio with respect to power, analyzes a correlation between the step coverage characteristic value and the index, and obtains an index value range within the characteristic value management value. Film forming apparatus.