JPH08285692A - Semiconductor processing technology including measurement ofradiated and heated main body by pyrometer and equipment forexecuting technology thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate restriction on the arrangement of probe by locating two probes on the boundary of a processing chamber such that a first probe includes no lamp and a second probe includes the field of view of first probe and a lamp. SOLUTION: A wafer 120 is set in a processing chamber 160 while being surrounded by a quartz envelope 130 and a lamp 140 is disposed on the side bank of envelope. Two probes are mounted on one side of the chamber 160 while facing the openings closely each other. More specifically, inlet face of a wafer probe 100 is directed toward the wafer 120 which is included in the field of view Ω1 of an optical baffle 150 for limiting the field of view and the lamp 140 is excluded therefrom. A lamp probe 110 is arranged on a plane 170 and provided with a field of view Ω2 including the entirety of wafer 120 in the field of view Ω1 of probe 100 and the lamp 140 located on the side face of the field of view Ω1 . This arrangement eliminates restriction on the arrangement of probe, e.g. arranging and removing of prove.

Description

Detailed Description of the Invention

[0001]

FIELD OF THE INVENTION This invention relates to the field of pyrometry, and more particularly to pyrometric observation of radiantly heated semiconductor wafers inside a reactor or furnace designed for rapid thermal processing.

[0002]

[Related Application Information] This application was filed on April 2, 1993, and was filed on April 19, 1994 in US Pat.
US Patent Application No. 08 / 0,5,416
No. 42,028, a continuation-in-part of U.S. patent application Ser. No. 08 / 227,844, filed April 14, 1994, the disclosure of which is incorporated by reference.

[0003]

In rapid thermal processing (RTP), workpieces such as semiconductor wafers are subjected to specific temperature cycles of arbitrary complexity. For this reason, RTP is useful in performing heat-dependent processes such as diffusion and annealing during integrated circuit manufacturing. However, some of these processes require temperature control within a range of ± 10 ° C or less. Such precise control is possible only when the temperature of the wafer can be measured with relatively high accuracy.

Optical pyrometry is one useful method of controlling the temperature of the wafer during RTP. High temperature measurement technology 1
As of October 13, 1992, C.I. W. Schi
US Pat. No. 5,154, assigned to etinger et al.,
No. 512. An overview of this technique is shown in FIG. In accordance with this technique, a first light pipe probe 10 having an entrance opening facing the wafer 20 is provided and a second probe 30 having an opening facing one of the opposing row 40 of lamps is provided. The row of lamps is typically a straight row of quartz / tungsten / iodine lamps located outside the processing chamber 50. First probe 10
Extracts a sample of the radiation diverged and reflected by the wafer and sends the sample of the extracted radiation to detector 60. The second probe 30 extracts a sample of the radiation emitted by the lamp and sends the sample of the extracted radiation to the detector. The probe 30 receives both the direct radiation from the lamp and the reflected radiation from the reflector 80. Since the emissivity ε of the wafer is estimated from the signal of the probe, the temperature of the wafer is inferred from Planck's law of radiation regarding the heat dissipation rate of the wafer, the emissivity of the wafer, and the temperature T of the wafer.

As mentioned above, the signal of the first probe is the sum of the divergent and reflected radiation. The divergence from a lamp lit by alternating current has an alternating component called "ripple", so that enough information is available to distinguish between the divergent and reflected components. Since the heat dissipation from the wafer has no significant AC component, the reflectivity of the wafer is estimated as the ratio of the ripple amplitudes of the signals of the first and second probes, respectively. This reflectivity is estimated and then corrected to obtain a value for the heat dissipation of the produced wafer.

