WO2020059011A1 - Substrate processing apparatus, semiconductor device manufacturing method, and program - Google Patents

Abstract

Provided is a configuration which: includes a control unit which causes a substrate processing system to execute a process recipe including a plurality of steps; and during the execution of the process recipe, for a step satisfying a predetermined collection condition, collects component data on a component to be monitored in the substrate processing system, generates a correlation curve indicating a correlation between the collected component data, compares the generated correlation curve with an initial correlation curve as a reference stored in advance to determine whether or not a difference between the correlation curve and the initial correlation curve exceeds a predetermined threshold value, and generates an alarm, when the difference exceeds the threshold value.

Description

Substrate processing apparatus, semiconductor device manufacturing method and program

The present disclosure relates to a substrate processing apparatus, a method of manufacturing a semiconductor device, and a program.

In the semiconductor manufacturing field, in order to improve the operation rate and production efficiency of equipment, information on equipment is accumulated, and the information is used to analyze equipment abnormalities and monitor equipment status. For example, Patent Literature 1 discloses a technique for identifying an abnormality factor using a plurality of monitor data in an abnormality analysis. Further, for example, Patent Literature 2 discloses a technique for displaying a plurality of monitor data and event data in an abnormality analysis.

JP 2018-078271 A WO 2017/168676

However, in the abnormality analysis, even if a plurality of data are simply used, there may be an abnormality that is difficult to grasp unless the correlation of each data changes over time.

The present disclosure has an object to provide a configuration capable of preventing defective production of a substrate due to a temporal change in a correlation of a plurality of data and improving a production yield.

According to one aspect of the present disclosure,
A configuration that includes a control unit that operates a substrate processing system by executing a process recipe including a plurality of steps,
The control unit includes:
During the execution of the process recipe, for a step that satisfies a predetermined collection condition, collects component data on a component to be monitored in the substrate processing system,
Generate a correlation curve showing the correlation of the collected component data,
Comparing the generated correlation curve with an initial correlation curve serving as a reference stored in advance, to determine whether the difference between the correlation curve and the initial correlation curve exceeds a predetermined threshold,
A configuration is provided for generating an alarm when the threshold is exceeded.

According to the present disclosure, it is possible to provide a technique capable of preventing defective production of a substrate due to a temporal change in a correlation of a plurality of data and improving a production yield.

It is a perspective view showing the substrate processing device used suitably for one embodiment. FIG. 1 is a side sectional view showing a substrate processing apparatus suitably used in one embodiment. FIG. 3 is a diagram illustrating a functional configuration of a control unit suitably used in one embodiment. FIG. 3 is a diagram illustrating a functional configuration of a main controller suitably used in one embodiment. FIG. 1 is a diagram illustrating a configuration example of a substrate processing system including a component to be monitored, which is preferably used in one embodiment. FIG. 7 is an explanatory diagram showing a specific example illustrating a component to be collected, a component data collection condition, and a method of calculating component data for generating correlation data in each step of a process recipe performed in an embodiment. is there. FIG. 4 is an explanatory diagram illustrating a specific example of a correlation curve generated in one embodiment. FIG. 4 is an explanatory diagram illustrating a specific example of a screen displayed in one embodiment. It is an illustration example of a cause judgment table for each combination pattern of each sensor information used in one embodiment. FIG. 11 is a diagram illustrating a configuration example of a substrate processing system including a component to be monitored, which is suitably used in a modification of the embodiment. It is an illustration example of a cause judgment table for each combination pattern of each sensor information used in a modification of one embodiment.

<One embodiment>
Hereinafter, an embodiment of the present disclosure will be described with reference to FIGS. 1 to 9.

(1) Configuration of Substrate Processing Apparatus First, a configuration example of a substrate processing apparatus according to an embodiment will be described with reference to the drawings. However, in the following description, the same components will be denoted by the same reference symbols, and repeated description may be omitted. In addition, in order to make the description clearer, the width, thickness, shape, and the like of each part may be schematically illustrated in comparison with an actual embodiment, but this is merely an example, and the interpretation of the present invention is not described. It is not limited.

(Overview of substrate processing equipment)
As shown in FIGS. 1 and 2, a substrate processing apparatus (hereinafter, also simply referred to as an apparatus) 1 to which the present disclosure is applied includes a housing 2, and a lower part of a front wall 3 of the housing 2 can be maintained. (Front maintenance opening) 4 is provided, and the opening 4 is opened and closed by a front maintenance door 5.

A pod loading / unloading port 6 is opened on the front wall 3 of the casing 2 so as to communicate between the inside and the outside of the casing 2, and the pod loading / unloading port 6 is opened and closed by a front shutter 7. A load port 8 is installed on the front side, and the load port 8 is configured to position the mounted pod 9. The pod 9 is a hermetically sealed substrate transfer container, which is carried into and out of the load port 8 by an in-process transfer device (not shown).

A rotatable pod shelf 11 is provided at an upper portion in a substantially central portion in the front-rear direction in the housing 2, and the rotatable pod shelf 11 is configured to store a plurality of pods 9. . The rotary pod shelf 11 includes a column 12 that is vertically erected and is intermittently rotated, and a plurality of stages of shelves 13 radially supported by the column 12 at respective positions of upper, middle, and lower stages. The shelf 13 is configured to store a plurality of pods 9 in a state of being placed thereon. A pod opener 14 is provided below the rotary pod shelf 11, and the pod opener 14 has a configuration on which the pod 9 can be placed and a lid of the pod 9 can be opened and closed.

A pod transport mechanism 15 is provided between the load port 8 and the rotary pod shelf 11 and the pod opener 14. The pod transport mechanism 15 can hold the pod 9 and can move up and down, and can move forward and backward in the horizontal direction. The pod 9 is transported between the load port 8, the rotary pod shelf 11, and the pod opener 14.

サ ブ A sub-housing 16 is provided at a lower portion in a substantially central portion in the front-rear direction in the housing 2 over the rear end. A pair of wafer loading / unloading ports 19 for loading / unloading a wafer (hereinafter, also referred to as a substrate) 18 into / from the sub-casing 16 is vertically arranged on the front wall 17 of the sub-casing 16 in two vertical stages. The pod openers 14 are respectively provided for the upper and lower wafer loading / unloading ports 19.

