DE19934299A1 - Method and device for calibrating emissivity-independent temperature measurements - Google Patents

Abstract

The invention relates to a method or device for calibrating temperature measurements that are carried out using at least one first radiation detector in order to measure one heat radiation emitted by at least one substrate. Said method consists of the following steps: at least one reference substrate carrying at least one material with a known melting point temperature is heated; the thermal radiation of the reference substrate is measured during heating and/or cooling following said heating; a measured value plateau occurring during the measuring process is compared with the known melting point temperature. A device for carrying out the inventive method is also disclosed.

Description

The invention relates to a method and a device for calibrating temperature measurements with little least a first radiation detector for measuring one thermal radiation emitted by at least one substrate be made.

Methods and devices of this type are for example se in connection with the manufacture of semiconductor sub known in a reaction chamber. There will be emissivity-independent methods preferred in which a heat radiation coming from a substrate in the we considerably independent of the respective emissivity substrate with the actual temperature of the sub strats related. Such a procedure ren compensates for emissivity differences between under different substrates. To achieve an emissivi measurement methods that are independent of the So-called "ripple technology", for example in the US-A-5 490 728 and the unpublished DE 197 54 386 A is described by the same applicant, and uses the cavity principle, in which, for example, a mirrored chamber from an object to be measured one side is complete, making an approximation to a cavity radiator with emissivity 1 is reached.

Another method that verifies the cavity principle is not prepublished in, for example described in DE 197 37 802 by the same applicant, which, to avoid repetition, the subject of this An report is made. In the known methods Wafers used on their top or bottom Thermocouple (TC-ThermoCouple) is glued on. The meas Experience shows that deviations from TC to TC are very high  small, as long as the TCs from wire pairs of the same series were manufactured. The scatter then lies in a Be range from approximately 1 to 2 ° K. Because of this low Scattering can be the temperature measurements on Emissivi Independence can be set, because it ins especially on a small spread between the TC's is coming. The measurement uncertainty of the absolute temperature value is, however, significantly higher, with the measurement uncertainties are at best in the range of 2-3 K, and in less good cases are above 10 K-20 K.

These measurement uncertainties result from differences factors. Among other things, the thermal voltage depends of the thermocouple in addition to the temperature and also from the alloy tion, but due to the manufacturing process certain variations subject to. Furthermore, between the TC and its Ver more provided multiple electrical connections that each form thermocouples at the same time, so that asymmetry trical transitions generate additional thermal voltage In addition, the gluing points of the TCs have egg absorbance other than that of the glue points giving wafer surface. The equilibrium temperature of the TCs is therefore not only a matter of thermal conduction a wafer and TC, which would be ideal; Temperature of the TC additionally by the radiant heat of the Lamp affects, which is why the temperature of the TCs often does not exactly match the temperature of the substrate.

The measurement uncertainties primarily arise due to sources of error that are the same for all TCs Way as measurement error. TCs are therefore measuring transducers that have a small spread, but a large one have absolute measurement inaccuracy.  

To achieve better measurement accuracy, the TCs in a TC calibrator, i. H. an oven with very ho internal temperature distribution with respect to a Absolute temperature measurement calibrated. Several uncalibrated te TCs were placed in the oven along with a reference TC introduced, the reference TC in turn at a separate calibration service using a transfer pyrome calibrated with respect to a primary reference. These multiple calibrations require different che devices, they are very expensive, and due the many steps result in numerous possibilities that errors are introduced into the calibration, which in turn results in measurement uncertainties in the end.

From US 5,265,957 a device and a method for calibrating a temperature sensor is also shown, in which a wafer with a plurality of calibration rinsing of a reference material is provided with a melting point in the range from 150 ° C to 550 ° C. During the heating of such a wafer, the effective reflectivity of the wafer is measured by the temperature sensor, and a first abrupt change in an output signal from the temperature sensor is equated with a temperature which corresponds to the melting point of the reference material. Then temperature sensor calibration parameters are calculated. The principle is illustrated in Fig. 6. The signal I of a temperature sensor (pyormeter) is recorded as a function of time t during the heating of the wafer. So that the reflectivity of the wafer changes at the phase transition of the reference material, the reference material must be arranged near the surface, so that it is in the penetration depth of the measuring wavelength. At the phase transition of the reference material occurs, as shown in Fig. 6, which he mentioned abrupt change in the pyrometer signal.