[0006]

However, the techniques described above provide an undesirable constraint on probe placement. One probe must only "see" the wafer and the other probe only "see" the lamp, since the AC signal of the probe is used to directly determine the reflectivity. I won't. In the prior art pyrometer of FIG. 1, the light pipe probe 10 is constructed and arranged to look only at the wafer 20, and the light pipe probe 30
Is configured and arranged to look only at the light source 40. This provides some constraints on probe placement. First
First, as shown in FIG. 1, the probes 10 and 30 project into the chamber. Particles are generated during the RTP treatment,
The generated particles adhere to the probe. As a result, the probe must be removed and cleaned, which is a difficult and time consuming process. Furthermore, since this process is sensitive to the position of the probe relative to the wafer and lamp, the probe must be reinserted in exactly the same position it was in before it was removed. Otherwise,
The pyrometer must be readjusted. Also, due to the special constraints of the RTP enclosure 50, the probe 10,
Providing space for 30 is difficult.

As a result, there is a desire for an optical pyrometry process and instrument that does not have the above constraints on probe placement. While useful, this technique requires the probe to be located near the wafer, or at least inside the oven enclosure. The probe is designed to
Directed to collect the radiation reflected from the wafer.

[0008]

SUMMARY OF THE INVENTION In the process and apparatus of the present invention, there are at least two controls for heating a wafer in a chamber.
Optical pyrometry is performed by using two optical probes. The first probe is positioned to collect the radiation emanating from the wafer, but not the direct radiation from the heating lamp. The radiation collected by the first probe is controlled by positioning the first probe in a manner that limits the radiation that can be collected. In one embodiment, the first probe is located at the boundary of the processing chamber. The first probe has a field of view such that the wafer is at an angle Ω ₁ forming the field of view of the first probe. The first probe is positioned so that the lamp does not enter the field of view between the first probe and the wafer.

In order to limit the field of view of this probe in a suitable manner, it is advantageous if the first probe does not extend into the chamber. In this embodiment, the first probe is located behind an iris diaphragm located on the inside surface of the oven chamber. The iris diaphragm limits the field of view of the first probe (referred to as the iris probe in this example) to the elongated portion of the wafer and the portion of the quartz tube interposed between the wafer and the iris diaphragm. The first probe is located at the boundary of the chamber and the radiation collecting part of the probe extends very little into the chamber.

The second probe is also located at the boundary of the chamber. However, this probe has a wider field of view than the first probe. In the context of the present invention, a wide field of view means
It is meant that the field of view of the second probe forming an angle Ω ₂ with the probe tip includes at least the field of view of the first probe and the lamp illuminating the field of view Ω ₁ of the first probe. As a result, the signals collected by the second probe are radiation from the wafer, quartz tube, lamp, and chamber interior surface. Advantageously, the radiation collecting portion of the probe does not extend into the chamber, but in other embodiments at least some of the radiation collecting portion of the probe extends from the wall of the chamber into the chamber. In one embodiment of the invention, the field of view formed by Ω ₂ is sampled by a lamp probe. In another embodiment, the image of the field of view formed by Ω ₂ reflected from the wall of the chamber in close proximity to the lamp probe is sampled.

Both the first and second probes are directed towards the heated object. The field of view of the second probe is the first
Since it includes the field of view of the probe, it is advantageous if the probes are located close to each other, i.e. the two probes are close to each other in the chamber. In this relationship, 2
It is advantageous if the two probes are fixed to the same wall in the chamber. This improvement makes it possible to measure the wafer temperature normally at a normal processing temperature with an accuracy of ± 10 ° C. or higher. An additional refinement is the detailed modeling of the radiation environment within the processing chamber leading to a more accurate interpretation of the signals from the first and second probes.

Accordingly, one embodiment of the process of the present invention is
Exposing the body to a controllable flux of electromagnetic radiation from one or more lamps illuminated by an alternating current, measuring the surface temperature of the body, and controlling the flux of electromagnetic radiation in response to the temperature measurement. By heating the body. The temperature measurement collects, in part, the thermal radiation transmitted from the body by the first probe and a portion of the reflected lamp radiation, and a portion of the lamp radiation transmitted in the direction of the second probe. Performed by collecting and detecting at least a portion of the radiation collected by the first and second probes in response to the first and second radiation signals. The temperature measurement is
Furthermore, it includes determining the amount of the time-varying component due to the time change of the lighting current with each of the first and second signals. Measuring temperature further comprises combining at least these quantities according to a mathematical formula from which the temperature can be inferred.

