JP2004028823A - Workpiece processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a workpiece processing apparatus for surely and easily detecting abnormality of a processed workpiece without an influence from background noise. <P>SOLUTION: The workpiece processing apparatus 30 is provided with a processing chamber 11 for processing the workpiece W, a workpiece support means 12 disposed in the processing chamber 11 and supporting the workpiece W, an acoustoelectric conversion element 13 disposed outside the processing chamber 11, receiving a sound wave and converting it into an electrical signal, a sound wave transmission means 31 for transmitting the sound wave generated in the processing chamber 11 to the acoustoelectric conversion element 13 and an abnormality detection means 15 for detecting the abnormality of the workpiece W based on an output signal from the acoustoelectric conversion element 17. The sound wave transmission means 31 noncontactly penetrates a wall 11a of the processing chamber 11 through a sound wave attenuation member 10. One end of the sound wave transmission means 31 contacts the workpiece W or the workpiece support means 12 in the processing chamber 11. The other end is connected to the acoustoelectric conversion element 13 outside the processing chamber 11. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a workpiece processing apparatus that performs various types of processing such as sputtering on a workpiece such as a semiconductor wafer, and more particularly to a technique for detecting an abnormality occurring in a workpiece in real time (immediately) during workpiece processing. Is.
[0002]
[Prior art]
Generally, a semiconductor wafer undergoes a complicated process before being sputtered, and various stresses are applied and accumulated in the process. Furthermore, in the sputtering process, thermal shock due to heating and cooling, impact due to collision of sputtered particles, and other stresses are applied from the electrostatic chuck, mechanical chuck, and sputtered film. It may become unbearable and may crack or chip.
[0003]
If the wafer is cracked or chipped during the sputtering process, a sputtered film is also formed on the stage surface on which the wafer is placed, resulting in non-uniform heat distribution on the stage surface and poor heating of the wafer. It leads to waking up.
[0004]
Conventionally, as a method of detecting breakage such as a crack or chip of a wafer, there is a method of irradiating the wafer with light after the processing is completed and detecting breakage of the wafer from the transmission or reflection of the light. However, this method cannot detect the damage during wafer processing in real time and cannot stop the processing immediately. There are problems such as the need to replace many parts and lengthening the downtime of the apparatus.
[0005]
Thus, for example, Japanese Patent Application Laid-Open Nos. 7-120440 and 2000-24919 disclose a method for detecting breakage of a workpiece in real time during workpiece processing using an AE sensor. This is to detect breakage of a workpiece by detecting an elastic wave generated along with breakage of the workpiece with an AE sensor.
[0006]
[Problems to be solved by the invention]
However, the elastic wave is also generated due to various factors other than the breakage of the workpiece, and the AE sensor detects such an elastic wave without distinction. If the detection signal of the AE sensor contains a lot of background noise due to elastic waves that are not caused by work breakage, it becomes difficult to identify the elastic waves caused by work breakage and the work breaks. It becomes impossible to reliably and easily detect whether or not.
[0007]
The present invention has been made in view of the above-described problems, and an object of the present invention is to perform workpiece processing capable of reliably and easily detecting an abnormality of a workpiece during workpiece processing in real time without being affected by background noise. To provide an apparatus.
[0008]
[Means for Solving the Problems]
The work processing apparatus of the present invention is provided with a processing chamber for processing a work, a work supporting means for supporting the work, and an outside of the processing chamber. An acoustoelectric transducer for converting to an acoustic wave, elastic wave transmitting means for transmitting an elastic wave generated inside the processing chamber to the acoustoelectric transducer, and an abnormality for detecting an abnormality of the workpiece based on the output signal of the acoustoelectric transducer Detecting means,
Further, the elastic wave transmission means penetrates the wall portion of the processing chamber in a non-contact manner through the elastic wave attenuating member, one end is brought into contact with the work or the work supporting means inside the processing chamber, and the other end Is connected to the acoustoelectric transducer outside the processing chamber.