The method shown in US 5,265,957 has considerable disadvantages. For example, due to the sudden change in the pyrometer signal, the melting temperature T _{m1 can} not be assigned a clear pyrometer value, which results in a systematic measurement error ΔI for the calibration method.

The invention is therefore based on the object, a Ver drive and a device of the type mentioned create the calibration of Temperature measurements in a simple and inexpensive way can be carried out with greater accuracy.

The object is achieved by a Ver driving of the type mentioned resolved, which follow the process steps comprises: heating a little Mostly a reference material with a known melting point temperature bearing reference substrate on / or over the Melting point temperature; Measuring the thermal radiation of the Re reference substrates during the heating and / or one of the Warming following cooling; Relating one measurement plateaus occurring during the measurement process with the known melting point temperature.

By heating the one on the reference substrate Reference material the temperature of the reference sub rises strats and the reference material until they reach the Melting point temperature of the reference material reached. On When the melting point temperature is reached, the tempera rises Do not continue until the reference material is complete has passed from the solid to the liquid phase, until the latent heat is added to the reference material has been. When it cools down, this process reverses known way around. Since the melting point temperature of the Re reference material is known exactly, can during the  Warming and / or a cooling following the warming measured plateau therefore with the known Melting point temperature can be related where by simply calibrating an absolute temperature measurement is achieved.

The measured value plateau is advantageously during the Heating and / or cooling of the reference substrate determined. However, the determination of the measurement is preferred value plateaus made during the cooling as the Reference material in the molten state before the full constant solidification in particularly good thermal contact to the reference substrate.

The method according to the invention has compared to that US 5,265,957 described above significant Advantages on. Since the process is not subject to changes in the Emissivity of the reference substrate is dependent possible the reference material with a thick protection layer, or the reference material inside of the reference substrate. On an order in close to the surface can be dispensed with. this has the advantage that contamination of the process chamber through the reference material can be avoided. However, this is the basic requirement for a broad application of the Re reference substrates in semiconductor technology.

Another advantage results from the occurrence of a measured value plateau at the known melting temperature, as is shown schematically in FIG. 7. In contrast to the method described in US Pat. No. 5,265,957, the method according to the invention makes it possible to clearly assign a radiation detector signal to a melting point temperature due to the plateau. The systematic error ΔI mentioned above is minimized in the method according to the invention.

The formation of a plateau also has the advantage that a thermal equilibrium is formed between the reference material and the reference substrate during the time t _p , thereby minimizing possible temperature differences. It is also advantageous that optical properties of the surface of the reference substrate do not change in the method according to the invention. If, for example, the reference substrate (with the exception of the reference material located inside) is selected from the same material as the substrates to be processed later (e.g. Si), the reference substrate has the same emissivity as these substrates. As a result, the calibration of the radiation detector can be used directly when processing substrates, even without emissivity compensation. Only if the substrates z. B. due to different surface properties have a different (spectral) emissivity, an emissivity compensation or correction is to be performed, as shown below. If the substrates can be processed without emissivity compensation, this also has the advantage that only one radiation detector is required for the temperature measurement. Such a temperature measurement method with only one radiation detector is not possible with the method described in US Pat. No. 5,265,957, since in this method the emissivity of the surface of the reference substrate changes at the phase transition.

So that the plateau can be easily measured, should the mass of the reference material is at least 1% of the total mass of the reference substrate. This minor mas  can be chosen because the specific Melting heat is a multiple of the specific heat capacity act is. So that the optical properties of the The surface of the reference substrate does not change, the Protective layer of the reference material preferably at least equal to three times the optical attenuation length that the Protective layer for the measuring wavelength of the radiation detector tors has been chosen.