[0013]

DETAILED DESCRIPTION OF THE INVENTION Illustrated in FIG. 2 is an RTP reactor having a wafer probe 100 and a lamp probe 110 arranged in accordance with a generally preferred embodiment of the present invention. Wafer 120 is enclosed within a processing chamber having a quartz envelope 130. The numbers indicate the two probes located outside the envelope 130. In another arrangement, one or both probes are placed within the envelope. As shown, the lamps 140 are installed in banks of two opposite sides of the envelope 130. (In another reactor design, only a bank of lamps on one side is used.) Each individual lamp is typically cylindrical, and lamp 140 is shown in cross section in FIG.

Entrance surface 150 of wafer probe 100
Are directed toward the wafer. As shown, the wafer probe 100 is positioned so that the lamp is not in its field of view, which is Ω ₁ . Wafer probe 100
Is a light pipe probe with a field-limiting light baffle 150, such as an iris diaphragm, with a diameter of 1.5 mm. Wafer probe 100 is, for example, an elongated region approximately 1 cm wide and 2 cm long,
The emission of the wafer from a relatively small spot on the surface of the wafer 120 is captured. Wafer probe 100 is preferably located near gap 175 between a set of consecutive lamps in a bank of lamps.

Since the lamp is represented in cross section, Ω
₁ , Ω ₂ is actually rotated 90 ° with respect to the lamp 140, as shown in FIG. That is, Ω ₁ , Ω ₂
Are actually parallel to the direction of the individual cylindrical lamps 140. However, a constraint on Ω ₁ and Ω ₂ , namely Ω ₁ and Ω
Lamp 140 is not for the portion of such wafer _1,
At least one lamp for Ω ₂ means
It still controls the placement of the probes 100, 110.

Enclosure 160 at least partially encloses a quartz envelope 130 and a lamp 140. The enclosure 160 has an inner surface 170.
AG Asso, Sunnyvale, CA
The Model 4100 Heat Pulse Reactor, available from Ciates, is one example of a suitable reactor. The majority of the reactor surface 170 is gold plated as originally supplied. Surface 170 is typically diffusely reflective. However, in other embodiments, surface 170 is specular.

The lamp probe 110 is preferably located on the surface 170. The probe 110 is the probe 100.
Has a field of view Ω ₂ that includes both the entire portion of the wafer 120 within the field of view Ω ₁ and the lamp 140 flanking Ω ₁ .
In one embodiment of the invention, lamp probe 110
Directly detects the radiation from the field of view formed by Ω ₂ . In another embodiment, the lamp probe is a lamp
The radiation is detected indirectly by detecting the image in the field of view Ω ₂ that is reflected from the chamber wall 170 proximate to the probe 110. Unlike the prior art, the lamp probe 110 uses the wafer probe 10
Wafer probe 10 collecting radiation from the same direction as
There is nothing between 0 and the lamp probe 110. However, the probe 110 has a wider field of view within the chamber than the probe 100.

In a generally preferred embodiment, the probe 110
Is a diffusion window light pipe with a diameter of 1.5 mm. In the context of the present invention, a diffusion window is a substance that transmits light but not necessarily images. It is advantageous if the probe 110 extends a short distance into the enclosure 160.