[0009]
When a solid deforms or breaks, an elastic wave called AE (Acoustic Emission) is generated. The elastic wave propagates through the solid and reaches the acoustoelectric converter, and the acoustoelectric converter converts the elastic wave into an electric signal and outputs it to the abnormality detecting means. This makes it possible to detect an abnormality that generates an elastic wave. According to this method, it is possible to catch in real time a minute abnormality such as an elastic wave due to cracking or deformation of an object in operation, which is effective when constant monitoring is required.
[0010]
The elastic wave generated due to the abnormality of the workpiece in the processing chamber is transmitted to the acoustoelectric transducer disposed outside the processing chamber via the elastic wave transmitting means.
Since the elastic wave transmission means interposes an elastic wave attenuating member between the wall portion of the processing chamber, the elastic wave that is not caused by the abnormality of the workpiece transmitted through the wall portion of the processing chamber is generated by the elastic wave attenuating member. It is possible to suppress the elastic wave that is attenuated and enters the acoustic wave transmitting means and the acoustoelectric transducer as background noise.
[0011]
As a material of the elastic wave attenuating member, for example, a resin such as a fluorine-based resin, a fiber such as rubber, glass fiber, or cotton can be used.
[0012]
Acoustoelectric transducers include crystal oscillators, electrostrictive ceramics such as lithium niobate, barium titanate (BaTiO3), lead zirconate (PbZrO3), lead titanate (PbTiO3) solid solutions (so-called PZT), ferrites, etc. A magnetostrictive vibrator or the like is used. Further, since the acoustoelectric transducer is disposed outside the processing chamber, the processing chamber may be in any atmosphere such as in vacuum, high pressure, or plasma.
[0013]
The abnormality detection means detects the occurrence of abnormality based on, for example, various waveform feature parameters obtained from the output signal of the acoustoelectric transducer.
For example, it is known that elastic waves generated when a hard work such as a silicon wafer or a glass substrate breaks or deforms tend to be high frequency, and the output signal waveform of the acoustoelectric transducer is Fourier transformed to obtain a spectrum for each frequency. The occurrence of an abnormality can be reliably detected by looking at the screen.
In addition, since mechanical vibration component noise usually has a low frequency component, it is also effective to remove the low frequency component using a high-pass filter.
In addition, for example, the elastic wave caused by the occurrence of the abnormality of the workpiece can be specified also by the duration in which the amplitude equal to or greater than a certain threshold value is maintained.
[0014]
Examples of abnormalities in the workpiece being processed include cracking, chipping and deformation of silicon wafers and glass substrates, cracking of powder, cracking during plastic injection molding, and the like.
[0015]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0016]
(First embodiment)
FIG. 1 is a schematic diagram showing an overall configuration of a work processing apparatus 30 according to a first embodiment of the present invention. The work processing apparatus 30 is generated inside the processing chamber 11, the work supporting means 12 disposed inside the processing chamber 11, the AE sensor 13 disposed outside the processing chamber 11, and the processing chamber 11. An elastic wave transmission unit 31 that transmits an elastic wave to the AE sensor 13, an amplifier 14 that amplifies a weak output signal from the AE sensor 13, and an abnormality detection unit 15 that receives the output signal of the amplifier 14 are provided. The workpiece processing apparatus 30 is a semiconductor manufacturing apparatus that performs a sputtering process or the like on a semiconductor wafer W that is a workpiece, for example.
[0017]
The AE sensor 13 is an acoustoelectric transducer that detects an elastic wave generated when the workpiece W is cracked, converts it into an electrical signal, and outputs the electrical signal.
[0018]
The elastic wave transmission means 31 has a rod shape, one end of which is in contact with the work support means 12 in the processing chamber 11 and the other end is connected to the AE sensor 13 outside the processing chamber 11 (configuration of the AE sensor 13). The elastic wave can be transmitted to the piezoelectric element by contacting the element piezoelectric element).
When the inside of the processing chamber 11 is relatively hot, for example, a ceramic elastic wave transmission means 31 is used. If the temperature in the processing chamber 11 is not so high, for example, a metal such as stainless steel may be used as the material of the elastic wave transmission means 31.