In order to achieve an emissivity-independent temperature measurement range, the reference substrate is heated ge according to a particularly preferred embodiment of the Er invention with at least one radiant heating source, the emitted radiation with at least one characteristic modulated chemical parameters, advantageously active mod is at least one of the radiation emitted radiation with at least one second radiation detector is determined, and wherein the radiation determined by the first radiation detector to compensate for the reflection from the reference substrate radiation from the radiation source by the second radiation detector determined radiation corri is greeded.

Due to the characteristic known modulation of the Radiation source it is possible to distinguish between those of the Ge subject itself emitted radiation, which for the Er averaging the temperature of the reference substrate required Lich, and the reflected radiation of the radiation to distinguish source. For further advantages and individual units regarding the modules known as the ripple technique tion and evaluation of the radiation from the radiation source is not previously published on US-A-5 490 728, as well as published DE 197 54 386 A from the same applicant  taken to the extent that the content of the present Be be made.

In this way it is possible regardless of knowledge A calibration based on the emissivity of the reference substrate to achieve with respect to the absolute temperature. To do that Measuring system for independence regarding the emissivi It can be pre-calibrated using TC wafers be briered, since it is primarily a low Scattering arrives between the TCs.

The modulation advantageously becomes a characteristic tion of the radiation emitted by the radiation source when correcting it from the first radiation detector averaged radiation, creating a sub separation of the radiation from the actually to be measured Radiation emitted by the reference substrate itself is particularly simple, reliable and quantitative nau is made possible.

The radiation emitted by the radiation source is preferably amplitude, frequency and / or phase mod liert. Depending on the existing conditions and requirements The choice of modulation type can be selected, whereby the Modulation type especially with regard to the one Expertise and reliability of the modulation process but also the evaluation method and the detection method rens is selectable.

According to a preferred embodiment of the invention the radiant heating source consists of several lamps, whereby at least modulates the radiation from one of the lamps is, but advantageously the radiation from all Lam pen is modulated.  

The degree of modulation or the Mo depth of control controlled, whereby he or she is known is, and the detection and evaluation is simplified.

The emissivity independence of the temperature measurement can can also advantageously be achieved in that the Measurement of the thermal radiation emitted by the substrate one side of the substrate, which is at least one Forms part of a cavity radiator.

According to a further preferred embodiment of the Er are several reference materials, each with un different melting point temperatures on the Refe renzsubstrat provided, and it will be during the heating cooling and / or cooling the respective measured value plateaus determined, and each with one of the known enamel point temperatures related.

The present invention is of great advantage for Ka calibration of a temperature measurement in connection with a device for the thermal treatment of substra ten in an oven in which the substrates quickly and with the most accurate, predetermined temperature curve possible be heated and cooled, applicable.

The object is also achieved by egg ne device for calibrating temperature measurements of the type mentioned solved the following indicates: a reference material with a known melting point temperature attached to a reference substrate a radiation source for heating the reference substrate, its emitted radiation by means of a modulation preferably active with at least one character teristic parameters can be modulated, and at least a second radiation detector, for measuring the of the  at least one radiation source emitted, and means for correcting the first Radiation detector determined radiation, for compensation on the radiation of the Radiation source through that of the second radiation detector determined radiation.

By using a reference substrate with a attached reference material with known Melting point temperature in combination with the active Mo In turn, the advantages mentioned above result le emissivity-independent simple calibration a temperature measurement.

According to one embodiment of the invention, the front towards a cavity radiator that is at least partially is formed by the substrate.

Preferably, the cavity radiator is ver mirrored chamber formed, one wall at least is partially formed by the substrate. Advantageous The cavity radiator can also be inserted through a plate are formed, which are arranged parallel to the substrate is.

According to a preferred embodiment, the Reference substrate provided a cover that between itself and the reference substrate a chamber for receiving of the reference material. Preferably the cam sealed from the environment to prevent that the reference material is the device to be calibrated contaminated. To ensure even pressure (partial pressure of the reference material) in the chamber, this is advantageously evacuated, which is also the case brings with it that no liquid reference material under the pressure of a heated residual gas by possibly  existing micro cracks is pressed outwards. Advantage the entire reference substrate is unfortunately glazed.

