JPH09219347A - Method and equipment for managing semiconductor manufacturing process - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to set a management standard of each parameter to show the condition of a semiconductor manufacturing equipment to the most appropriate value by finding a correlation between the equipment condition data and the product data on the basis of the time when a semiconductor device is processed. SOLUTION: The equipment condition data 12 to show the condition of a semiconductor manufacturing equipment 11 and the measurement times are stored in a storage device 17 through a control system 15. The products 13 which have been through all the processes are evaluated for electric characteristics by an electric characteristic measuring equipment and the evaluation data of the products and the times when the lot of the products is processed are stored in the storage device 17 through the control system 15. On the basis of the times when the semiconductor devices have been processed, a correlation 16 between the equipment condition data 12 and the product data 14 is found. By this method, a management standard of each parameter to show the condition of the semiconductor devices can be set to the most appropriate value.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor process management method and a semiconductor process management apparatus.

[0002]

2. Description of the Related Art A semiconductor device is manufactured through a semiconductor process having 100 or more steps, and is manufactured using a plurality of complicated semiconductor manufacturing apparatuses. Therefore, there are many cases where the relationship between the parameter indicating the state of each manufacturing apparatus and the characteristics of the device manufactured by using each manufacturing apparatus has not been clearly obtained.

In the semiconductor process, each process must be strictly controlled so that the yield of manufactured devices is improved.

A conventional method for controlling a semiconductor process will be described below with reference to the drawings.

FIG. 4 is a schematic diagram showing the flow of management device and management data of a conventional semiconductor process. In FIG. 4, 51 is a semiconductor manufacturing apparatus, and 52 is a semiconductor manufacturing apparatus 51.
Device status data indicating the status of the device, 53 is device status data 5
2 is an apparatus state database for storing 2, 54 is an electric characteristic measuring apparatus for measuring electric characteristics of semiconductor devices processed by the semiconductor manufacturing apparatus 51, 55 is device characteristic data obtained by the electric characteristic measuring apparatus 54, and 56 is It is a device characteristic database that stores device characteristic data 55.

In the conventional management method using the semiconductor process management apparatus, the apparatus state data 52 and the device characteristic data 55 are stored in the respective databases 53 and 55, and the fluctuation of each individual data is extracted. Was there. For example, as shown in FIG. 5, a method of examining how the degree of vacuum of a certain manufacturing apparatus changes with time and repairing the manufacturing apparatus when an abnormality is found is used. Similarly, with respect to the device characteristic data 55, whether or not the value of the device characteristic is significantly different from the past value is managed on a daily basis. In addition, device characteristics may have clear standards based on past experimental results, etc.
We have used a method of investigating or taking countermeasures by assuming that an abnormality has occurred when the standard is not met.

[0007]

However, the conventional method for controlling a semiconductor process has no problem if the influence of each parameter on each characteristic of the device is clear, but most of the parameters are controlled. The correlation between them is not clear, and it is unclear whether the range to be managed is optimal. Therefore, even if each parameter is managed on the basis of a certain management standard, the characteristic of the device is deteriorated. Further, when the device characteristics fluctuate and a non-standard characteristic value is generated, it takes a huge amount of time and a high technology to investigate the cause.

The present invention solves the above-mentioned conventional problems all at once,
It is possible to set the management standard for each parameter that indicates the state of the semiconductor manufacturing equipment to an optimum value, and when a defective location occurs in the device characteristics, the equipment that caused the failure can be easily selected from among multiple equipment. The purpose is to enable easy extraction and easy analysis.

[0009]

In order to achieve the above-mentioned object, the means for solving the problems according to the invention of claim 1 is to provide a method for controlling a semiconductor process, which is a state of a semiconductor manufacturing apparatus for processing a semiconductor device at a predetermined time interval. A step of storing device state data indicating the above, a step of storing product data such as a yield and electric characteristics of semiconductor devices processed by the semiconductor manufacturing apparatus, and the apparatus based on the time when the semiconductor devices are processed. And a step of obtaining a correlation between the state data and the product data.

