Adhesive bonding and miscellaneous chemical manufacture – Differential fluid etching apparatus – With measuring – sensing – detection or process control means
Reexamination Certificate
2000-08-30
2003-01-07
Mills, Gregory (Department: 1763)
Adhesive bonding and miscellaneous chemical manufacture
Differential fluid etching apparatus
With measuring, sensing, detection or process control means
C156S345420, C156S345460, C156S345490, C118S712000, C118S7230MA, C118S7230MR
Reexamination Certificate
active
06503364
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a plasma processing apparatus and, more particularly, to a plasma processing apparatus which is capable of forming a fine pattern during semiconductor manufacture. More particularly, the invention relates to a window for measuring optical emission in a processing chamber.
During semiconductor manufacture, for fine processing, such as etching, film forming, and ashing, a plasma processing apparatus is widely used. Among them, there is an etching apparatus which converts process gas introduced into a vacuum chamber (reactor) to plasma using a plasma generation means, causes the plasma to react on the semiconductor wafer surface, executes fine processing, and also performs a predetermined etching process by exhausting volatile reaction products.
In this plasma etching apparatus, the optical emission intensity of specific radicals or ions is measured for plasma optical emission generated by etching of a wiring layer or insulating film; and, for example, the end point of the etching process is detected from a change with time of the optical emission intensity during the etching process. The radical composition is determined from the dependency of the optical emission intensity of the radical species on the process parameter so as to construct the process, and changes in the plasma chemistry are monitored.
However, when the etching process of a wafer is actually repeated over a period of several hours to several tens of hours, reaction products and a part of the process gas gradually adhere to the inner surface of the window for plasma optical emission measurement, and so the transmission factor of the measurement window is reduced. Therefore, problems arise in that the end point detection accuracy is reduced and the radical composition cannot be monitored accurately.
As a countermeasure for this problem, in Japanese Patent Application Laid-open 8-111403, an arrangement is described in which a detection window for plasma optical emission is installed at the top of a cylindrical hollow support portion; and, hence, adhesion of reaction products onto the inner surface of the detection window is suppressed, whereby an etching apparatus for stably determining the end point of etching over a long period of time is achieved. In Japanese Patent Application Laid-open 9-199476, an etching apparatus is described wherein, in the same way as with the aforementioned patent application, a cylindrical light guide is used, and a detection window for plasma optical emission is installed at the end thereof. More particularly, the length of the light guide is set to 10 times or more of the inner diameter; and, furthermore, inactive gas is introduced into the light guide, and the inside of the light guide is positively pressured, so that adhesion of reaction products is suppressed.
In Japanese Patent Application Laid-open 9-330917, a method is proposed in which some non-through holes are installed in a detection window in a block-type shape, and an end point detector is installed at the position opposite to the bottom thereof. Furthermore, Japanese Patent Application Laid-Open 1-232725 describes a method in which a heater is installed in a detection window, and the window portion is heated, so that adhesion of reaction products is suppressed. Also, Japanese Patent Application Laid-open 63-128632 describes a method in which, when a magnetic field is applied by a coil installed around an intermediate cylinder attached to a light measuring window, reaction products and ions are deflected and prevented from adhering to the light measuring window.
However, the aforementioned methods cannot completely suppress the reduction of the transmission factor of the optical emission measuring window, and so a problem arises in that the detection intensity of optical emission reduces gradually. For example, in the method using the non-through holes described in Japanese Patent Application Laid-open 9-330917, although the reduction of the transmission factor is suppressed more than that of other methods, data showing that the detection intensity is reduced to less than about 50% for about 10 hours of discharge is available.
Also, in Japanese Patent Application Laid-open 9-199476 mentioned above, it is indicated that, when the length of the light guide is set to 10 times or more of the inner diameter, the deposition amount of reaction products on the optical emission measuring window is reduced to {fraction (1/10)} or less compared with a case in which the length is 5 times or less of the inner diameter. Also, in this case, the deposition amount is not reduced to zero and the detection intensity reduces with the progress of time. In this patent application, it is also described that, when an inactive gas of positive pressure (for example, 101 mTorr for a processing pressure of 100 mTorr) is introduced into the light guide, reaction products find it hard to enter inside the light guide. When the processing pressure is 100 mTorr or more, the gas flow is in the area of viscous flow; and, furthermore, the mean free path of molecules is 1 mm or less, so that the molecules are discharged before they can adhere to the inner wall of the light guide, and an effect that the molecules do not easily enter the light guide can be expected to some extent. However, as the process is refined, the processing pressure is reduced to several Pa or less (several tens mTorr or less), and, in this pressure area, the flow is a molecular flow, so that an effect of discharge by the gas flow cannot be expected. Since the mean free path of molecules becomes longer, such as several mm to several tens mm, it may be considered that the probability of arrival of reaction products at the detection window is increased, and the deposition amount on the measuring window is increased.
Furthermore, in the method which calls for installing a heater on the end point detection window and heating the window portion, it becomes necessary to separately add a heating device and a safety mechanism in the etching apparatus; and, in the method for deflecting reaction products and ions using a magnetic field, a coil for generating the magnetic field is required separately, and both methods resultantly cause a complication of the apparatus and an increase in cost, and hence they are not practical. In the method which calls for using a magnetic field, the magnetic field causes disturbance of the plasma. Furthermore, in these methods, the effects of heating and the use of a magnetic field are not indicated quantitatively and are not clear.
Each of the aforementioned considerations relates to end point detection, but it is also desirable to detect dynamic intensity changes during processing, so that a gradual decrease of the transmission factor for many hours disturbs the signal detection very little. However, for example, when changes in the radical composition under a continuous discharge test are to be monitored, when the transmission factor of the measuring window reduces, the apparent detection intensity reduces and changes cannot be monitored accurately, so that it is necessary to measure the radical optical emission intensity without being adversely affected by such a disturbance factor.
Particularly, when a silicon oxide film is to be etched using a CF babies etching gas, such as C
4
F
8
, radical optical emission of CF and CF
2
, which are etchants, is important for the process characteristics. However, the optical emission of them is in the area within the range of the peak wave length from 200 nm to 300 nm, and, in this area, the transmission factor of the measuring window is reduced greatly by the deposited film of the CF series, so that the reduction in the optical emission detection intensity appears more remarkably. As a result, without being adversely affected by the reduction in the transmission factor due to adhesion of the deposited film on the measuring window, it is necessary to measure plasma optical emission stably over a long period of time.
SUMMARY OF THE INVENTION
The present invention i
Masuda Toshio
Shirayone Shigeru
Suehiro Mitsuru
Takahashi Kazue
Usui Tatehito
Alejandro Luz L.
Mills Gregory
LandOfFree
Plasma processing apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Plasma processing apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Plasma processing apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3073451