The reference substrate advantageously has at least a recess for receiving the at least one reef reference material, whereby the bottom of the reference sub strats as thin as possible in the area of the reference material is a temperature difference between the bottom of the reference substrate and the reference material if possible to keep low. For stability reasons, several Recesses preferred. To mechanical stresses on due to different expansion coefficients between to avoid the reference substrate and the reference material the walls of the recesses are preferably abge slants.

The reference substrate advantageously has the same che size and / or shape and / or the same weight as the substrate, the temperature of which after calibration should be measured to the reference substrate with a existing handling system for the substrate to be measured to handle. In this way a calibration or a temporary recalibration can be automated. The There is an advantage over conventional TC substrates not only in increasing the accuracy and re producibility, but also that a TC substrate not traded automatically due to the connection wires and the wires after installing the TC- Substrate must be clamped. This fact falls away with the described reference substrate.

This is advantageously at least one reference mat rial a metal, as for metals, especially highly pure ones Metals, the melting point temperatures are known very precisely or are defined.  

According to a further embodiment of the invention different reference materials with differ Chen melting point temperatures on the reference substrate provided for a calibration at different Ensure temperatures.

At least one reference material is advantageously present different locations on the reference substrate orders, which can ensure that the Refe the field of view of the first radiation detector covers.

Advantageously, the reference substrate carries little least a reference material on one of the radiation side facing away from the detector. To ensure high emissivity of the Reaching reference substrates are advantageous on a side of the radiation detector facing the Reference substrate structures, such as micro channels le, provided.

To primary references with different emissivities a variety of reference substrates can be obtained with different optical layers are provided. By provide primary references with different Emissivities arise on a possibility fully maneuverable pre-calibration with TC wafers to breed.

Because of the relatively light manufacturing and ther mixing properties of ceramic material is the Reference wafers are preferably made from the same.

The present device is of great advantage Calibration of a device for fast thermal  Treatment of substrates with an oven in which the sub strate quickly and with the most accurate possible the temperature curve is heated and cooled, suitable.

The invention is described below in connection with the Example of a device for heating semiconductors wafers explained with reference to the figures. It demonstrate:

Fig. 1 a longitudinal section through a rapid heating for the treatment of semiconductor wafers in a schematic representation;

Fig. 2 is a cross section along the Fig. 1 section line II-II,

Figure 3 is a plan view Renz substrate. An inventive Refe, wherein a cover Ab was omitted for clarity,

Fig. 4 is a sectional view of the Refe rence inventive substrate with cover,

Fig. 5 Renz substrate is an enlarged partial sectional view of the Refe,

Fig. 6 shows a temperature-time curve for a Referenzsub strat according to the prior art,

Figure 7 is a temperature-time curve for the fiction, modern reference substrate.

Fig. 8 strat equipped with a thermocouple reference substrate.

The embodiment of a rapid heating furnace for treating semiconductor wafers 2 shown in FIGS . 1 and 2 shows a reaction chamber 1, preferably made of quartz glass, with a semiconductor wafer 2 located therein. The reaction chamber 1 is surrounded by a housing 3 which has lamps 4 , 5 at the top and bottom, the radiation of which is directed onto the reaction chamber 1 . A schematically illustrated pyrometer 6 (cf. in particular FIG. 2) with a large entrance angle measures the radiation emitted by the semiconductor wafer 2 , as well as the radiation of the lamps 5 reflected on the semiconductor 2 , which are formed as rod lamps in the exemplary embodiment shown. An arrangement of this type is described, for example, in the unpublished DE 197 37 802 A or the unpublished DE 197 54 386 A of the same applicant, so that in order to avoid repetition, reference is made to it, and the applications in this respect to the content of the present description be made.

Another pyrometer 7 receives the light emitted by the lamps 5 directly via optical lines or light channels 8 . In order to avoid repetitions with regard to this so-called lamp pyrometer 7 and the arrangement for irradiating the lamp pyrometer 7 with the light from the lamp 5 , reference is made to the unpublished DE 197 54 385 A by the same applicant, which in this respect makes the content of the present application ge.