[0019]
An elastic wave attenuating member 10 made of resin, rubber or the like is embedded in the wall (side wall) 11a of the processing chamber 11. The elastic wave transmitting means 31 penetrates the elastic wave attenuating member 10 and has one end positioned inside the processing chamber 11 and the other end positioned outside the processing chamber 11. That is, the elastic wave transmission means 31 is in contact with the elastic wave attenuating member 10 and not in contact with the wall portion 11 a in a portion penetrating the wall portion 11 a of the processing chamber 11. Therefore, an external elastic wave (an elastic wave not caused by an abnormality of the workpiece W) transmitted through the wall portion 11a is attenuated by the elastic wave attenuating member 10 and the arrival at the AE sensor 13 can be suppressed.
[0020]
The workpiece support means 12 includes a plate 12b made of, for example, copper installed on the installation surface 11b in the processing chamber 11, and a plate 12a made of, for example, carbon on which the workpiece W is placed. Further, an elastic wave attenuating member 9 made of resin or rubber is incorporated between the plates 12a and 12b. Therefore, an external elastic wave (an elastic wave not caused by an abnormality of the workpiece W) transmitted through the installation surface 11 b is attenuated by the elastic wave attenuating member 9, and can be prevented from reaching the elastic wave transmission means 31.
[0021]
In addition, although not shown, the work processing apparatus 30 is provided with a loading / unloading means for loading / unloading the wafer W into / from the processing chamber 11, an exhaust port, a vacuum pump, a sputtering target, a process gas introduction means, and the like.
[0022]
FIG. 2 is a block diagram showing the configuration of the abnormality detection means 15. The abnormality detection unit 15 includes a display unit (for example, a liquid crystal display) 25, an arithmetic processing unit (for example, a CPU) 27, a storage unit (for example, a hard disk) 26, and alarm units (for example, lamps) 16 and 17. The lamp 16 is lit in green when normal, and the lamp 17 is lit in red when an abnormality occurs.
[0023]
The work processing apparatus 30 of the present embodiment is configured as described above, and the operation thereof will be described next.
[0024]
The workpiece (semiconductor wafer) W carried into the processing chamber 11 is sputtered while being heated on the workpiece supporting means 12. When the work W is cracked or chipped during the processing, the elastic wave generated along with the crack or chip is from the work support means 12 and is in contact with the work support means 12. It propagates to the transmission means 31 and further reaches the AE sensor 13 via the elastic wave transmission means 31.
[0025]
The AE sensor 13 detects the elastic wave, converts it into an electrical signal, and outputs it to the amplifier 14. The output signal of the AE sensor 13 amplified by the amplifier 14 is input to the abnormality detecting means 15 and sequentially processed. The arithmetic processing means 27 built in the abnormality detection means 15 calculates various waveform characteristic parameters from the output signal of the AE sensor 13. Further, the arithmetic processing means 27 compares the calculated waveform feature parameter with the waveform feature parameter (stored in the storage unit 26) obtained in the normal processing in which no cracks or the like have occurred in the wafer W in advance. Thus, it is determined whether or not an abnormality such as a wafer crack has occurred in the processing chamber 11.
[0026]
Examples of the waveform feature parameters include those shown below.
FIG. 3 shows the waveform of the output signal of the AE sensor 13 when an abnormality occurs, and various waveform characteristic parameters will be described with reference to this.
[0027]
Threshold value: A reference level for recognizing a signal with an amplitude larger than this as abnormal (the amplitude of background noise during normal operation is smaller than this threshold value)
Count: Number of times of a series of signal waveforms exceeding a threshold Number of events; Maximum value obtained by counting a series of signal waveforms exceeding a threshold as one pulse group; Maximum voltage reached by a signal waveform of one event Energy: The total amount of elastic energy released by a single event, the area within the envelope of the waveform shown in FIG. 3, or the root mean square of the maximum value; the root mean square rise time; Time duration from start of signal to maximum; time from start of signal to end of crossing threshold
Furthermore, the following are mentioned as the waveform feature parameters indicating the central tendency, variation, and distribution of the waveform feature parameters within a certain time or cycle time.