According to the structure of claim 1, since the correlation between the apparatus state data and the product data is obtained on the basis of the time when the semiconductor device is processed, the device state parameter among the parameters indicating the state of each manufacturing apparatus is determined. For parameters for which the causal relationship with the characteristics is not technically clear,
Since the correlation can be clarified statistically, the range to be controlled for each manufacturing apparatus can be clarified.

Further, even if a defective portion occurs in the device characteristics, the cause can be easily found by comparing the recorded product data with the device state data at the time of the defective occurrence.

According to a second aspect of the present invention, the configuration of the first aspect is added with a configuration further including a step of controlling the state of the semiconductor manufacturing apparatus based on the correlation.

The invention of claim 3 adds to the structure of claim 1 further comprising a step of analyzing a cause of failure based on the correlation when a defective portion occurs in the semiconductor device. It is a thing.

According to a fourth aspect of the present invention, there is provided a semiconductor process management apparatus, means for storing apparatus state data indicating a state of a semiconductor manufacturing apparatus that processes semiconductor devices at predetermined time intervals, and the semiconductor manufacturing apparatus. Configuration comprising means for storing product data such as yield and electrical characteristics of semiconductor devices, and means for obtaining a correlation between the device status data and the product data with reference to the time when the semiconductor device is processed It is what

According to the structure of claim 4, since the correlation between the apparatus state data and the product data is obtained on the basis of the time when the semiconductor device is processed, the parameter of the device among the parameters showing the state of each manufacturing apparatus is determined. For parameters for which the causal relationship with the characteristics is not technically clear,
Since the correlation can be clarified statistically, the range to be controlled for each manufacturing apparatus can be clarified.

Further, even if a defective portion occurs in the device characteristics, the cause can be easily found by comparing the recorded product data with the device state data at the time of the defective occurrence.

According to a fifth aspect of the present invention, in addition to the configuration of the fourth aspect, a configuration further including means for storing the correlation is added.

According to a sixth aspect of the present invention, in addition to the configuration of the fourth aspect, a configuration further including means for controlling the state of the semiconductor manufacturing apparatus based on the correlation is added.

According to a seventh aspect of the present invention, in addition to the configuration of the fourth aspect, when a defective portion is generated in the semiconductor device, a configuration is further provided for analyzing the cause of the defectiveness based on the correlation. It is a thing.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A semiconductor process management apparatus and a management method using the management apparatus according to an embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a schematic diagram of a semiconductor process management apparatus according to an embodiment of the present invention. In FIG. 1, 11
Is a semiconductor manufacturing apparatus, 12 is apparatus state data indicating the state of the semiconductor manufacturing apparatus 11, 13 is a product of each device type processed by the semiconductor manufacturing apparatus 11 in a lot unit, 1
4 is product data indicating the yield in lot units and electric characteristics such as device power consumption, operation characteristics and wiring resistance measured by the electric characteristic measuring device, and 15 is device state data based on the time when the product 13 is processed. A control system 17 for obtaining correlation data 16 showing the correlation between the product data 14 and the product data 14, analyzing the correlation data 16 and controlling the semiconductor manufacturing apparatus 11 according to the analysis result.
Is a storage device such as a database for recording the device status data 12, the product data 14, and the correlation data 16.

The operation of the semiconductor process management apparatus configured as described above will be described below. Semiconductor devices
100 or more processes are processed and completed in the semiconductor manufacturing apparatus 11 for each unit called a lot. The semiconductor manufacturing apparatus 11 includes a CVD apparatus, an electric furnace, a cleaning apparatus, a stepper, a dry etching apparatus and the like. For example, CV
The device D constantly manages and controls the device state such as the vacuum degree of the reaction chamber, the temperature of the stage, the electric power of the heater for controlling the temperature, the flow rate of the gas and the opening / closing angle of the valve for controlling the vacuum degree. . The device status data 12 obtained from these devices is used together with the measurement time in the control system 1
5 to the storage device 17.

