Measurement of substrate temperature in a process chamber...

Details Measurement of substrate temperature in a process chamber... Measurement of substrate temperature in a process chamber...

2001-11-28

2003-05-13

Paschall, Mark (Department: 3742)

Electric heating

Heating devices

With power supply and voltage or current regulation or...

C219S497000, C219S121430, C118S725000, C156S345420, C392S416000

Reexamination Certificate

active

06563092

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to methods and apparatus for measurement and control of temperature and more particularly for measurement and control of a substrate temperature within a semiconductor process chamber.
2. Description of the Related Art
Photoresist layers are typically applied and patterned over surfaces of a substrate prior to the formation of features during the manufacture of semiconductor devices. Upon completion of these processes the patterned photoresist must be removed through photoresist stripping or ashing. Quite often a photoresist asher removes a photoresist layer by reacting free radical oxygen atoms with the resist material at an elevated temperature. In addition, the photoresist asher often incorporates a microwave or radio frequency (RF) plasma generator to produce free radical oxygen atoms and oxygen ions, which in turn strip the photoresist through high temperature oxidation. The temperature of the substrate directly impacts the rate of removal of the resist during the process. Photoresist stripping is employed primarily after implant operations where selected areas of a substrate are implanted or “doped” by elements such as boron and phosphorous to define the transistors in an integrated circuit. In another application, patterned photoresist defines regions of dielectric or metal layers that must be removed to form the interconnect wiring that constitute an integrated circuit (IC).
The introduction of new materials and the shrinking of feature sizes in integrated circuits are requiring the development of new photoresist formulations and processes. Tight temperature control of the substrate in an ashing chamber is critical as devices on the substrate can become damaged if the temperature overshoots a set point temperature. On the other hand, the process may not run efficiently if the temperature is not high enough. For example, photoresist stripping rate has been measured to increase by approximately 300 Å per minute for each 1° C. change in substrate temperature. One skilled in the art would appreciate that an inaccurate temperature measurement and control can have disastrous consequences for overall device yield. In ashing chambers where the substrate is being heated by lamps, closed loop temperature control is applied to accurately maintain a temperature set point. In a closed loop temperature control system, the temperature of an object, such as the substrate, is measured and the feedback from the temperature measurement is used by a power control system that controls the intensity of a heating source to increase or decrease the temperature.
FIG. 1A
displays a block diagram
100
representing a prior art closed loop controller for the temperature of a substrate in an ashing chamber. In diagram
100
, a sensor
104
measures the temperature of a substrate
102
. A signal corresponding to the temperature measured by the sensor is sent to the controller, which in turn controls the intensity of heat lamps
108
according to the difference in values of the temperature of the substrate and a set point temperature.
The sensor
104
of diagram
100
is a critical component of the control loop since it needs to provide accurate temperature measurement and must be capable of withstanding the harsh environment of the ashing chamber.
FIG. 1B
illustrates a detailed view of one prior art sensor employed to measure substrate temperature, where lamps are used to heat the substrate. This sensor consists of a thermocouple bead
111
contacting the substrate through an aluminum pad
103
where the aluminum pad
103
and thermocouple is attached to a pin supporting the substrate
102
.
FIG. 1B
illustrates a diagram of the contact between the substrate
102
and a thermocouple sensor pad
103
. Since the backside surface of the substrate and the aluminum pad surface are not completely smooth, a gas gap
110
exists at the interface of the two surfaces. The gas gap
110
exists even though substrate support extension
101
provides a gimballing effect to support pad
103
against the backside surface of the substrate since the corresponding surfaces are not completely smooth. The gas gap
110
causes inaccuracies in temperature measurement especially under the operating condition for ashing processes as is explained further below. The probe body
105
encases thermocouple wires
109
a
and
109
b
, which are routed through a high vacuum seal since the chambers typically operate under low pressures below 2 torr.
As mentioned above, the ashing process is performed at a low process pressure, typically below 2 Torr. Therefore, very little gas exists in the gas gap
110
to conduct the heat from the substrate to the contact pad. As a result of the gas being evacuated from the gas gap
110
, the effective thermal conductivity is low, which in turn makes it difficult to accurately measure the temperature of the substrate. This method of temperature measurement requires detailed calibration of each individual sensor. Such calibration is normally performed using instrumented substrates. Consequently, the accuracy of such calibration is dependent on the reproducibility of the quality of the substrate-pad contact. Additionally, the characteristics of thermal interaction between substrate and pad vary with both pressure and substrate temperature. This necessitates detailed calibration over an extensive range of temperatures and operating pressures. Furthermore, the contact between the substrate and the aluminum pad is different for each substrate, thus injecting additional variables into the temperature measurement process, not to mention the poor calibration resulting from the substrate-to-substrate inconsistencies. The aluminum pad also oxidizes over time, thereby changing the characteristics of the pad for each process, which in turn further throws off the calibration.
Another type of sensor used in ashing chambers is an optical emissivity sensor. Processing chambers, where the substrate is supported by an RF-excited platen to generate the plasma, cannot use thermocouples since the wires of the thermocouple act as antennas. Because the high voltages induced in the thermocouple lead wires can damage sensitive electronic circuitry to which the thermocouple wires are connected, unshielded thermocouples are typically not used in the presence of RF-excited platens. Therefore, an optical emissivity sensor may be used to measure the temperature of the substrate to avoid this antenna effect.
FIG. 1C
illustrates block diagram
112
representing a prior art processing chamber employing an optical emissivity probe
124
to measure the temperature of the substrate
102
. The chamber
126
includes a microwave source
114
and a radio frequency (RF) source
118
. When the chamber
126
is operating in the microwave mode, i.e. microwave source activated and RF source deactivated, pins
128
lift the substrate
102
off of the platen
120
. In the RF mode the substrate
102
rests on the platen
120
. Here, the sensor assembly, including the contact pad
122
, is used as one support in conjunction with the pin
128
when the substrate is elevated for microwave processing. Typically, microwave processing is performed at elevated temperatures as high as 300° C. for which lamps
108
are employed.
The optical sensor
124
of
FIG. 1C
measures temperature of the substrate
102
by detecting the emitted infrared radiation from the backside of the pad
122
which is in contact with the substrate
102
. Similar problems as encountered with thermocouples persist with the optical sensor
124
. The pad
122
for the optical sensor also oxidizes over time. Accordingly, the emissivity of the backside of the pad changes with time. In addition, the optical sensor
124
must be contained in a light-proof housing
123
. Since the lamps
108
emit high intensity radiation, any light leak through the housing
123
of the optical sensor
124
, could trigger the optical sensor to measure temperature of the lamps
108
, which is much high

Affiliated with

Also associated with

Rating

Measurement of substrate temperature in a process chamber... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Measurement of substrate temperature in a process chamber..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Measurement of substrate temperature in a process chamber... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3040057