FIG. 3 shows in more detail the portion of 200 represented in FIG. As shown in FIG. 3, the probe 100
Opening 21 in wall 160 of the oven enclosure
Located at the bottom of 0. This opening 210 is used for the probe 100.
Limit the field of view. The probe 110 is the enclosure 1
Located in the opening 220 in the wall of 60. Opening 220
The purpose of is to avoid contact between the tip of the probe 110 and the wall 160 of the oven enclosure. The probes 100, 110 are optically connected to gold-plated channels 211, 212, respectively. The gold-plated channels 211, 212 have corresponding optical filters 21.
5, 225 through the corresponding photoelectric detectors 230, 23
Optically connected to 1. An example of a suitable photoelectric detector is the EG of Montgomeryville, PA.
& Indium arsenide (I
nAs) photoelectric detector (part number 420040).
Photoelectric detectors 230 and 231 are water-cooled heat sinks 232.
And is electrically connected to the print board 233. The printed board shown in FIG. 3 is equipped with two preamplifier detectors 234 and a thermistor preamplifier 235. Another printed circuit board 240 is provided, which has two ac / dc amplifiers 241 and an electrical connector 245 for transmitting signals to a data processor (not shown). The photodetector, filters, and printed circuit board are all within the magnetic shield 250.

The method is predicated on the fact that the wafer does not carry the radiation of the lamp. This condition is met if the temperature of the wafer is above 600 ° C., and the silicon wafer substrate is such that the resistivity of the silicon is below about 0.01 Ωcm (for example, the concentration of dopant is at least about 2 × 10 5). This condition is met if enough dopant is incorporated (to ¹⁵ dopant atoms / cm ³ ).
One example of a suitable dopant is a p + dopant such as boron. This condition is also met if the wafer is coated with a metal film that is opaque to the lamp radiation. Titanium is an example of such a metal film.

The probe signals (from the wafer probe 100 and lamp probe 110) are used to control the heating of the wafer in the chamber. It should be noted that the variation of the lamp radiation in the reactor is more complex than the simple ripple (ie the AC signal component of the lamp), which is twice the power supply frequency. In addition to harmonic fluctuations in the power supply frequency, there are, for example, fluctuations in the power of the lamp caused by the feedback circuit controlling the temperature of the wafer and non-harmonic fluctuations due to the finite reaction time of the lamp. As a result, it is preferable to calculate the AC components of these signals from the time-dependent first and second moments. Since the thermal radiation from the wafer contains very little AC component, the AC component of these signals represents the AC component of the wafer. That is, the optical signal from the probe is directed to a photodiode amplifier and the output of the amplifier is fed to an analog-to-digital converter (ADC) for digitized signal V ₁
Derive (t) (corresponding to the signal from the wafer probe 100) and V ₂ (t) (corresponding to the signal from the lamp probe 110). The sampling time t is usually
The interval δ is 0.2 ms. The amplifier has an analog filter of about 1 ms to cut the spurious signal noise due to the digitization.

One component of the digitized signal V ₁ from wafer probe 100 is the reflected lamp radiation (ie, the lamp radiation reflected from the wafer).
The reflected lamp radiation is also reflected by the lamp probe 11
It is also the ratio R of the digitized signal V ₂ from zero. This is expressed as a partial derivative.

[Equation 1] Usually, the AC power source that lights the heating lamp has one or more phases or groups. For example, A, B, C
If there are groups of three lamps connected to the phases of the three power supplies, then the variation component is a multiple of the period of the first harmonic of the AC power supply divided by the number of phases, ie, a quantity

[Equation 2] Is shifted by 0, 1 or 2 times,
Here, f is a power supply frequency (for example, 60 Hz in the United States). The ratio of V ₁ (t) due to the lamps of group A is A _1, and the ratios due to the remaining phases are B ₁ and C ₁ . Similarly, the ratios of V ₂ (t) due to the lamps of each phase are A ₂ , B ₂ , and C ₂ . The ratio is such that groups of lamps are selectively illuminated individually, and for each group V ₁
It is determined by measuring the AC components of (t) and V ₂ (t).

In general, the ratios are not the same for the two detectors. That is, A ₁ / A ₂ ≠ B ₁ / B ₂ ≠ C _{1 /}
Since it is C ₂ , different raw 3-phase patterns appear in the raw signals for the two detectors. Therefore, it is advantageous to suppress this difference by computer-analyzing the signal containing the periodicity associated with one phase. That is, the three phases are mathematically analyzed into one phase for further calculation. This is estimated by replacing the raw signal with its instantaneous value and a weighted sum of the two initial pre-time values delayed by τ _p and 2τ _p . This analytic signal is represented by V ₁ '(t) and V ₂ ' (t) and is given by the following algebraic expression.