FIGS. 3 to 5 show an inventive Refe rence substrate in the form of a reference wafer 10, which is used to calibrate the rometers Py shown in FIGS. 1 and 2 6. The reference wafer 10 has a flat, round shape which essentially corresponds to that of a semiconductor wafer 2 to be treated. The reference wafer 10 has a base body 12 made of ceramic material which has a circular elevation 13 in the central region. In the area of the elevation 13 circular recesses 15 are provided for receiving a reference material in the form of molten metal inserts 17 with known melting point temperature. In order to achieve better surface coverage, the recesses can also be hexagonal (honeycomb-shaped) and arranged over the entire cross section of the base body 12 . In order to avoid tension between the base body 12 and the molten metal insert, a metal tablet is placed loosely in the recess before first use, which always solidifies with a maximum permissible diameter after the first melting. As it cools down further, the metal contracts faster than the base body. With repeated heating, the metal melts just before mechanical stresses can occur.

As can be seen in Fig. 3, 19 circular recesses 15 are provided, in each of which there is a metal melt insert 17 with a known melting point temperature. The number of recesses 15 is, however, irrelevant to the invention and it can be adapted to the stability requirements of the base body 12 or other parameters. For example, it is possible to provide only one recess for receiving the molten metal insert 17 , care being taken to ensure that the recess is in the field of view of the pyrometer 6 . According to further embodiments, different metal melt inserts with different melting points are provided within the recesses instead of metal melt inserts 17 with the same, known melting points.

Above the elevation 13 and the molten metal inserts 17 there is a cover 20 which is connected in its edge region 22 to the base body 12 in a sealing manner. This connection can be made in a known manner by gluing, welding or another method known in the art. From the cover 20 and the elevation 13 or the Metallschmelzeinla conditions 17 , a chamber 23 is formed which is evacuated. Instead of providing a cover, it is also conceivable to glaze the entire reference wafer or to provide a combination of these two features.

Although the reference substrate was described using a specific exemplary embodiment, it should be noted that the reference substrate can have a different shape and configuration without departing from the inventive concept. For example, the elevation 13 in the central region could be omitted, and the recesses 15 could be formed directly in the base body 12 . Instead of providing the molten metal inserts 17 in recesses in the base body 12 , it is also possible to apply them flatly to the base body without recesses. Structuring the surface of the base body facing away from the reference material to increase its emissivity would also be conceivable.

To calibrate the pyrometer 6 of the above device for the treatment of semiconductor wafers, the reference wafer 10 is introduced into the reaction chamber 1 by means of an existing handling device (not shown in more detail), in the same position as the semiconductor wafer 2 shown in FIG. 1. The reaction chamber is subsequently closed and the reference wafer 10 is heated by means of the lamps 4 , 5 , the pyrometer 6 measuring the radiation emitted by the reference wafer 10 and the radiation from the lamp 5 reflected thereon. During the heating, the temperature of the molten metal insert rises in the same way until it reaches its melting point. From this point in time, the temperature of the molten metal insert does not continue to rise. Due to the good thermal conductivity and the resulting rapid temperature adaptation between the molten metal insert 17 and the reference wafer 10 , the temperature of the reference wafer 10 does not rise further until the molten metal insert 17 has completely changed from the solid to the liquid phase, ie until the molten metal insert 17 the latent heat was applied. The constant temperature is measured as a plateau of measurement by the pyrometer 6 and device by a Vorrich not shown with the known melting point temperature in relation.

After the molten metal insert 17 has completely melted and no longer absorbs latent heat, the temperature of the reference wafer 10 continues to rise. Thereupon the heating is stopped and the reference wafer 17 is cooled or cools down.

The process is reversed when it cools down. The metal melt insert 17 cools down to the solidification point and its temperature then remains essentially constant until the insert 17 has given up its latent heat completely and is again in the solid phase. This in turn results in a plateau of measured values in the pyrometer measurement, which can be related to the known melting point of the insert 17 .

The determination of the Meßwertplateaus during cooling is advantageous because the insert 17 to stand in the molten better thermal contact with the reference substrate 10 has, and hence the temperature equalization between the insert 17 and the reference substrate 10 is faster.

