Classifications

• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
  • H10P74/00—Testing or measuring during manufacture or treatment of wafers, substrates or devices
  • H10P74/23—Testing or measuring during manufacture or treatment of wafers, substrates or devices characterised by multiple measurements, corrections, marking or sorting processes

Definitions

• TECHNICAL FIELD This invention relates generally to semiconductor device manufacturing, and, more particularly, to a method and apparatus for determining measurement frequency based on hardware age and usage
• a number of discrete steps are needed to produce a packaged semiconductor circuit device from raw semiconductor material
• the starting substrate is usually a slice of single crystal silicon referred to as a wafer Circuits of a particular type are fabricated together in batches of wafers called "lots" or "runs"
• the fab ⁇ cation process creates regular arrays of a circuit on the wafers of a lot During processmg, the individual wafers m a lot may go through individual processmg steps one at a time or as a batch At the completion of wafer processing, the wafers are tested to determine circuit functionality Later the wafers are sliced, the functioning devices are packaged, and further testing occurs prior to use by the customer
• Such performance measurements include defect count measurements, thickness measurements (I e , indicative of deposition rate), and resistivity measurements, for example
• the number of defects is often predictable as a function of tool hardware life
• the resistivity and deposition rate also change with hardware life, and/or mamtenance history
• defects are more probable when the tool is first put together and at the end of its life
• the source of metal to be sputtered is referred to as a target
• the target is depleted as sputtering is conducted, and, thus, the target is changed frequently
• the resistivity and/or deposition rate of the layer being deposited change slowly after a target change and after a wet clean
• a chemical vapor deposition (CVD) tool reactive gases are introduced into the processing chamber through a gas supply header, commonly referred to as a showerhead Over time, process materials and/or byproducts collect on the showerhead, eventually leading to degraded performance As the showerhead becomes obstructed, the deposition rate of the CVD tool decreases and becomes more erratic
• the gas supplied to the showerhead often passes through an in-line filter Over time, the filter may become obstructed, thus reducmg the amount of reactive gases it passes This reduction could also eventually cause a reduction in the deposition rate of the tool
• preventative maintenance tasks such as cleanings, tool refurbishment, showerhead replacement, in-line filter replacement, and target replacement
• intervals are less than the expected time where the likelihood of instabilities increases
• a processmg line including a processing tool, a measurement tool, and an automatic process controller
• the processing tool is adapted to process articles
• the measurement tool is adapted to measure a characteristic of selected articles at a measurement frequency
• the automatic process controller is adapted to change the measurement frequency based on a usage characteristic of the processing tool
• Another aspect of the present invention is seen in a method for monitoring a processing tool
• the method includes processmg a plurality of articles in the processing tool, measurmg a characteristic of selected articles at a measurement frequency, and changing the measurement frequency based on a usage characteristic of the processing tool
• Figure 1 is a simplified block diagram of a processmg line in accordance with the present invention
• Figure 2 is a defect count curve for a deposition tool of Figure 1 following a cleaning or refurbishment
• Figure 3 is a deposition rate curve for the deposition tool of Figure 1 following a cleaning or part replacement
• Figure 4 is a flow diagram of a method for monitoring a processing tool in accordance with the present mvention
• the processing line 10 includes a processmg tool 20 for processing semiconductor wafers 30 and a measurement tool 40 for taking performance measurements on some of the wafers 30 processed by the processing tool 20 to measure the efficacy of the process implemented by the processing tool 20
• An automatic process controller 50 interfaces with the measurement tool 40 and the processing tool 20, at least in part, to determine the measurement frequency used by the measurement tool 40
• the particular process performed by the processing tool 20 and the particular performance measurement taken by the measurement tool 40 may vary widely Various illustrative examples will be described m greater detail hereinafter, however, application of the present invention is not limited to the particulars described in such examples Rather, the instant invention is applicable to a wide variety of processmg tools 20 related or not to semiconductor processing, and the performance measurement may be selected from a wide range of performance measurements applicable to the particular product being processed or the particular process being performed
• the automatic process controller 50 adjusts the measurement frequency of the measurement tool 40 based on usage characteristics of the processing tool 20
• Exemplary usage characte ⁇ stics include the time elapsed smce a preventative maintenance task has been performed (e g , cleanmg) and the time elapsed since a particular part in the processing tool 20 has been replaced (e g , target, shower head, in-line filter)
• the performance measurement taken by the measurement tool 40 to gauge the efficacy of the processing tool 20 may include parameters such as, but not limited to, process layer thickness, defect count, or resistivity
• the automatic process controller 50 comp ⁇ ses a computer executing application software adapted to perform the functions described
• the automatic process controller 50 may be coupled to other computers (not shown) over a computer network (not shown) Mamtenance on the processing tool 20 may be tracked by a factory control system accessible to the automatic process controller 50 Alternatively, the automatic process controller 50 may receive usage characteristics via manual input from the operator of the processing tool 20
