Voltage control sensor and control interface for radio...

Details Voltage control sensor and control interface for radio... Voltage control sensor and control interface for radio...

2000-04-25

2003-05-13

Paschall, Mark (Department: 3742)

Electric heating

Metal heating

By arc

C219S121430, C219S121580, C118S7230IR, C156S345420

Reexamination Certificate

active

06563076

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to plasma processing systems and, more particularly, to methods and apparatus for controlling radio frequency delivery in a plasma reactor through monitoring and feedback of an electrical parameter, in particular a peak voltage.
2. Background Art
Ionized gas, or plasma, is commonly used during the processing and fabrication of semiconductor devices. For example, plasma can be used to etch or remove material from semiconductor integrated circuit wafers, and to sputter or deposit material onto semiconducting, conducting or insulating surfaces.
With reference to
FIG. 1A
, creating a plasma for use in manufacturing or fabrication processes typically begins by introducing various process gases into a plasma chamber
10
of a plasma reactor, generally designated
12
. These gases enter the chamber
10
through an inlet
13
and exit through an outlet
15
. A workpiece
14
, such as an integrated circuit wafer is disposed in the chamber
10
held upon a chuck
16
. The reactor
12
also includes plasma density production mechanism
18
(e.g. a TCP coil). A plasma inducing signal, supplied by a plasma inducing power supply
20
is applied to the plasma density production mechanism
18
. the plasma inducing signal is preferably a radio frequency (RF) signal. A dielectric window
22
, constructed of a material such as ceramic, incorporated into the upper surface of the chamber
10
allows efficient transmission of the first RF signal from the TCP coil
18
to the interior of the grounded chamber
10
. This first RF signal excites the gas molecules within the chamber, generating a plasma
24
.
The plasma
24
formed within the chamber
10
includes electrons and positively charged particles. The electrons, being lighter than the positively charged particles tend to migrate more readily, causing a sheath to form at the surfaces of the chamber
10
. A self biasing effect causes a net negative charge at the inner surfaces of the chamber. This net negative charge, or D.C. sheath potential acts to attract the heavier positively charged particles toward the wall surfaces. The strength of this D.C. bias in the location of the workpiece
14
largely determines the energy with which the positively charged particles will strike the workpiece
14
and correspondingly affects the desired process (e.g. etch rate, or deposition rate).
The present invention will be more readily understood by bearing in mind the distinction between DC bias and DC sheath potential. DC bias is defined as the difference in electrical potential between a surface within the chamber
10
and ground. DC sheath, on the other hand is defined as the difference between the plasma potential and the potential of a surface within the chamber as measured across the plasma sheath.
The workpiece is held upon a chuck
16
is located at the bottom of the chamber
10
and constitutes a chuck electrode
26
. A bias RF power source
28
supplies a biasing RF signal to the chuck electrode
16
. Alternatively, in some systems the both the plasma density signal and bias signal are in fact a single signal produced by a single power source.
This second excitation signal, preferably in the form of a RF signal, at the second electrode increases the DC bias at the location of the workpiece, depending on the disposition of the RF electric field within the chamber
10
, and this increases the energy with which the charged particles strike the workpiece. Variations in the RF signal supplied to the second electrode
16
produce corresponding variations in the D.C. bias at the workpiece affecting the process.
With continued reference to
FIG. 1A
, the bias RF power source
28
described above supplies a R.F. signal to the chuck electrode
26
. This signal passes through a match network
30
disposed between the bias RF power source
28
and the chuck electrode
26
. The match network
30
matches the impedance of the RF signal with the load exhibited by the plasma. A similar match network
31
is provided between the power inducing power source
20
and the TCP coil
18
. As discussed above, the control and delivery of the RF signal at the chuck electrode
26
is of fundamental importance in plasma processing. Significant variance in actual power delivered may unexpectedly change the rate of the process. Unfortunately, the match network
30
generates significant losses in the RF signal. Furthermore, these losses are variable and, to a degree, unpredictable. Therefore, simply supplying a predetermined RF signal power from the RF power source
28
does not ensure that a predictable and consistent RF signal will be delivered at the electrode
26
.
With continued reference to
FIG. 1A
, one method which has been used to attach the workpiece
14
to the chuck
16
has been to provide the chuck with clamps
32
which contact the surface of the workpiece along its edges to hold the workpiece to the chuck. Using such a chuck
16
(and to the extent that the workpiece is somewhat conductive) it is possible to measure the D.C. bias directly by installing a pickup
33
at the electrode
26
and transmitting a voltage signal to a voltage sensor
34
. The power source could then be feedback controlled to maintain a constant measured D.C. bias. However, using such clamps
32
to attach the workpiece
14
to the chuck
16
presents multiple problems. For one, valuable surface area may be wasted on the workpiece due to its engagement with the clamps
32
. In addition, any such contact of clamps
32
to the workpiece
14
is undesirable due the risk of damage to the workpiece
14
, and the generation of particles.
With reference to
FIG. 1B
, another method which has been used to hold the workpiece onto the electrode has been to provide an electrode in the form of an electrostatic chuck
36
. In its most general sense an electrostatic chuck includes an electrode
38
which is covered with an insulator
40
. The electrically conductive workpiece
14
, which is generally semiconductive, sits on the electrically insulating material. When a DC voltage is applied to the electrode
38
, the electrode and workpiece
14
become capacitively coupled resulting in opposite electrical charges on each, attracting the workpiece
14
and electrode
38
toward one another. This acts to hold the workpiece against the chuck
36
.
More particularly, the electrostatic chuck
36
can be understood with reference to
FIG. 1C
in addition to FIG.
1
B. In this bipolar implementation, the electrode
38
of the electrostatic chuck
36
includes first and second electrically conducive portions
42
and
44
, which are electrically isolated from one another. A DC voltage from a D.C. voltage source
46
, passes through a filter
47
before being applied between the first and second portions
42
and
44
of the electrode
38
. This causes the desired electrostatic attraction between the electrode
38
and the workpiece
14
, thereby holding the workpiece to the chuck
36
.
With reference to
FIG. 1D
, a simpler version of electrostatic chuck is illustrated. This simpler form of electrostatic chuck, termed a mono polar chuck
37
is shown in plan view in
1
D. By applying a DC potential between the workpiece
14
and the chuck an electrostatic charge on each holds the workpiece to the chuck. It will be appreciated by those skilled in the art that numerous other forms of electrostatic chuck are possible as well.
However, use of such an electrostatic chuck
36
renders a direct measurement of the D.C. bias at the workpiece impractical. End users are averse to having their sensitive semiconductor products touched by any mechanical probe or electrically conductive item such as a voltage sensor. In addition, it would be difficult to maintain sensor accuracy and longevity in the plasma environment. Correlating the D.C. voltage by measuring the power of the RF signal at the electrode
16
is also difficult and does not provide an accurate measurement of the D.C. sheath potential due, in part, to the capacitive coupling b

Affiliated with

Also associated with

Rating

Voltage control sensor and control interface for radio... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Voltage control sensor and control interface for radio..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Voltage control sensor and control interface for radio... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3090995