Since the measured heat radiation from the reference substrate can change depending on the emissivity of the substrate, to achieve emissivity independence during the process described above, the radiation emitted by the lamps 4 , 5 is preferably active and in a defined manner modulated, and the radiation emitted by the lamps 5 is measured directly with the lamp pyrometer 7 . The output signals of the pyrometers 6 and 7 are fed to an evaluation circuit, not shown. It determines the Conference wafer from Refe 10 radiation emitted by, sets the falling on the pyrometer 6 radiation, consisting emitted from the wafer 10 and reflected radiation with the detected radiation by the pyrometer 7 in relation ship. This is possible because the radiation emitted by the lamps 5 is modulated in a known manner. This modulation is also included in the radiation picked up by the pyro meter 6 , so that by comparing or relating the degrees of modulation and / or the depth of modulation of the radiation picked up by the pyrometers 6 and 7 , compensation of the lamp radiation reflected by the reference wafer 10 in the radiation captured by the wafer pyrometer 6 is possible. As a result, the ratio between the radiation emitted by the wafer 10 and the reflected radiation can be determined in order to enable the emissivity of the wafer 10 to be determined. Due to the emissivity, the known melting point temperature of the insert 17 can now be related to the radiation emitted by the reference wafer 10 in order to provide a calibration with respect to the absolute temperature.

Such a modulation and determination of the one Radiation emitted is, for example, in the unpublished DE 197 54 386 A same An notifier, and described in US-A-5 490 728, so that referenced to avoid repetition  becomes, and these publications to the extent of the content of the present description are made.

To achieve an essentially emissivity independent measurement instead of the one described above Ripple technology, for example, the so-called cavity Principle are used in which by using a mirrored chamber is attempting a cavity beam imitate.

A dimensioning proposal for the reference substrate is given below. In order to ensure that the reference substrate heats up as uniformly as possible, it is preferably dimensioned such that the thermal mass per unit area over the substrate is constant. This can e.g. B. can be achieved by varying the thickness of the reference materials so that the following equation is satisfied:

d ₁ .ρ ₁ .c ₁ + d ₂ .ρ ₂ .c ₂ + d ₃ .ρ ₃ .c ₃ = d ₄ .ρ ₄ .c ₄ (1).

Here, d ₁ , d ₂ , d ₃ and d _{4 are} the thicknesses of the reference substrate on the molten metal insert, the thickness of the molten metal, the thickness of the cover of the molten metal or the thickness of the reference substrate in areas without a molten metal insert. The associated densities and heat capacities are denoted by ρ ₁ , ρ ₂ , ρ ₃ , ρ ₄ and c ₁ , c ₂ , c ₃ , c ₄ .

The power density P (power / area) z. B. an RTP system and the heating rate R (ram rate) can be simplified using the following equation:

d ₄ .ρ ₄ .c ₄ .R = P (2).

This relationship defines the required power density P at a predetermined heating speed R in order to heat the reference substrate homogeneously (see equation (1)). When the reference material reaches its melting point, the (area-related) energy E _L , which results from the following equation, is required to melt it:

E _L = d ₂ .ρ ₂ .c _s (3)

where c _{s is} the specific heat of fusion of the reference material. If the power density P of the RTP system is kept unchanged, the plateau time t _p (see FIG. 7) can be estimated using the following equation:

t _p = E _L / P (4)

where E _L and P can be seen from equations (2) and (3), respectively. This allows a plateau-time ramp rate product to be specified:

Z = t _p .R = d ₂ .ρ ₂ .c _s / (d ₄ .ρ ₄ .c ₄ ) (5).

This product contains only the material-specific sizes and the respective layer thicknesses. This allows for a given product Z using equation (5) and equation (1) when choosing z. B. d ₁ , calculate the thickness of the molten metal d ₂ and the thickness of the reference substrate d ₄ at the edge region. For simplification, d ₃ = 0 is assumed here. However, d ₃ = d ₁ ge can also advantageously be selected, in particular when c ₁ = c ₃ . Is z. B. d ₁ = 1 mm (or d ₁ + d ₃ = 1 mm at c ₁ = c ₃ ), so for a reference substrate made of silicon with a plateau time ramp rate product of 100 K for the melting metals Germa nium or aluminum melting metal thicknesses d ₂ of 0.071 mm or 0.2 mm, the outer region of the reference substrate having thicknesses d ₄ of 1.074 mm or 1.3 mm.