The pod opener 14 includes a mounting table 21 on which the pod 9 is mounted, and an opening / closing mechanism 22 for opening and closing the lid of the pod 9. The pod opener 14 is configured to open and close a wafer entrance of the pod 9 by opening and closing a lid of the pod 9 mounted on the mounting table 21 by an opening and closing mechanism 22.

The sub-housing 16 constitutes a transfer chamber 23 which is airtight from a space (pod transfer space) in which the pod transfer mechanism 15 and the rotary pod shelf 11 are provided. A wafer transfer mechanism (substrate transfer mechanism) 24 is installed in the front area of the transfer chamber 23. The substrate transfer mechanism 24 has a required number (five in the drawing) of the substrates 18 to be mounted. A wafer mounting plate 25 is provided. The wafer mounting plate 25 can be moved directly in the horizontal direction, can be rotated in the horizontal direction, and can be moved up and down. The substrate transfer mechanism 24 is configured to load and unload the substrate 18 to and from the boat (substrate holder) 26.

(4) In the rear area of the transfer chamber 23, a standby unit 27 that accommodates and stands by the boat 26 is configured. Above the standby unit 27, a vertical processing furnace 28 is provided. The processing furnace 28 has a processing chamber (reaction chamber) 29 formed therein. The lower end of the processing chamber 29 is a furnace port, and the furnace port is opened and closed by a furnace port shutter 31. ing.

ボ ー ト A boat elevator 32 as an elevating mechanism for elevating the boat 26 is installed between the right end of the housing 2 and the right end of the standby section 27 of the sub-housing 16. A seal cap 34 as a cover is horizontally mounted on an arm 33 connected to the elevator of the boat elevator 32. The cover 34 vertically supports the boat 26, and transfers the boat 26 to the processing chamber 29. The furnace port can be hermetically closed in a state where the furnace is charged.

The boat 26 is configured so that a plurality of (for example, about 50 to 125) substrates 18 are aligned in the center thereof and held in multiple stages in a horizontal posture.

A clean unit 35 is disposed at a position facing the boat elevator 32 side. The clean unit 35 is configured by a supply fan and a dustproof filter for supplying a clean atmosphere or clean air 36 that is an inert gas. I have. Between the substrate transfer mechanism 24 and the clean unit 35, a notch aligning device (not shown) as a substrate aligning device for aligning the circumferential position of the substrate 18 is provided.

The clean air 36 blown from the clean unit 35 flows through the notch alignment device (not shown), the substrate transfer mechanism 24, and the boat 26, and is then sucked by a duct (not shown) and exhausted to the outside of the housing 2. Or is blown into the transfer chamber 23 by the clean unit 35.

Next, the operation of the device 1 will be described.
When the pod 9 is supplied to the load port 8, the pod loading / unloading port 6 is opened by the front shutter 7. The pod 9 on the load port 8 is carried into the housing 2 through the pod carry-in / out port 6 by the pod transport device 15 and is placed on the designated shelf 13 of the rotary pod shelf 11. After the pod 9 is temporarily stored on the rotary pod shelf 11, the pod 9 is transferred from the shelf 13 to one of the pod openers 14 by the pod transfer device 15 and transferred to the mounting table 21, or 8 and transferred directly to the mounting table 21.

At this time, the wafer loading / unloading port 19 is closed by the opening / closing mechanism 22, and the transfer chamber 23 is filled with the clean air 36 flowing therethrough. Since the transfer chamber 23 is filled with nitrogen gas as clean air 36, the oxygen concentration in the transfer chamber 23 is lower than the oxygen concentration inside the housing 2.

The pod 9 placed on the mounting table 21 has its opening-side end face pressed against the edge of the opening of the wafer loading / unloading port 19 on the front wall 17 of the sub-housing 16, and the lid is removed by the opening / closing mechanism 22. , The wafer entrance is opened.

When the pod 9 is opened by the pod opener 14, the substrate 18 is taken out of the pod 9 by the substrate transfer mechanism 24, transferred to a notch aligning device (not shown), and aligned with the notch aligning device. Thereafter, the substrate transfer mechanism 24 carries the substrate 18 into the standby section 27 located behind the transfer chamber 23 and charges (charges) the boat 26.

(4) The substrate transfer mechanism 24 that has transferred the substrate 18 to the boat 26 returns to the pod 9, and loads the next substrate 18 into the boat 26. While the substrate transfer mechanism 24 in one (upper or lower) pod opener 14 is loading the substrate 18 into the boat 26, the other (lower or upper) pod opener 14 is separated from the rotary pod shelf 11 by another. The pod 9 is transported and transferred by the pod transport device 15, and the opening of the pod 9 by the other pod opener 14 proceeds simultaneously.

(4) When a predetermined number of substrates 18 are loaded into the boat 26, the furnace port of the processing furnace 28 that has been closed by the furnace port shutter 31 is opened by the furnace port shutter 31. Subsequently, the boat 26 is lifted by the boat elevator 32 and is loaded (loaded) into the processing chamber 29.

After loading, the furnace opening is hermetically closed by the seal cap 34. In this embodiment, a purge step (pre-purge step) in which the processing chamber 29 is replaced with an inert gas at this timing (after loading) is provided.

(4) The processing chamber 29 is evacuated to a desired pressure (degree of vacuum) by a gas exhaust mechanism (not shown) such as a vacuum pump. Further, the processing chamber 29 is heated to a predetermined temperature by a heater driving unit (not shown) so as to have a desired temperature distribution. In addition, a processing gas controlled at a predetermined flow rate is supplied by a gas supply mechanism (not shown), and the processing gas contacts the surface of the substrate 18 in the process of flowing through the processing chamber 29, and A predetermined process is performed. Further, the processing gas after the reaction is exhausted from the processing chamber 29 by the gas exhaust mechanism.

When a preset processing time has elapsed, an inert gas is supplied from an inert gas supply source (not shown) by a gas supply mechanism, and the processing chamber 29 is replaced with the inert gas. Is returned to normal pressure (after-purge step). Then, the boat 26 is lowered by the boat elevator 32 via the seal cap 34.