In addition, the electric characteristics of the devices of the lots of products 13 which have been processed in all steps are evaluated by using an electric characteristic measuring device. The product data 14 includes the LSI yield in which it is determined whether or not the device performs a desired operation as a whole, the result of checking the presence or absence of wiring short circuit or disconnection, and the transistor characteristic evaluation using a specific pattern. included. This product data 14 is also sent to the storage device 17 via the control system 16 together with the time when the lot was processed.

The control system 15 pairs the device state data 12 with the product data 14 processed in that state based on the time when the device for each lot is processed by the semiconductor manufacturing apparatus 11. These pairs are grouped by data indicating the device status and output to a monitor screen, a printer, etc., so that the device status data 12 and the product data 1
It becomes possible to clearly know the correlation with 4. Therefore, it is possible to determine the standard for device management and to perform a quick analysis when the device characteristics change.

An example of the database structure stored in the storage device 17 will be described below with reference to the drawings. As shown in FIG. 2, the measurement date and time of the device state and the device state data 12
The vacuum degree of the CVD apparatus, the lot name processed on the measurement date and time of the apparatus state, and the device yield which is the product data 14 are defined as one record.

Thus, the storage device 1 such as a database
By recording the device state data 12 and the product data 14 associated with the measurement date and time of the device state in 7, it is possible to manage the device on a regular basis. Therefore, the device state data 12 and the product data 14 are recorded. Will be able to analyze quickly.

Hereinafter, the storage device 1 configured as described above
An example of the correlation between the device state data 12 and the product data 14 in 7 will be described with reference to the drawings. 3 shows that the heater power is newly added to the degree of vacuum of the CVD apparatus,
It is a figure which shows the correlation with the parameter which shows two apparatus states, a vacuum degree and heater electric power, and the device yield of a certain kind processed by this CVD apparatus. The heater power is plotted on the horizontal axis and the degree of vacuum is plotted on the vertical axis.
Are plotted by black circles. Also, each performance data 2
The yield distributions of the devices processed by the respective device states at 0 are shown using three types of ellipses with yields of 50% or more, 80% or more, and 90% or more.

According to the present embodiment, even data that is technically difficult to correlate with the device yield, such as the vacuum degree and heater power of the CVD apparatus shown in FIG. 3, is statistically processed. This will clarify the scope of management.

Although the two parameters are correlated with the yield in FIG. 3, even if one parameter is correlated with the yield, or three or more parameters are correlated with the yield. Clearly, the scope of control can be clarified.

In the semiconductor process, when a characteristic change such as a decrease in device yield occurs, it is necessary to pursue the cause of the change and take countermeasures. At that time, although the cause may be easily found, it often takes a long time to investigate the cause when the state of the device is slightly changed.

In the present embodiment, the distribution as shown in FIG. 3 is investigated for each parameter of the apparatus that was processing the device at the time when the characteristic change of the device occurs, and the device characteristic is selected from them. Since it is possible to check whether or not there is a parameter having a value in the distribution region that has an influence, it is possible to promptly determine the parameter that is the cause.

Conventionally, in this survey, after outputting the results, each parameter had to be surveyed one by one, and a huge amount of time was spent. However, by using the control system 15,
Among all parameters, it is possible to extract and output the parameters distributed outside the preset characteristic range.

As described above, by adopting the system configuration shown in FIG. 1, even if a defective portion occurs in a device which is a product, the problem can be promptly extracted.

[0034]

As described above, according to the semiconductor process management method of the first aspect of the present invention, the causal relationship with the device characteristics among the parameters indicating the state of each manufacturing apparatus is technically clarified. Even for parameters that have not been set, the correlation can be clarified statistically, and the range to be controlled for each manufacturing device can be grasped, so the control criteria for each parameter should be optimized. You can set it.

Further, even when a defective portion occurs in the device characteristics, the recorded product data can be compared with the device state data at the time of the defective occurrence, so that the cause can be easily determined.

According to the semiconductor process management method of the second aspect of the present invention, since the method further comprises the step of controlling the state of the semiconductor manufacturing apparatus based on the correlation, the management standard of each parameter for each manufacturing apparatus. Can be reliably set to the optimum value.