The proposed dimensioning examples clearly show the thickness ratios of the reference substrate and reference material under the process condition of a constant thermal mass per surface element (equation (1)). Furthermore, the plateau time t _{p can be} estimated if a ram rate R is specified. If Z, as in the example described, is 100 K, the plateau time t _{p is} approximately 10 seconds if a ram rate R of 10 K / s is used.

Another example of application of the reference substrate 12 is shown in FIG. 8. Here, the reference substrate is additionally equipped with at least one thermocouple TC (TC ₁ , TC ₂ , TC ₃ , TC ₄ ), which is attached to the reference substrate using conventional means (adhesives). This results in the possibility of measuring the thermocouple TC with respect to the reference substrate and determining possible influences of the fastening means of the thermocouple TC (gluing or cementing) on a TC temperature determination.

The invention was previously based on a preferred management example explained. However, those skilled in the art are Refinements and modifications possible without there is left by the idea of the invention. That invented The method according to the invention is in particular also together hang with other devices or emissivity-independent current measuring method as the previously described with before partly usable to a reliable with simple means  sig, reproducible calibration of temperature measurements gene.

Claims (37)

1. A method for calibrating temperature measurements, which are carried out with at least one first radiation detector ( 6 ) for measuring heat radiation emitted by at least one substrate ( 2 ), characterized by the following method steps:

• a) heating at least one reference material ( 17 ) with a known melting point temperature carrying the reference substrate ( 10 ) on and / or above the melting point temperature;
• b) measuring the thermal radiation of the reference substrate ( 10 ) during the heating and / or a cooling following the heating; and
• c) Relating a measurement plateau occurring during the measurement process to the known melting point temperature.