(4) Regarding unloading of the substrate 18 after the processing, the substrate 18 and the pod 9 are discharged to the outside of the housing 2 in a procedure reverse to the above description. The unprocessed substrate 18 is further loaded into the boat 26, and the processing of the substrate 18 is repeated.

(Functional configuration of control unit)
Next, a functional configuration of a control unit (control system) 200 centering on a main controller 201 as an operation unit will be described with reference to FIG.
As shown in FIG. 3, the control unit 200 includes a main controller 201, a transport controller 211 as a transport controller, a process controller 212 as a processing controller, an apparatus management controller 215 as a data monitoring unit, It has. The device management controller 215 functions as a data collection controller, collects device data inside and outside the device 1, and monitors the soundness of the device data DD inside the device 1. In the present embodiment, the control unit 200 is housed in the device 1. Further, the transport system controller 211, the process system controller 212, and the device management controller 215 have the same configuration as the main controller 201.

Here, the apparatus data DD refers to data (hereinafter, also referred to as control parameters) relating to substrate processing such as a processing temperature, a processing pressure, and a flow rate of a processing gas when the apparatus 1 processes the substrate 18, and the quality of a manufactured product substrate. (E.g., the film thickness of the formed film and the accumulated value of the film thickness) and the component data (e.g., the quartz reaction tube, heater, valve, mass flow controller (hereinafter, MFC), etc.) of the apparatus 1 , Set values, measured values, etc., which are data generated by operating each component when the apparatus 1 processes the substrate 18.

The main controller 201 is electrically connected to the transport controller 211 and the process controller 212 by a LAN line LAN1 such as 100BASE-T, for example, so that transmission / reception of each device data DD and download / upload of each file can be performed. It has a possible configuration.

(4) An external host computer 300 and management device 310 are connected to the main controller 201 via a communication network LAN2 such as 100BASE-T. For this reason, even when the device 1 is installed in a clean room, the host computer 300 and the management device 310 can be arranged in an office or the like outside the clean room.

The device management controller 215 is connected to the main controller 201 via a LAN line, is configured to collect device data DD from the main controller 201, quantify the operation state of the device, and display it on a screen. The device management controller 215 will be described later in detail.

The transport system controller 211 is connected to a substrate transport system 211A mainly including the rotary pod shelf 11, the boat elevator 32, the pod transport device 15, the substrate transfer mechanism 24, the boat 26, and a rotation mechanism (not shown). ing. The transport system controller 211 is configured to control transport operations of the rotary pod shelf 11, the boat elevator 32, the pod transport device 15, the substrate transfer mechanism 24, the boat 26, and a rotation mechanism (not shown). .

The process controller 212 includes a temperature controller 212a, a pressure controller 212b, a gas flow controller 212c, and a sequencer 212d. The temperature controller 212a, the pressure controller 212b, the gas flow controller 212c, and the sequencer 212d constitute a sub-controller and are electrically connected to the process system controller 212. Uploading is possible.

A heating mechanism 212A mainly composed of a heater, a temperature sensor, and the like is connected to the temperature controller 212a. The temperature controller 212a is configured to control the temperature inside the processing furnace 28 by controlling the temperature of the heater of the processing furnace 28. The temperature controller 212a is configured to perform switching (on / off) control of the thyristor, and to control electric power supplied to the heater element wire.
A gas exhaust mechanism 212B mainly composed of a pressure sensor, an APC valve as a pressure valve, and a vacuum pump is connected to the pressure controller 212b. The pressure controller 212b controls the opening degree of the APC valve and the switching (on / off) of the vacuum pump based on the pressure value detected by the pressure sensor so that the pressure in the processing chamber 29 becomes a desired pressure at a desired timing. It is configured to control.
The gas flow controller 212c is configured by the MFC 212c.
The sequencer 212d is configured to control the supply and stop of the gas from the processing gas supply pipe and the purge gas supply pipe by opening and closing the valve 212D.

The process controller 212 having such a configuration is configured to control the MFC 212c and the valve 212D so that the flow rate of the gas supplied to the processing chamber 29 becomes a desired flow rate at a desired timing.

Note that the main controller 201, the transport system controller 211, the process system controller 212, and the device management controller 215 according to the present embodiment can be realized using an ordinary computer system without using a dedicated system. For example, by installing the program from a recording medium (such as a USB key) that stores the program for executing the above-described process in a general-purpose computer, each controller that executes a predetermined process can be configured.

Then, each controller including the main controller 201, the transport system controller 211, the process system controller 212, the device management controller 215, and the like starts the provided program and executes it under the control of the OS in the same manner as other application programs. Thereby, a predetermined process can be executed.

(Configuration of main controller)
Next, the configuration of the main controller 201 will be described with reference to FIG.
The main controller 201 includes an operation display unit 227 including a main controller control unit 220, a hard disk 222 as a main control storage unit, a display unit for displaying various information, and an input unit for receiving various instructions from an operator. It is configured to include a transmission / reception module 228 as a main controller communication unit that communicates with inside and outside. The main controller 220 includes a CPU (Central Processing Unit) 224 and a memory (RAM, ROM, etc.) 226 as a temporary storage unit, and is configured as a computer having a clock function (not shown).

The hard disk 222 includes recipe files such as recipes in which processing conditions and processing procedures of the substrate are defined, a control program file for executing these recipe files, a parameter file in which parameters for executing the recipes are defined, Further, in addition to an error processing program file and an error processing parameter file, various screen files including an input screen for inputting process parameters, various icon files, and the like (all not shown) are stored.

In addition, on the operation screen of the operation display unit 227, each operation as an input unit for inputting operation instructions to the substrate transfer system 211A, the heating mechanism 212A, the gas exhaust mechanism 212B, and the gas supply system 212C shown in FIG. Buttons can also be provided.

The operation display unit 227 is configured to display an operation screen for operating the device 1. The operation display unit 227 displays information based on device data DD generated in the device 1 via the operation screen on the operation screen. The operation screen of the operation display unit 227 is, for example, a touch panel using liquid crystal. The operation display unit 227 receives input data (input instruction) of the operator from the operation screen and transmits the input data to the main controller 201. Further, the operation display unit 227 provides an instruction to execute an arbitrary substrate processing recipe (hereinafter, also referred to as a process recipe) among a plurality of recipes stored in the hard disk 222 or a recipe developed in the memory (RAM) 226 or the like. Control instruction) and transmits it to the main controller 220.