For example, the calculation formulas for the count (x) and energy (y) mentioned above are shown below. (N is the number of events within a certain time or cycle time, Σ is the sum from i = 1 to n)
Count average value M _x = (1 / n) Σx _i
Energy average value M _y = (1 / n) Σy _i
Count distribution V _x = (1 / n−1) Σ (x _i −x) ²
Energy dispersion V _y = (1 / n−1) Σ (y _i −y) ²
Count standard deviation S _x = V _x ^1/2 
Energy standard deviation S _y = V _y ^1/2
Count variation count CV _x = S _x / M _x
Energy fluctuation count CV _y = S _y / M _y
Count kurtosis K _x = m _x4 / m ² _x2
m _xk = (1 / n) Σ (x _i −x) ^k
Energy kurtosis Ky = m _y4 / m ² _y2
m _yk = (1 / n) Σ (x _i −x) ^k
[0029]
Further, based on the frequency spectrum characteristics obtained by performing fast Fourier transform (FFT) on the signal waveform of FIG.
Total power (total energy); area peak spectrum of the power spectrum of one event; maximum peak frequency of the power spectrum of one event; frequency average (center) frequency at the maximum power spectrum value of one event; one event Power spectrum average (center) frequency spectrum height ratio; ratio of average and maximum power spectrum of one event
FIG. 4 shows an output signal waveform of the AE sensor 13 obtained in a test for artificially generating an abnormal sound.
[0031]
FIG. 4A shows an output signal waveform of the AE sensor 13 when a silicon wafer is placed on the floor beside the processing chamber 11 and beaten with an aluminum plate.
FIG. 4B shows an output signal waveform of the AE sensor 13 when a silicon wafer is placed in the same place as FIG. 4A and is struck lightly by an aluminum plate than in the case of FIG. 4A.
FIG. 4C shows an output signal waveform of the AE sensor 13 when the processing chamber 11 is directly hit with an aluminum plate.
In the case of FIGS. 4A and 4B, silicon wafers of the same material, shape, and dimensions were used.
[0032]
As can be seen from these results, when the wafer breaks, a signal waveform is obtained in which the rise is sharp and the amplitude gradually decreases thereafter. Further, it is possible to recognize the magnitude of the crack from the magnitude of the amplitude of the signal waveform.
[0033]
FIG. 5 is a frequency spectrum graph obtained by subjecting each signal of FIG. 4 to A / D conversion and then fast Fourier transform. 5A corresponds to FIG. 4A, FIG. 5B corresponds to FIG. 4B, and FIG. 5C corresponds to FIG. 4C. When the wafer is broken, it can be seen that there is a peak at the frequency (around 190 kHz to 230 kHz) regardless of the size of the crack.
[0034]
Among the waveform feature parameters mentioned above, for example, a threshold value, a rise time, and a duration are used for abnormality detection discrimination. In addition, the abnormality detection accuracy can be improved by combining a plurality of waveform characteristic parameters instead of determining them by one. Furthermore, in addition to the determination using the waveform feature parameter, if the output signal waveform of the AE sensor 13 is displayed on the display means 25 and this is also monitored by the operator at the same time, the abnormality can be detected. It can be done more reliably.
[0035]
When the abnormality detection unit 15 detects the occurrence of an abnormality, the abnormality detection unit 15 outputs a stop signal to the control unit of the work processing apparatus 1 and receives the stop signal to control the work processing apparatus 1. The unit immediately stops the work processing apparatus 1. At the same time, the abnormality detection means 15 outputs an abnormality occurrence signal to the lamp 17 and lights the lamp 17 in red to notify the operator of the occurrence of the abnormality. Alternatively, an alarm may be issued with sound instead of lighting the lamp 17.
[0036]
As described above, according to the present embodiment, an elastic wave caused by an external mechanical vibration transmitted from the wall 11a or the installation surface 11b of the processing chamber 11 is applied to the elastic wave attenuating members 9, 10. And can be suppressed from being transmitted to the elastic wave transmission means 31 and the AE sensor 13. Therefore, the AE sensor 13 can detect an elastic wave caused by an abnormality of the workpiece W such as a crack of the workpiece W while reducing the influence of the background noise, and whether or not an abnormality has occurred in the workpiece W. Certainly and easily identified.