(Equation 3)

[Equation 4]

The weighting factors a ₁ , b ₁ etc. are determined by the matrix inversion and the linear algebra of the ratios A ₁ , B ₁ etc. of the groups of ramps. The equations for the coefficients a ₁ , b ₁ , c ₁ of equation (3) are given by:

(Equation 5)

(Equation 6)

(Equation 7) This time,

(Equation 8) Is. The coefficients a ₂ , b ₂ , c ₂ of equation (4) are determined in a similar manner.

The next step is the calculation of the variation components in the analyzed signal to calculate the ratio R as expressed in equation 1. This ratio is obtained by first constructing the partitions of V ₁ and V ₂ , and the least squares fit is used to obtain the diagonal lines of this relationship. The statistical weight given to each point of least-squares fit decreases exponentially as the point gets older. That is, the old points are given less statistical weight than the new points. The formulas for making this calculation will be described sequentially below. First, the reference signals U ₁ , U ₂
Is calculated by the recursive filtering method from the input signals V ′ ₁ , V ′ ₂ using the following formula.

[Equation 9]

[Equation 10] In formulas 9 and 10, δ ₁ is the time between signal samples and τ ₁ is a time constant (usually 1 ms) approximately equal to the time of the AC radiation produced by the heating lamp. These numerical values have a relation of τ ₁ > δ ₁ .

The fluctuation component of the signal is then calculated as the difference between the input signal and the reference signal. The calculation of R described above is performed using a least squares fit of the diagonals, where the four intermediate terms are the moving averages of the lines, algebra, and crossing terms.

[Equation 11]

(Equation 12)

(Equation 13)

[Equation 14] In the above equation, δ ₂ is approximately equal to δ ₁ and τ ₂ > τ ₁ . Normally, τ ₂ is about 10 ms.

The above terms are then combined and averaged in the following way.

(Equation 15)

[Equation 16] In these equations, δ ₃ is approximately equal to δ _1, and τ ₃ > τ
₂ The slope is then calculated by

[Equation 17] Here, δ ₄ is greater than or equal to δ ₃ , τ ₄ is approximately equal to τ _3, and τ ₄ > δ ₃ .

The quasi DC components of the digital signals V ₁ and V ₂ are
Obtained from the averaged values filtered using the following inductive formula.

(Equation 18)

[Formula 19] Where τ ₅ is almost equal to τ ₄ . Quasi DC component of the digitized signal V ₁ was expressed in S _1, the quasi-DC component of the digitized signal V ₂ is represented by S _2. Then equation 9
To 19 are larger time constants, for example, τ ₁ '= 1.
25τ ₁ is used to recalculate the values u ′, v ′, w′X ′. The factor 1.25 is empirically determined.

Next, the values of w and w'are combined to determine R, and the values of X and X'are combined to form S ₁ , S ₂ as follows.
To decide.

(Equation 20)

[Equation 21]

[Equation 22] The combination of terms with and without'in equations (20), (21), (22) corrects for the time delay inherent in inductive filtering. The coefficient of 2 for the term without'is determined empirically.

The pyrometer scale is adjusted to ensure that the temperature of the heated body is determined accurately in the manner described above. The relationship with the wafer temperature is expressed by the following equation.

(Equation 23) Here, the adjustment factors f ₁ and ε (R) (effective radiation amount) are determined by the adjustment, λ is the wavelength of the detector, T is the wafer temperature, and h, c, k _B are basic constants of Planck's radiation law. is there.

Equation (23) provides the definition of ε (R). The relationship between ε (R) and R is almost a straight line for ε (R) larger than 0.3, and is expressed by the following equation.