In the present embodiment, when the device management controller 215 starts up, by executing various programs and the like, each stored screen file and data table are expanded, and by reading the device data DD, the operating state of the device is read. Are configured to be displayed on the operation display unit 227.

A switching hub or the like is connected to the main controller communication unit 228, and the main controller 201 communicates with the external computer 300 and other controllers (211 212, and 215) in the apparatus 1 and the like via a network. Is configured to perform transmission and reception.

The main controller 201 also transmits device data DD such as the status of the device 1 to an external host computer 300, for example, a host computer via a network (not shown). The substrate processing operation of the apparatus 1 is controlled by the control unit 200 based on each recipe file, each parameter file, and the like stored in the main controller storage unit 222.

(2) Procedure of Substrate Processing Method Next, a substrate processing method having a predetermined processing step, which is performed using the apparatus 1 according to the present embodiment, will be described. Here, the case where the predetermined processing step is a substrate processing step, which is one step of a semiconductor device manufacturing process, will be described as an example.

(4) In performing the substrate processing process, the process recipe is developed in a memory such as a RAM in the process controller 212, for example. Then, an operation instruction is given from the main controller 201 to the process controller 212 and the transport controller 211 as needed. The substrate processing step performed in this manner includes at least a loading step, a film forming step, and a discharging step.

(Transfer process)
The main controller 201 issues an instruction to drive the substrate transfer mechanism 24 to the transport system controller 211. Then, while following instructions from the transport system controller 211, the substrate transfer mechanism 24 starts the transfer processing of the substrate 18 from the pod 9 on the mounting table 21 to the boat 26. This transfer processing is performed until the loading (wafer charging) of all the planned substrates 18 into the boat 26 is completed.

(Loading process)
When a predetermined number of substrates 18 are loaded in the boat 26, the boat 26 is lifted by the boat elevator 32 that operates according to an instruction from the transfer system controller 211, and is loaded into the processing chamber 29 formed in the processing furnace 28. (Boat road). When the boat 26 is completely loaded, the seal cap 34 of the boat elevator 32 hermetically closes the lower end of the manifold of the processing furnace 28.

(Deposition process)
Next, the inside of the processing chamber 29 is evacuated by a vacuum exhaust device such as a vacuum pump so as to have a predetermined film forming pressure (degree of vacuum) while following instructions from the pressure control unit 212b. Further, the inside of the processing chamber 29 is heated by a heater to a predetermined temperature while following an instruction from the temperature control unit 212a. Subsequently, the rotation of the boat 26 and the substrate 18 by the rotation mechanism is started while following instructions from the transport system controller 211. Then, while maintaining a predetermined pressure and a predetermined temperature, a predetermined gas (processing gas) is supplied to the plurality of substrates 18 held by the boat 26 to perform predetermined processing (for example, film formation) on the substrates 18. Processing) is performed. Before the next unloading step, the temperature may be lowered from the processing temperature (predetermined temperature).

(Unloading process)
When the film forming process on the substrate 18 placed on the boat 26 is completed, the rotation of the boat 26 and the substrate 18 by the rotation mechanism is stopped, and the seal cap 34 is Is lowered to open the lower end of the manifold, and the boat 26 holding the processed substrate 18 is carried out of the processing furnace 28 (boat unloading).

(Recovery process)
Then, the boat 26 holding the processed substrate 18 is cooled very effectively by the clean air 36 blown out from the clean unit 35. When the substrate 18 is cooled to, for example, 150 ° C. or lower, the processed substrate 18 is removed from the boat 26 (wafer discharge) and transferred to the pod 9, and then the new unprocessed substrate 18 is transferred to the boat 26. Is performed.

(3) Monitoring Process of Apparatus State Next, regarding the control processing performed by the control unit 200 in the course of performing the substrate processing step, here, the control of the apparatus state performed by the apparatus management controller 215 of the control unit 200 when performing the film forming step A specific description will be given using a monitoring process as an example.

In the film forming process, as shown in FIG. 5, a plurality of types of processing gases N2-1, N2-2, and N2-3 are adjusted in a flow rate by a corresponding gas flow rate controller (MFC) 212c. The substrate 18 is supplied to the processing chamber (reaction chamber) 29 into which the substrate 18 has been loaded at the timing set for each. Examples of the plurality of types of processing gases N2-1, N2-2, and N2-3 include a first element-containing gas that is a source gas, a second element-containing gas that is a reactive gas or a reforming gas, and a gas that acts as a purge gas. There are active gases and the like. Further, the gas is exhausted from the processing chamber 29 by an APC valve (hereinafter simply referred to as a valve) 212B-1 and a vacuum pump (hereinafter simply referred to as a pump) 212B-2 of the gas exhaust mechanism 212B. The pressure inside is adjusted. The pressure in the reaction chamber 29 is detected by the pressure sensor PG1.

That is, the substrate 18 carried into the reaction chamber 29 is processed by a substrate processing system including at least the reaction chamber 29, the MFC 212c, the valve 212B-1, the pump 212B-2, the pressure sensor PG1, and the like. Then, in the film forming process, the process controller 212 controls the MFC 212c, the valve 212B-1, and the pump 212B-2 such that the flow rates of various gases supplied to the reaction chamber 29 become desired at desired timings. I do.

At this time, the device management controller 215 can function as a data collection controller and collect device data DD inside and outside the device 1. More specifically, the device management controller 215 includes, as device data DD, at least data on the total actual flow rate (unit: slm) of various gases after flow rate adjustment in each MFC 212c in order to monitor the operation state of each MFC 212c. From each MFC 212c, and from the pressure sensor PG1, data on the actual pressure (unit: Pa) of the reaction chamber 29 in order to monitor the operation state of the valve 212B-1 and the pump 212B-2. Is possible.

That is, at least one component on the gas supply side of the reaction chamber 29 and one component on the gas exhaust side of the reaction chamber 29 are each selected as one or more components to be monitored. Specifically, each MFC 212c, which is a component on the gas supply side, and a pressure sensor PG1, which is directly affected by the operation state of the valve 212B-1 and the pump 212B-2, which are components on the gas exhaust side, are monitored. Selected as a part. Then, the device management controller 215 collects data on the total actual flow rate obtained by each MFC 212c and data on the actual pressure obtained by the pressure sensor PG1 as component data on the component to be monitored. I have.