• the measurement frequency of the measurement tool 40 is mcreased or decreased, based, at least in part, on the expected stability of the processing tool 20 During times of expected instability, the automatic process controller 50 increases the measurement frequency of the measurement tool 40, and during times of expected greater stability, the measurement frequency is decreased It will be appreciated that mtermediate frequencies may be used dunng transitions from expected periods of instability to periods of greater stability
• Exemplary period of increased instability mclude immediately following a cleaning operation, immediately following a target, showerhead, or in-line filter replacement, near the expected end-of- fe of a target, showerhead, or filter, near the anticipated time that a cleaning should be performed, etc
• Periods of greater stability may be defined as those periods other than those described above as being unstable, otherwise referred to as normal or steady-state operating conditions
• Figure 2 illustrates a defect count curve for a deposition tool following a cleanmg or refurbishment
• the defect count may be high
• the measurement frequency of the measurement tool 40 is mcreased as compared to its default value
• a steady-state period 110 follows the initial period 100, and the measurement frequency of the measurement tool 40 is decreased to the default value
• an end-of-life period 120 is entered where the defect count again may increase
• the measurement frequency of the measurement tool 40 is again mcreased
• the transition times between the steady-state period 110 and the initial and end-of-life periods 100, 120 are not completely predictable
• the determination of the times to transition from one period 100, 1 10, 120 to another may be based on predetermined time frames or, alternatively based on the performance measurements themselves
• the particular frequencies for the initial period 100 and the end-of-life period 120 may not necessarily be the same
• the automatic process controller 50 directs the measurement tool 40 to transition to the default frequency Likewise, if the measurement tool 40 is measuring at the default frequency, and the automatic process controller 50 detects a susta ed increase in the defect count, the automatic process controller 50 directs the measurement tool 40 to transition to the increased frequency
• Figure 3 illustrates a deposition rate curve for a deposition tool following a cleanmg or showerhead replacement
• the deposition rate is generally stable Dur g the initial period 200
• the measurement frequency of the measurement tool 40 is set by the automatic process controller 50 to the default frequency
• the deposition rate decreases
• the period marked by a marked decrease in deposition rate is the end-of-life period 210 Durmg the end-of-life period 120
• the measurement frequency of the automatic process controller increases the measurement frequency of the measurement tool 40 to the mcreased value
• the transition to the increased frequency could be used to trigger an alert or message to an operator of the processing tool 20 to evaluate the conditions and possibly schedule a maintenance task Usmg such an approach, the maintenance interval could be effectively lengthened when the transition to the higher frequency occurs at a point in time later than the expected time for entering the end-of-life interval 120, 210 This approach identifies both early degraded conditions as well as conditions more favorable than predicted where maintenance mtervals can be extended
• the specific frequencies for the default frequency and the increased frequency are application dependent
• the default frequency may be every N lots, every shift, every month, etc
• the increased frequency may be some multiple (e g , two or three times) the default frequency
• the automatic process controller 50 may be configured to adaptively change the frequency and or transition times based on the actual history of the measurement tool 40 For example, the automatic process controller 50 may attempt maintain a constant ratio between the percent failed and the percent measured That is, during periods with a low number of failures the frequency may be adaptively decreased Likewise, during periods with higher rates, the frequency may be adaptively increased In such an embodiment, there may not be distinct intervals, but rather a continuum of adaptively changing measurement frequencies based on the current failure rate
• the automatic process controller 50 may be adapted to change the measurement frequency of the measurement tool 40 to identify a number of degraded conditions For example, the automatic process controller 50 may increase the measurement frequency around times of expected cleanings, target replacement, shower-head replacement, and in-line filter replacement These intervals may or may not overlap
• FIG. 4 a flow diagram of a method for monitoring a processing tool 20 is provided
• a plurality of articles are processed in the processing tool 20
• a characteristic of selected articles is measured at a measurement frequency
• the measurement frequency is changed based on a usage characteristic of the processing tool 20, in block 320

Landscapes

• General Factory Administration (AREA)
• Chemical Vapour Deposition (AREA)

Priority Applications (2)

Applications Claiming Priority (2)

Publications (1)

Family

ID=23902975

Family Applications (1)

Country Status (5)

Cited By (20)

Families Citing this family (12)

Citations (2)

Family Cites Families (10)

• 2000
  • 2000-01-07 US US09/479,180 patent/US6469518B1/en not_active Expired - Lifetime
  • 2000-09-06 WO PCT/US2000/024411 patent/WO2001052319A1/en not_active Ceased
  • 2000-09-06 EP EP00961575A patent/EP1245041A1/en not_active Ceased
  • 2000-09-06 JP JP2001552441A patent/JP2003520434A/ja not_active Withdrawn
  • 2000-09-06 KR KR1020027008824A patent/KR100708426B1/ko not_active Expired - Fee Related

Patent Citations (2)

Also Published As

Similar Documents

Legal Events