[0037]
Then, when the workpiece W being processed in the processing chamber 11 is cracked or chipped, the processing can be stopped immediately, so that the portion other than the workpiece W is processed through the cracked or chipped portion. Can be prevented. As a result, the restoration work of the apparatus can be minimized, and a reduction in the operating rate of the apparatus can be suppressed.
[0038]
Further, since the AE sensor 13 is disposed outside the processing chamber 13, it is not damaged by the temperature in the processing chamber 11 or the process gas.
[0039]
Further, if the output signal of the AE sensor 13, the waveform feature parameter calculated from the output signal, the abnormality detection result based on the waveform feature parameter, and the like are stored in the storage means 26, these stored various Data can be reflected in the subsequent processing as past processing history. For example, it is possible to know at which point in the processing the work breaks, and the frequency thereof, and to change the processing conditions for the next and subsequent times to help prevent the occurrence of an abnormality.
Of course, the various data stored in the storage means 26 can also be used as comparison data when determining whether or not there is an abnormality.
[0040]
(Second Embodiment)
Next, a second embodiment of the present invention will be described. In addition, the same code | symbol is attached | subjected to the same component as the said 1st Embodiment, and the detailed description is abbreviate | omitted.
[0041]
FIG. 6 shows a cross-sectional view of the bottom wall portion of the processing chamber 11 ′ in the second embodiment. The bottom wall portion of the processing chamber 11 ′ includes a fixing portion 11′a made of, for example, aluminum and an attachment portion 5 made of, for example, stainless steel. The mounting portion 5 is airtightly fixed to the fixing portion 11′a by bolts 6 with elastic wave attenuating members 7a and 7b made of resin or rubber.
[0042]
Further, an O-ring 8 made of, for example, rubber is interposed between the mounting portion 5 and the elastic wave transmitting means 31 faces the processing chamber 11 ′, one end of which is the bottom surface of the work supporting means 12 (plate 12b). Touching. The other end is connected to the AE sensor 13. The O-ring 8 has a function of a seal that hermetically blocks the inside and outside of the processing chamber 11 ′, and also functions as an elastic wave attenuation member that attenuates an elastic wave transmitted through the attachment portion 5 and suppresses arrival at the elastic wave transmission means 31. To do. The elastic wave attenuating members 7a and 7b also have a function of attenuating elastic waves and a function of a seal that hermetically blocks the inside and outside of the processing chamber 11 ′.
[0043]
Also in the present embodiment, as in the first embodiment, elastic waves caused by external mechanical vibration transmitted from the wall portion of the processing chamber 11 ′ are converted into elastic wave attenuating members 7 a and 7 b, O-rings. 8, it is possible to suppress transmission to the elastic wave transmission means 31 and the AE sensor 13. Therefore, the AE sensor 13 can detect an elastic wave caused by an abnormality of the workpiece W such as a crack of the workpiece W while reducing the influence of the background noise, and whether or not an abnormality has occurred in the workpiece W. Certainly and easily identified.
[0044]
As mentioned above, although each embodiment of this invention was described, of course, this invention is not limited to these, A various deformation | transformation is possible based on the technical idea of this invention.
[0045]
As processing for semiconductor wafers as workpieces, in addition to sputtering processing, cleaning, drying, lithography, development, CVD, plating, vapor deposition, ion implantation, resist coating, etching, annealing, CMP, polishing, dicing, etc. As an example, it is possible to detect abnormality during these processes.
[0046]
In addition, the display unit 25, the arithmetic processing unit 27, the storage unit 26, and the alarm units 16 and 17 shown in FIG. 2 are not limited to the form provided in the abnormality detection unit 15, and may be provided separately from the abnormality detection unit 15. .
[0047]
Further, the elastic wave transmission means 31 may be brought into direct contact with the workpiece W with one end thereof being pointed conically.
[0048]
In the first embodiment shown in FIG. 1, the elastic wave attenuating member 9 may be interposed between the installation surface 11b and the workpiece support means 12 (plate 12b).