However, collection of component data by the device management controller 215 may be performed as needed. In the present embodiment, as described below, during execution of a process recipe, component data is collected only for steps that satisfy a predetermined collection condition.

Specifically, as shown in FIG. 6, a process recipe in which a procedure, conditions, and the like when executing a substrate processing step including a film forming step is formed of a plurality of steps (recipes in the figure) See step #). In this case, the device management controller 215 collects component data only for steps that satisfy predetermined collection conditions.

The predetermined collection conditions include, for example, the processing time of each step constituting the process recipe to be executed, the open / closed state of the valve 212B-1, and the operating state of the pump 212B-2. More specifically, for example, the device management controller 215 determines that the processing time of the step is equal to or longer than a predetermined time (5 seconds in the present embodiment) (see Time [sec] in the figure), and the valve 212B-1 is opened (Open). ) State (see Valve in the figure) and the pump 212B-2 is in the operation (ON) state (see Pump in the figure), and the total data obtained by each MFC 212c for each step as part data is shown. Data on the actual flow rate and data on the actual pressure obtained by the pressure sensor PG1 are collected (see a bold frame in the figure).

After collecting the component data of each step satisfying the collection condition in this way, the device management controller 215 subsequently generates a correlation curve indicating the correlation between the collected component data as shown in FIG. . Here, in the present specification, the correlation curve indicates a relational expression indicating a relationship between the component data on the input side of the reaction chamber 29 and the component data on the output side of the reaction chamber 29. In particular, in the present embodiment, a relational expression between the actual gas flow rate controlled by each MFC 212c supplied to the reaction chamber 29 and the pressure sensor PG1 for detecting the pressure in the reaction chamber 29 is shown. The correlation curve shown in FIG. 7 is configured by plotting the actual measurement value of the pressure sensor PG1 on the vertical axis and the actual flow rate of the gas supplied to the reaction chamber 29 on the horizontal axis. That is, the apparatus management controller 215 determines the total actual value of each MFC 212c collected in each step satisfying the collection condition in a coordinate space in which the measured value of the pressure sensor PG1 is the vertical axis and the actual gas flow rate to the reaction chamber 29 is the horizontal axis. The correlation curve is generated by plotting the data on the flow rate and the data on the actual pressure by the pressure sensor PG1 in association with each other. Also, depending on the process recipe, a large amount of data may be collected, and simply plotting the data may not be enough to draw a correlation curve. In such a case, the device management controller 215 may be configured in advance to generate a correlation curve (relational expression) by obtaining an approximate curve using, for example, the least square method.
Then, by referring to the correlation curve generated in this way, the device management controller 215 determines, for each step, the total data by each MFC 212c with respect to the data on the actual pressure by the pressure sensor PG1 (ie, the actual measurement value of the pressure sensor PG1). Data on the actual flow rate (that is, the actual gas flow rate to the reaction chamber 29) can be calculated.

The correlation curve is generated each time the process recipe is executed. After generating the correlation curve, the device management controller 215 then compares the generated correlation curve with an initial correlation curve stored in advance as a reference. The comparison with the initial correlation curve is performed each time the process recipe is executed.

Here, the initial correlation curve is a correlation curve serving as a reference for judging abnormality of the generated correlation curve (presence or absence of a change in the correlation curve). The initial correlation curve indicates a state in which the substrate processing unit including the reaction chamber 29 and the like is exhibiting a predetermined film forming performance (that is, in a reference batch, for example, the substrate processing unit exhibits the film forming performance without any problem. (Corresponding state). It is assumed that the initial correlation curve is stored in advance in a storage unit accessible by the device management controller 215 (for example, the main control storage unit 222 of the main controller 201).

{Circle around (7)} Then, the device management controller 215 determines whether or not the difference between the correlation curve and the initial correlation curve exceeds a predetermined threshold. Specifically, for example, for the pressure in the reaction chamber 29 at an arbitrary flow rate, a difference between a value calculated from the correlation curve and a value calculated from the initial correlation curve is obtained, and the difference is set to a predetermined threshold. It is determined whether or not it has exceeded. At this time, the arbitrary flow rate does not necessarily have to be at one point, but may be at a plurality of points. In that case, the differences at the respective points are obtained, and it is determined whether or not the total value of the respective differences exceeds a predetermined threshold. It is assumed that the threshold value on which the determination is based is stored in advance in the storage unit 222 accessible by the device management controller 215, similarly to the initial correlation curve.

(4) As a result of the determination, when the difference between the correlation curve and the initial correlation curve exceeds the threshold, the device management controller 215 requests the main controller 201 to generate an alarm. In response to this, the main controller 201 outputs an alarm on the screen of the operation display unit 227, or outputs an alarm to the external computer 300 via the network. Note that the device management controller 215 determines that the correlation curve is normal when the difference between the correlation curve and the initial correlation curve falls within the threshold value.

The specific mode of the alarm output will be described with reference to FIG. According to FIG. 8, which component has an error can be color-coded and displayed in the device schematic diagram.

As shown in FIG. 8, each component including a component to be monitored is displayed on the screen as a schematic configuration diagram (apparatus schematic diagram), and the component to be monitored is displayed in a list format. It is displayed on the screen by a table table (part management table or the like). Then, on the display screen, the location determined to be the cause of the change of the correlation curve can be distinguished from other locations by, for example, changing the display color.

Specifically, for example, when the MFC zero point voltage is deviated or the MFC flow rate deviation is abnormal, the display color of the MFC design or the corresponding column is changed to a predetermined color (for example, error display). Change the color to yellow) so that it can be distinguished from other parts. Further, for example, when a pump abnormality has occurred, the display color of the design of the pump or the corresponding column is changed to a predetermined color (for example, yellow which is an error display color) so that it can be distinguished from other parts. I do.

Note that the specific mode of the alarm output is not limited to the example given here, but may be another mode as long as it is a preset mode. For example, instead of changing the display color as described above, a predetermined symbol (for example, an! Mark) that calls attention may be displayed together.