2. The method according to claim 1, characterized in that for heating the reference substrate ( 10 ) little least a radiation source ( 4 , 5 ) is used. 3. The method according to claim 1 or 2, characterized records that the temperature measurement emissivity is independent. 4. The method according to any one of the preceding claims, characterized in that the radiation emitted by the at least one radiation source ( 5 ) with at least one second radiation detector ( 7 ) is averaged that the radiation emitted by the at least one radiation source ( 5 ) with little least one characteristic parameter is modulated, and that the radiation determined by the first radiation detector ( 6 ) is compensated for compensating for the radiation from the radiation source ( 5 ) reflected by the reference substrate by the radiation determined by the second radiation detector ( 7 ). 5. The method according to claim 4, characterized in that the modulation for characterizing the radiation emitted by the radiation source ( 5 ), is used in the correction of the radiation from the first radiation detector ( 6 ) he averaged. 6. The method according to claim 4 or 5, characterized in that the radiation from the radiation source ( 5 ) given abge is amplitude, frequency and or phase modulated. 7. The method according to any one of claims 2 to 6, characterized in that the radiation source ( 5 ) is formed by a plurality of lamps, and that the radiation is modulated by at least one of the lamps. 8. The method according to any one of claims 3 to 7, characterized characterized in that the degree of modulation or the Mo dulation depth of the radiation is controlled. 9. The method of claim 1 or 2, characterized in that the measurement of the reference substrate (10) radiant heat emitted is performed on the side of the reference substrate (10) that faces ei nem cavity radiator. 10. The method according to any one of the preceding claims, characterized in that a plurality of reference materials ( 17 ) each with different melting point temperatures on the reference substrate ( 10 ) are seen before, and that the respective measured value plateaus determined during heating and / or cooling are in each case be related to one of the known melting point temperatures. 11. The method according to any one of the preceding claims, characterized in that the reference substrate for Calibration of a device for thermal loading act of substrates is introduced into this. 12. A device for calibrating substantially emissivity-independent temperature measurements, which are carried out with at least one first radiation detector ( 6 ) for measuring the heat radiation emitted by at least one substrate ( 2 ), characterized by at least one reference material ( 17 ) with a known melting point temperature, the a reference substrate ( 10 ) is attached; at least one radiation source ( 4 , 5 ) for heating the reference substrate ( 10 ); at least one second radiation detector ( 7 ) for measuring the radiation emitted by the at least one radiation source ( 5 );
a modulation device for modulating the radiation emitted by the at least one radiation source ( 5 ) with at least one characteristic parameter; and
a device for correcting the radiation determined by the first radiation detector ( 6 ) on the basis of the radiation determined by the second radiation detector ( 7 ). 13. The apparatus according to claim 12, characterized in that the radiation source ( 4 , 5 ) is formed by a plurality of lamps, and that the radiation can be modulated at least egg ner of the lamps. 14. Device according to one of claims 12 or 13, characterized in that the degree of modulation or the depth of modulation of the radiation emitted by the radiation source ( 4 , 5 ) is controllable. 15. The apparatus according to claim 12 or 13, characterized through a cavity radiator that at least partially is formed by the substrate. 16. The apparatus according to claim 15, characterized in net that the cavity radiator through a mirror te chamber is formed, in which at least one Wall area is formed by the substrate. 17. The apparatus according to claim 15, characterized in net that the cavity emitter in a space between the substrate and one parallel to it arranged plate is formed. 18. Device according to one of claims 12 to 17, characterized by a cover ( 20 ) located above the reference material ( 17 ) on the reference substrate ( 10 ). 19. The apparatus according to claim 18, characterized by a between the cover ( 20 ) and the reference substrate ( 10 ) defined chamber ( 23 ). 20. The apparatus according to claim 19, characterized in that the chamber ( 23 ) is sealed from the environment. 21. Device according to one of claims 19 or 20, characterized in that the chamber ( 23 ) is evacuated. 22. Device according to one of claims 12 to 21, characterized in that the reference substrate ( 10 ) is glazed. 23. Device according to one of claims 12 to 22, characterized in that the reference substrate ( 10 ) has at least one recess ( 15 ) for receiving the at least one reference material ( 17 ). 24. The device according to claim 23, characterized in that the walls of the recesses ( 15 ) are chamfered. 25. Device according to one of claims 12 to 24, characterized in that the reference substrate ( 10 ) has the same size and / or shape and / or the same weight as the substrate ( 2 ), the temperature of which is measured after calibration shall be. 26. Device according to one of claims 12 to 25, characterized in that the at least one reference material ( 17 ) is a metal, in particular a highly pure metal. 27. The device according to one of claims 12 to 26, characterized in that on the reference substrate ( 10 ) different reference materials ( 17 ) are provided with different melting point temperatures. 28. Device according to one of claims 12 to 27, characterized in that at least one reference material ( 17 ) is arranged at different locations on the reference substrate ( 10 ). 29. Device according to one of claims 12 to 28, characterized in that at least one reference material ( 17 ) on a side facing away from the first radiation detector ( 6 ) of the reference substrate ( 10 ) is brought to. 30. Device according to one of claims 12 to 29, characterized in that one of the first radiation detector ( 6 ) facing side of the reference substrate ( 10 ) has structures, in particular microchannels. 31. The device according to any one of claims 12 to 30, characterized by a plurality of reference substrates ( 10 ), each with different optical layers. 32. Device according to one of claims 12 to 31, characterized in that the reference substrate ( 10 ) is made of a ceramic material. 33. Device according to one of claims 12 to 32, there characterized by that the device for potash bration of a device for rapid thermal Treatment of substrates in an oven, suitable is. 34. Reference substrate for temperature calibration with white at least a reference material of known enamel point temperature, located inside the reference  substrate, the reference material being at least 1% the mass of the reference substrate. 35. Reference substrate according to claim 34, characterized records that the reference material by a sub strat material with a thickness of at least three opti attenuation lengths of a selected world length is surrounded. 36. Reference substrate according to claim 35, characterized records that the thickness is at least 3 microns. 37. reference substrate according to one of claims 34 to 36, characterized in that the thermal mass per Area unit constant across the substrate is.