[0049]
Further, if a plurality of sets of the AE sensor 13 and the elastic wave transmission means 31 penetrating the wall portion of the processing chamber through the elastic wave attenuating member 10 are used for the processing chamber 11, one processing chamber is used. It is possible to identify where an abnormality has occurred.
[0050]
【The invention's effect】
As described above, according to the present invention, an elastic wave caused by an abnormality occurring in a workpiece during workpiece processing can be reliably detected in real time while reducing the influence of background noise, and easily It can be identified as an elastic wave caused by a workpiece abnormality. As a result, the damage caused by the abnormality of the workpiece can be minimized.
[Brief description of the drawings]
FIG. 1 is a schematic perspective view showing a configuration of a work processing apparatus according to a first embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of an abnormality detection unit.
FIG. 3 is an output signal waveform diagram of an AE sensor that detects an abnormal elastic wave caused by a workpiece abnormality.
FIG. 4 is a waveform diagram of an output signal of an AE sensor obtained by an experiment, where A is when a large crack is generated in the workpiece, B is when a small crack is generated in the workpiece, and C is hitting the processing chamber itself. It is the output signal waveform figure of the AE sensor at the time of hitting.
FIG. 5 is a graph of a frequency spectrum obtained by Fourier transforming each of FIGS.
FIG. 6 is a cross-sectional view of a main part of a work processing apparatus according to a second embodiment of the present invention.
[Explanation of symbols]
7a, 7b ... elastic wave attenuation member, 8 ... elastic wave attenuation member, 9 ... elastic wave attenuation member, 10 ... elastic wave attenuation member, 11 ... treatment chamber, 11 '... treatment chamber, 11a ... wall portion, 11b ... installation surface , 12 ... Work support means, 13 ... Acoustoelectric transducer (AE sensor), 15 ... Abnormality detection means, 17 ... Alarm lamp, 25 ... Display means, 26 ... Storage means, 27 ... Arithmetic processing means, 30 ... Work processing device 31 ... elastic wave transmission means, W ... work (semiconductor wafer).

Claims (9)

A processing chamber for processing workpieces;
A workpiece support means disposed inside the processing chamber and supporting the workpiece;
An acoustoelectric transducer disposed outside the processing chamber and receiving an elastic wave to convert it into an electrical signal;
Elastic wave transmission means for transmitting elastic waves generated inside the processing chamber to the acoustoelectric transducer;
A workpiece processing apparatus comprising an abnormality detection means for detecting an abnormality of the workpiece based on an output signal of the acoustoelectric transducer,
The elastic wave transmission means passes through the wall portion without contact with the wall portion of the processing chamber with an elastic wave attenuating member interposed, and has one end in contact with the workpiece or the workpiece supporting means inside the processing chamber. The work processing apparatus, wherein the other end is connected to the acoustoelectric transducer outside the processing chamber. The work processing apparatus according to claim 1, wherein the work support means is installed with an elastic wave attenuating member interposed between the work support means and an installation surface inside the processing chamber. The work processing apparatus according to claim 1, wherein an elastic wave attenuating member is incorporated in the work support means. A plurality of sets of the acoustoelectric transducer and the elastic wave transmission means are provided for the processing chamber, and the occurrence position of the abnormality is specified from a time difference between output signals of the acoustoelectric transducer. The work processing apparatus according to claim 1. The work processing apparatus according to claim 1, further comprising an arithmetic processing unit that calculates a waveform characteristic parameter from the output signal of the acoustoelectric transducer. Alarm means for notifying the occurrence of the abnormality is provided,
The workpiece processing apparatus according to claim 1, wherein the abnormality detection unit outputs an abnormality occurrence signal to the alarm unit when the abnormality is detected. The work processing apparatus according to claim 1, wherein when the abnormality is detected, the operation of the processing chamber is automatically stopped. The work processing apparatus according to claim 1, further comprising a storage unit that stores a processing result of the output signal of the acoustoelectric transducer. The workpiece processing apparatus according to claim 1, wherein the workpiece is a semiconductor wafer.

Images