Further, as shown in FIG. 8, when an icon (for example,! Mark) indicating a place where the display has been changed (an alarm occurrence place) is touched, the screen is changed to an abnormality history information display screen of the place. .

行 う By performing such an alarm output, the operator of the apparatus 1 can recognize a portion requiring repair or maintenance. For example, even when repair or maintenance is performed as a countermeasure against defective production of the substrate 18 that may occur with aging, downtime of the apparatus 1 can be reduced as much as possible.

Further, by performing such an alarm output, the operator of the apparatus 1 has a temporal change in the correlation between the plurality of component data as compared with the case defined by the initial correlation curve. You can recognize that. Therefore, for example, even if a situation occurs in which defective production of the substrate 18 may occur due to a change over time, this can be determined and recognized each time the process recipe is executed. It is possible to improve the production yield by preventing it before it happens.

(4) Alarm Cause Determination Process Next, in the above-described device state monitoring process, when it is determined that an alarm output is to be performed, the alarm cause determination process performed by the device management controller 215 will be specifically described.

え ば By performing the above-described alarm output, the operator of the apparatus 1 can recognize the temporal change of the correlation curve. However, it is difficult to specify the cause of the change in the correlation curve only by the alarm output. Therefore, in the present embodiment, in addition to the alarm output, a determination process for specifying the cause of the change in the correlation curve is performed.

Specifically, for the alarm cause determination process, a cause determination table for each combination pattern of the sensor information (hereinafter, also simply referred to as a cause determination table) is prepared in advance. As shown in FIG. 9, the cause determination table is configured by a table in which a combination pattern of error items for a component to be monitored is defined, and is stored in the storage unit 222 accessible by the device management controller 215. Shall be

In such a cause determination table, an error item relating to the gas supply side component of the reaction chamber 29 and an error item relating to the gas exhaust side component of the reaction chamber 29 are included as error items for the monitored component. Items are provided. More specifically, the error items related to the components on the gas supply side include, for example, the zero point voltage of each MFC 212c and the deviation between the set flow rate and the actual flow rate in each MFC 212c. Further, as an error item related to the components on the gas exhaust side of the reaction chamber 29, for example, there is a leak rate of the reaction chamber 29 that can be obtained from the detection result by the pressure sensor PG1. Then, a cause determination table is configured by a combination of the zero point voltage of each MFC 212c, the deviation between the set flow rate and the actual flow rate in each MFC 212c, and the leak rate of the reaction chamber 29.

When the difference between the correlation curve and the initial correlation curve exceeds the threshold, the device management controller 215 generates an alarm as described above, while generating an error item for each monitored component (for example, there is a change / None). Each error item can be confirmed by monitoring sensor information from the corresponding monitoring target component. Then, the result of the check is compared with a corresponding combination pattern in a cause determination table (table), and a part to be monitored that has an abnormality is identified, thereby performing a determination process on the cause of the alarm. I have.

Specifically, for example, as shown in Case 1 in FIG. 9, when there is an abnormality only in the MFC zero point voltage and there is no other change, the device management controller 215 determines the cause of the change in the correlation curve (described above). Can be concluded as a change in the actual flow rate of the supplied gas due to a change in the MFC zero point voltage. Similarly, for example, as shown in Case 2 in FIG. 9, when there is an abnormality only in the MFC deviation and no other change is observed, the device management controller 215 determines that the cause of the change (alarm) in the correlation curve is And the change in the actual flow rate of the supply gas due to the MFC failure. Further, for example, as shown in Case 3 in FIG. 9, when there is only an abnormality in the leak rate and no other abnormality is found, the device management controller 215 determines that the correlation curve change (alarm) is caused in the furnace. It can be concluded that the leak amount has changed. In addition, for example, as shown in Case 4 in FIG. 9, when no abnormality is found in any of the cases, the device management controller 215 determines that the cause of the change in the correlation curve is deterioration of the pump 212B-2 or by-products. It can be determined that exhaust pipe blockage occurs due to this.

Although omitted in FIG. 9, as another case (Case), when an error item of both the MFC zero point voltage and the MFC deviation occurs, the device management controller 215 determines that the MFC has failed and an abnormality has occurred. The main controller 201 is instructed to stop the MFC (transmit a stop signal). Further, when the error items of both the MFC zero point voltage and the leak rate occur, it is concluded that the cause of the alarm is both of them. The same applies to the case of the MFC deviation and the leak rate.

で は At present, the cause of the abnormality due to the combination of error items including parts on the exhaust side is unknown. Accordingly, when a cause other than the leak rate of the MFC 212c or the reaction chamber 29 is considered as in Case 4 of FIG. 9, it is determined that the pump 212B-2 is deteriorated or the exhaust pipe is blocked by a by-product. The cause determination table according to the present embodiment is an example, and an error item can be added, and an error item relating to the valve 212B-1 or the pump 212B-2 can be added in the future. When the combination pattern of the error items increases in this way, the cause can be determined by the cause determination table for any abnormality, and the recovery processing can be performed.

(4) As described above, the device management controller 215 requests the main controller 201 to report a cause determined based on the cause determination table together with the alarm output. In response to this, the main controller 201 issues a report on the screen of the operation display unit 227, or issues a report to the external computer 300 via the network.

As a specific mode of report notification, for example, a screen shown in FIG. 8 can be used. Specifically, for example, if a deviation occurs in the MFC zero point voltage, the display color of the MFC design or the corresponding column is changed to a predetermined color (for example, yellow which is an error display color), When the cause of the change of the correlation curve is found while being distinguishable from other places, the information including the cause of the zero point shift is displayed according to a touch operation on (the symbol of) the MFC in the device schematic diagram. To Further, for example, when a pump abnormality occurs, the display color of the design of the pump or the corresponding column is changed to a predetermined color (for example, yellow which is an error display color) so that it can be distinguished from other parts. On the other hand, when the cause of the change of the correlation curve is found, the information including the cause of the pump abnormality is displayed according to the touch operation on (the design of) the pump in the apparatus schematic diagram. The specific mode of report notification is not limited to the example described here, but may be another mode as long as the mode is a preset mode. For example, data may be transmitted to a computer (PC) installed in a place (for example, an office) separated from the device 1 (not shown). When the error of the component whose cause is specified (for example, the MFC or the pump) is released, the identifiable display displayed on the device outline diagram or the component management table is returned to the original. You may.