[Equation 24] Here, f ₂ is the second adjustment coefficient. Adjustment factor f ₁ ,
The functional relationship between f ₂ and ε (R) and R is obtained by using a thermocouple to measure the temperature of the test wafer in a fast thermal anneal oven heated by a lamp. A minimum of two wafers with different R measurements and corresponding ε (R) greater than 0.3 are used to determine the coefficients f1, f2 needed to make this adjustment. Once the sensitivity of the sensor is known as a function of ε (R), the sensor is used to monitor and control the heating of the wafer.

At low temperatures, the radiation emanating from the wafer is
The equation S ₁ -RS ₂ is close to zero when the heating lamp is first ignited, since it is negligible compared to the reflected lamp radiation. However, in the technology using two sensors, due to defects such as non-uniformity of the output of the heating lamp, R
There is an error in the calculation of. To correct these defects, R is multiplied by the empirical factor F. F is determined by the following formula before the wafer is heated.

(Equation 25) In calculating R as described above, it is advantageous if the wavelength of the light detected by the photoelectric detector is from about 1 μm to about 3 μm. It is advantageous if a filter transmitting radiation with a wavelength of about 2 μm to about 3 μm is inserted between the probe and its corresponding detector. For example, it is advantageous if the filter has a transmission frequency band of about 2.4 μm to about 2.5 μm and the detector detects signals in that frequency band. One example of a suitable detector is an indium arsenide photodiode.

Signals in this wavelength range contain relatively little interference from oven heating lamps and quartz components. About 1μ
Wavelengths shorter than m contain noise from the reflected lamp radiation. Wavelengths longer than about 3 μm are affected by absorption and divergence of signals in this wavelength range by the quartz components of the oven. As a result, it is advantageous if the wafer probe and lamp probe are equipped with optical filters that transmit only signals in this wavelength range to the corresponding sensor. In one embodiment, the radiation that passes through the iris diaphragm for the wafer probe and the diffuser opening for the lamp probe is transmitted by a light guide to a filter that adds only infrared radiation to this wavelength range. After the signal has been transmitted through the filter, conventional detectors, such as InAs photodiodes cooled to about -30 ° C, convert radiation in the wavelength range into current. Such current is further converted into a voltage signal by the impedance conversion amplifier. To increase the amplitude of the AC signal and digitize the amplified AC component signal separately from the DC signal, it is advantageous to use analog circuits. These voltage signals are then transmitted to a computer programmed to carry out the calculations described above.

Example 1 A set of 12.5 cm diameter silicon wafers made of various different materials to exhibit emissivity ranging from about 0.2 to 0.9. Coated with. The film was one or more materials such as polysilicon, silicon dioxide, titanium silicide. Thermocouples were bonded to the shallow recesses of the wafer using zirconia-alumina refractory cement. The wafer was then placed in an oven and heated in the presence of the two probes as previously described. Data from thermocouples and probes were recorded when the wafers were annealed. The probe was equipped with an infrared filter so that only radiation in the wavelength range from 2.4 μm to about 2.5 μm was detected by the photoelectric detector.
The sensitivity factors f ₁ , f ₂ were obtained from the temperature calculated from equations (23), (24) and the best fit of the thermocouple temperature. The solid line shown in FIG. 4 is the linear relationship given by equation (24).

Once this tuning curve is obtained for a particular RTP system, the tuning curve is used to control the RTP processing of this system. The linear region of the adjustment curve is used for wafers with ε (R) greater than 0.3. The calibrated pyrometer is then used to control the temperature of the body during the process using the above equation, as previously described.

The silicon wafer (135) is coated with a 60 nm thick film of titanium, approximately 650.
Annealed at a temperature of ° C for 1 minute. Since the electrical resistance of titanium silicide films depends on the processing temperature, the electrical resistance of the film was measured after processing was completed to determine the temperature at which the wafer was processed. The amount of temperature determined using the method of the present invention that differs from the temperature determined from the resistance of the silicide is found in the bar graph of FIG. The average error is 3.3 ° C and the standard deviation is 3.
It was 6 ° C. As shown in Figure 5, the temperature difference between the two techniques was small.