According to the semiconductor process management method of the third aspect of the present invention, when a defect location occurs in the semiconductor device, the method further includes a step of analyzing the cause of the defect based on the correlation, so that the method can be reliably implemented. In addition, the cause of failure can be quickly investigated.

According to the semiconductor process control apparatus of the fourth aspect of the present invention, among the parameters indicating the state of each manufacturing apparatus, even if the causal relationship with the device characteristics is not technically clear, Since the correlation is statistically clarified, the range to be controlled for each manufacturing apparatus can be grasped, and the control standard for each parameter is set to an optimum value.

Further, even when a defective portion occurs in the device characteristics, the cause can be easily found by comparing the recorded product data with the device state data at the time of the defective occurrence.

According to the semiconductor process control apparatus of the fifth aspect of the present invention, since the means for storing the correlation is further provided, it is possible to constantly control the manufacturing apparatus. Therefore, the apparatus state data and the product data can be controlled. Will be able to analyze quickly.

According to the semiconductor process management apparatus of the sixth aspect of the present invention, since the means for controlling the state of the semiconductor manufacturing apparatus based on the correlation is further provided, the management standard of each parameter in each manufacturing apparatus is ensured. Will be set to the optimum value.

According to the semiconductor process management apparatus of the seventh aspect of the present invention, when a defective portion occurs in the semiconductor device, means for analyzing the cause of the defectiveness based on the correlation is further provided, so that the device can be reliably operated. In addition, the cause of failure can be quickly investigated.

[Brief description of drawings]

FIG. 1 is a schematic diagram of a semiconductor process management apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing a data structure in a storage device of a semiconductor process management apparatus according to an embodiment of the present invention.

FIG. 3 is a diagram showing a correlation between a parameter showing two device states of a CVD device and a device yield in a semiconductor process management method according to an embodiment of the present invention.

FIG. 4 is a schematic diagram showing a flow of a conventional semiconductor process management apparatus and management data.

FIG. 5 is a diagram showing the output of parameters indicating the state of the conventional semiconductor process management apparatus.

[Explanation of symbols]

 11 Semiconductor Manufacturing Equipment 12 Equipment Status Data 13 Products 14 Product Data 15 Control System 16 Correlation Data 17 Storage Device 20 Actual Data 51 Semiconductor Manufacturing Equipment 52 Equipment Status Data 53 Equipment Status Database 54 Electrical Characteristics Measuring Equipment 55 Device Characteristics Data 56 Device Characteristics The database

Claims (7)

[Claims] 1. A step of storing apparatus state data indicating a state of a semiconductor manufacturing apparatus for processing a semiconductor device at predetermined time intervals, and product data such as yield and electrical characteristics of the semiconductor device processed by the semiconductor manufacturing apparatus. A method of managing a semiconductor process, comprising: a step of storing; and a step of obtaining a correlation between the apparatus state data and the product data based on a time when the semiconductor device is processed. 2. The method for managing a semiconductor process according to claim 1, further comprising the step of controlling a state of the semiconductor manufacturing apparatus based on the correlation. 3. The method of managing a semiconductor process according to claim 1, further comprising a step of analyzing a cause of a defect based on the correlation when a defective portion occurs in the semiconductor device. . 4. A unit for storing apparatus state data indicating a state of a semiconductor manufacturing apparatus that processes semiconductor devices at predetermined time intervals, and product data such as yield and electrical characteristics of the semiconductor devices processed by the semiconductor manufacturing apparatus. An apparatus for managing a semiconductor process, comprising: a storage unit; and a unit that obtains a correlation between the apparatus state data and the product data based on a time when the semiconductor device is processed. 5. The semiconductor process management apparatus according to claim 4, further comprising means for storing the correlation. 6. The semiconductor process management apparatus according to claim 4, further comprising means for controlling a state of the semiconductor manufacturing apparatus based on the correlation. 7. The semiconductor process management apparatus according to claim 4, further comprising means for analyzing a cause of a defect based on the correlation when a defective portion occurs in the semiconductor device. .