レ ポ ー ト By performing such a report notification, the operator or the like of the apparatus 1 can quickly and accurately execute repair or maintenance. Therefore, for example, even when repair or maintenance is performed as a countermeasure against defective production of the substrate 18 which may occur with aging, downtime of the apparatus 1 can be reduced as much as possible.

(5) Effects of the present embodiment According to the present embodiment, one or more effects described below can be obtained.

(A) According to the present embodiment, a correlation curve indicating a correlation between component data collected during execution of a process recipe is generated, and a difference between the correlation curve and an initial correlation curve serving as a reference is determined in advance. When the threshold value is exceeded, an alarm is generated. Therefore, it is possible to prevent defective production of the board 18 due to a temporal change in the correlation of each component data (plural data), and to improve the production yield of the board 18.

(B) In the present embodiment, collection of each component data required for generating a correlation curve is performed for a step that satisfies a predetermined collection condition. Therefore, data collection only needs to be performed for steps that are considered to have a large effect on the correlation curve, and the processing load for data collection can be reduced.

(C) According to the present embodiment, a table of cause determination tables is prepared in advance, and when a difference between the correlation curve and the initial correlation curve exceeds a threshold value, occurrence of an error item for each component to be monitored is performed. Is checked and checked against the combination pattern in the cause determination table to perform a determination process for specifying the cause of the abnormality that causes the alarm. Therefore, the cause of the abnormality (that is, the part requiring repair or maintenance) can be quickly and accurately recognized. For example, even when repair or maintenance of the part where the abnormality occurs due to aging is performed, the downtime of the apparatus 1 is reduced. Can be reduced as much as possible, and as a result, the operation rate of the apparatus can be improved.

<Modification>
Next, a modified example of the present embodiment will be described with reference to FIGS. Here, only the portions different from the above-described embodiment will be described below, and the description of the same portions will be omitted.

The modified example described here is an example in which a plurality of pressure sensors are installed at various locations, and it is possible to narrow down the exhaust pipe blockage position by a by-product.
Specifically, as shown in FIG. 10, in addition to the pressure sensor PG1 for directly measuring the actual pressure in the reaction chamber 29 as in the case of the above-described embodiment, the distance between the reaction chamber 29 and the valve 212B-1 is increased. , A pressure sensor PG3 disposed upstream between the valve 212B-1 and the pump 212B-2, and a pressure sensor PG3 disposed upstream between the valve 212B-1 and the pump 212B-2. Pressure sensor PG4. With such a configuration, in addition to the actual pressure in the reaction chamber 29, the actual pressure at various points in the exhaust pipe can be measured.

In the case where a plurality of pressure sensors PG1 to PG4 are installed in each place as described above, a corresponding cause determination table (table) for alarm cause determination processing is prepared in advance. Specifically, as shown in FIG. 11, in addition to the error items described in the above-described embodiment, the cause determination table also defines the detection results of the pressure sensors PG1 to PG4 as error items. Prepare one composed of combinations.

By using such a cause determination table, the device management controller 215 can perform the following determination process regarding the cause of the alarm.
For example, in Case 1 in FIG. 11, the change in the supply gas actual flow rate due to the change in the MFC zero point voltage, in Case 2, the change in the supply gas actual flow rate due to the MFC failure, and in Case 3, the change in the in-furnace leak rate are the correlation curve changes. Can be determined as the cause.
Further, in Case 4 in FIG. 11, it can be concluded that the occurrence of the blockage of the pipe due to the accumulation of by-products between the reaction chamber 29 and the pressure sensor PG2 is the cause of the correlation curve change. Further, in Case 5, between the pressure sensor PG2 and the pressure sensor PG3, and in Case 6, between the pressure sensor PG3 and the pressure sensor PG4, it is concluded that pipe clogging caused by by-product accumulation has caused a change in the correlation curve. be able to. Further, in Case 7, it is possible to conclude that the blockage of the pipe due to the accumulation of by-products generated between the pressure sensor PG4 and the pump 212B-2 or the deterioration of the pump 212B-2 is the cause of the correlation curve change.

As described above, according to the modified example described above, a plurality of pressure sensors PG1 to PG4 are installed at various locations, and a cause determination table corresponding to the pressure sensors PG1 to PG4 is prepared in advance. It is possible to narrow down the location of the pipe blockage due to the by-product. Therefore, downtime can be further reduced, which is very preferable in improving the operation rate of the apparatus.

<Other embodiments>
As mentioned above, although one Embodiment of this invention and its modification were specifically described, this invention is not limited to said Embodiment or modification, and can be variously changed in the range which does not deviate from the summary. . For example, the automatic management of future parts will be briefly described below.

For example, the device management controller 215 compares the operation state of each MFC, which is a component on the gas supply side, and the APC valve and the vacuum pump, which are components on the gas exhaust side, with respect to a cause determination table prepared in advance (FIG. 9). The pressure sensor PG1 that is directly affected is automatically selected as each monitored component, and the device management controller 215 creates or selects an initial correlation curve for the selected monitored component and performs the initial correlation. It is conceivable to set a threshold value for a curve and automatically set component data collection conditions for creating a correlation curve in the present embodiment.

In this way, the device management controller 215 selects a monitoring target component, collects collected component data, creates a correlation curve, and compares the correlation curve with the initial correlation curve according to the cause determination table. Thus, the component to be monitored can be automatically monitored. It is possible to select an optimal component from the components constituting the substrate processing apparatus 1 and efficiently manage necessary components.

In the above-described embodiment or modification, the substrate processing apparatus and the semiconductor device manufacturing method used in the semiconductor manufacturing process have been mainly described. However, the present invention is not limited to these. For example, a liquid crystal display (LCD) The present invention is also applicable to a substrate processing apparatus for processing a glass substrate, such as an apparatus, and a method of manufacturing the same.

In the film forming process, any method may be used as long as the liquid material is vaporized and supplied to a processing chamber (reaction chamber) 29 in a processing furnace 28 to form a film on the surface of the substrate (wafer) 18. The type of film to be formed is not particularly limited. For example, the type of film formed in the film formation step may be a film containing a silicon compound (SiN, Si, etc.) or a film containing a metal compound (W, Ti, Hf, etc.) Can also be suitably applied.