[Brief description of drawings]

FIG. 1 is a schematic diagram of a prior art RTP system.

FIG. 2 is a schematic diagram of an RTP system including a set of radiation-sensitive probes according to the present invention.

FIG. 3 is a detailed view of a portion 200 of FIG.

FIG. 4 is a tuning curve of an RTP system obtained by using wafers with various backside emissivity.

FIG. 5 is a bar graph of temperature error determined by a series of experiments comparing temperature determination according to the method of the invention with temperature determined from the electrical resistivity of a heated wafer.

Claims (12)

[Claims] 1. A process of heating a body, exposing the body to a controllable flux of at least one electromagnetic radiation lit by a time-varying current, measuring a surface temperature of the body, and temperature. Controlling the flux of radiation in response to the measurement, the measuring step comprising: a) collecting the radiation emitted and reflected by the body at the first probe and detecting said radiation to be V _1. Directing the signal of the first probe; b) collecting radiation emitted and reflected by the body and at least one lamp at the second probe, detecting said radiation, and making it the signal V ₂ of the second probe. and directing a and determining the AC component of the c) V ₁ and V _2, and determining the DC component of d) V ₁ and V _2, the R from the AC component of e) V ₁ and V ₂ Judge And f) so that the surface temperature of the body is calculated from R to ε
(R) determining step; g) in step (a), the first probe effectively samples the radiation from the region of the body for the solid angle Ω ₁ of the first probe; and h) step ( In b), the second probe effectively samples the radiation from the region for the solid angle Ω ₂ including the region of the body for the solid angle Ω ₂ and at least one lamp. 2. The process of claim 1 wherein the radiation of step (b) comprises radiation directly incident on the second probe from the lamp. 3. The body has at least a first major surface, and in the step of exposing, the radiation of the lamp impinges on at least the first major surface and the diffusive reflective surface of the first major surface. 2. Located in close proximity, the lamp is located substantially in a plane intermediate the diffusive reflective surface and the first major surface.
The process described in. 4. The process of claim 1, wherein the probe is conditioned before being used in the process. 5. The process of claim 4, wherein the probe is tuned by determining the sensitivity of the probe as a linear function of ε (R). 6. The radiation emitted by the probe is polyphase and the signals V ₁ , V ₂ are analyzed before these signals are used to calculate the ratio R. process. 7. The process of claim 6, wherein R is determined by calculating the linear function dependence of the analyzed signals V ₁ , V ₂ . 8. An oven containing at least one lamp for establishing at least one adjustment factor, wherein the probe is later used to calculate ε (R) as a function of S1, S2, R. The process of claim 5, wherein the process is adjusted by measuring the temperature of the test wafer. 9. An apparatus for heating a body, comprising: an enclosure, a quartz envelope disposed within the enclosure, and a receiving body disposed within the quartz envelope for receiving a heated body. Mounting surface, multiple lamps located between the inner surface of the enclosure and the quartz envelope, and does not extend into the enclosure substantially,
Although against _1, the lamp positioned so as not part for Omega _1, installed in the first passage of the enclosure wall, a first radiation probe having a field of view defined by the solid angle Omega _1, substantially the enclosure Has a field of view defined by the solid angle Ω ₂ installed in the second passage of the wall of the enclosure, which does not extend into the interior of the enclosure, but is located in the part of the body for Ω ₁ and at least two lamps as for Ω ₂ . A second probe; a means for raising the temperature inside the enclosure; a detection means for detecting the signals from the first and second probes;
Equipment including. 10. The two probes are from about 1 μm to about 3 μm.
10. The device according to claim 9, which is equipped with an optical filter that blocks the transmission of signals with wavelengths outside the μm range to the detector. 11. The instrument of claim 9, wherein the field of view of the first probe is limited by a light baffle. 12. The device of claim 11, wherein the light baffle is an iris diaphragm.