成膜 In addition, the film forming process performed in the film forming step includes, for example, a process for forming a CVD (chemical vapor deposition), a PVD (Physical Vapor Deposition), an oxide film, a nitride film, a process for forming a metal-containing film, and the like.

Further, in the above-described embodiment or the modified example, the substrate processing apparatus for performing the film forming process and the method for manufacturing the semiconductor device have been described. However, the present invention is not limited to these. For example, another substrate processing apparatus ( Exposure apparatus, lithography apparatus, coating apparatus, CVD apparatus using plasma, etc.) can also be applied.

DESCRIPTION OF SYMBOLS 1 ... Substrate processing apparatus, 18 ... Substrate (wafer), 29 ... Processing chamber (reaction chamber), 200 ... Control part, 201 ... Main controller, 212B ... Gas exhaust mechanism, 212B-1 ... APC valve, 212B-2 ... Vacuum Pump, 212c: Gas flow controller (MFC), 215: Device management controller, DD: Device data, PG1 to PG4: Pressure sensors

Claims (15)

1. A substrate processing apparatus that includes a control unit that executes a process recipe including a plurality of steps to operate a substrate processing system,
   The control unit includes:
   During the execution of the process recipe, for a step that satisfies a predetermined collection condition, collects component data on a component to be monitored in the substrate processing system,
   Generate a correlation curve showing the correlation of the collected component data,
   Comparing the generated correlation curve with an initial correlation curve serving as a reference stored in advance, to determine whether the difference between the correlation curve and the initial correlation curve exceeds a predetermined threshold,
   When the threshold is exceeded, it is configured to generate an alarm,
   Substrate processing equipment.
2. The control unit includes:
   It has a table in which a combination pattern of error items for the component to be monitored is defined,
   When the threshold value is exceeded, the occurrence of the error item is confirmed from the component data collected for each of the monitoring target components, the error item is checked against a corresponding combination pattern in the table, and the monitoring target component in which an error has occurred is detected. And a cause of the abnormality occurring in the part.
   The substrate processing apparatus according to claim 1.
3. The components to be monitored are configured such that at least one or more components on the gas supply side of the reaction chamber and components on the gas exhaust side of the reaction chamber included in the substrate processing system are selected. The substrate processing apparatus according to claim 1.
4. The components to be monitored are a flow controller and a pressure sensor,
   The substrate processing apparatus according to claim 3.
5. 2. The method according to claim 1, wherein the predetermined collection conditions include a processing time of the step, an open / close state of an exhaust valve included in the substrate processing system, and an operating state of an exhaust device included in the substrate processing system. 3. Substrate processing equipment.
6. The correlation curve is formed by plotting an actual measurement value of a pressure sensor included in the substrate processing system on a vertical axis and an actual flow rate of gas supplied to a reaction chamber included in the substrate processing system on a horizontal axis. The substrate processing apparatus according to claim 1, wherein:
7. The control unit checks whether the collection condition is satisfied for each step, and in the matched step, the actual value of the gas supplied to the reaction chamber with respect to the actual measurement value of the pressure sensor that measures the pressure of the reaction chamber. The substrate processing apparatus according to claim 1, wherein the substrate processing apparatus is configured to calculate a flow rate.
8. 2. The substrate processing apparatus according to claim 1, wherein the initial correlation curve corresponds to the correlation curve when the substrate processing unit is exhibiting a predetermined film forming performance. 3.
9. As error items for the monitored component, there are an error item related to a gas supply side component of the reaction chamber included in the substrate processing system and an error item related to a gas exhaust side component of the reaction chamber. 3. The substrate processing apparatus according to claim 2, wherein each of the substrate processing apparatuses is provided.
10. The error item for the monitored component is configured by a combination of a zero point voltage of a mass flow controller included in the substrate processing system, a deviation between a set flow rate and an actual flow rate in the mass flow controller, and a leak rate of the reaction chamber. The substrate processing apparatus according to claim 9, wherein:
11. 2. The substrate processing apparatus according to claim 1, wherein the control unit is configured to generate the correlation curve each time the process recipe is executed and to compare the correlation curve with the initial correlation curve. 3.
12. 10. The substrate processing apparatus according to claim 9, wherein the control unit is configured to generate an alarm when a sum of differences between the measured values at the plurality of arbitrary points exceeds the threshold.
13. A display unit configured to display on the screen the component to be monitored as a table table in a list format, the substrate processing system including the component to be monitored as an apparatus schematic diagram,
    The control unit is configured to, when the threshold value is exceeded, generate an alarm, and display the display unit so as to be able to discriminate between the component to be monitored that has caused the abnormality that caused the alarm and the component where the abnormality has not occurred. 2. The substrate processing apparatus according to claim 1, wherein the substrate processing apparatus is configured to display the information.
14. A method of manufacturing a semiconductor device having a substrate processing step of operating a substrate processing system by executing a process recipe including a plurality of steps,
    The substrate processing step includes:
    During the execution of the process recipe, for a step that satisfies a predetermined collection condition, a step of collecting component data on a component to be monitored in the substrate processing system,
    Generating a correlation curve indicating a correlation between the collected component data;
    Comparing the generated correlation curve and an initial correlation curve serving as a reference stored in advance, and determining whether a difference between the correlation curve and the initial correlation curve exceeds a predetermined threshold value,
    Generating an alarm when the threshold is exceeded;
    A method for manufacturing a semiconductor device having:
15. A procedure for operating a substrate processing system by executing a process recipe including a plurality of steps;
    During the execution of the process recipe, for a step that satisfies a predetermined collection condition, a procedure of collecting component data on a component to be monitored in the substrate processing system,
    A procedure for generating a correlation curve indicating a correlation between the collected component data,
    Comparing the generated correlation curve with an initial correlation curve serving as a reference stored in advance, and determining whether a difference between the correlation curve and the initial correlation curve exceeds a predetermined threshold value,
    A procedure for generating an alarm when the threshold value is exceeded;
    That causes a substrate processing apparatus to execute the above through a computer.

Images