Method for setting up CVD reactor

Abstract

The invention relates to a method for predicting a change in a value ([lambda] i) of a layer characteristic or a local variable on a support location (5 ') on which a layer is deposited on a substrate (7). If the independent processing parameter (qi) of one bearing position (5 ') varies, the value of the independent parameter (qi based on the bearing position (5') varies, and the value (lambda i) at the other bearing position (5 ') also varies. The method suggests depositing a starting layer, i.e. Formula (I), with independent starting parameters, and depositing a test layer, i.e. Formula (I), with different test parameters therefrom. The sensitivity matrix (Si, j) can be formed from determined starting values, formulae (II), (II) and test values, formulae (III), (III), the elements of which represent, respectively, the effect of the variation of each individual parameter (qI) on each value ([lambda] i) of the layer characteristic or local variable. A correction parameter ([delta] qi) can also be calculated by reversing a sensitivity matrix (Si, j) in such a way that an independent parameter (q) is preset and a preset target value is achieved by means of these independent parameters, i.e. Formulae (IV), (IV), # imgabs0 #

Description

Field of the Invention

[0001] The present invention relates to a method for predicting changes in target values of layer properties of a layer or a layer sequence deposited on a plurality of substrates at support positions with different positions in a processing chamber, or changes in target values of local variables affecting layer growth at the support positions. The present invention also relates to a method for setting up a device for simultaneously depositing a layer or a layer sequence on a plurality of substrates in a processing chamber respectively, wherein gases are fed into the processing chamber according to preset processing parameters and / or a temperature is set in the processing chamber, and the processing parameters include independent parameters that can be changed individually, and these parameters are individually corresponding to different substrates.

[0002] The present invention also relates to a device having a control device for simultaneously depositing a layer or a layer sequence on a plurality of substrates respectively, and a method for simultaneously depositing a layer or a layer sequence on a plurality of substrates respectively. Background Art

[0003] US2018 / 0340259 A1 describes a process for simultaneously depositing a plurality of layers on a plurality of substrates. Different gases can enter the processing chamber from different azimuth positions by an intake member, and the bottom of the processing chamber is composed of a rotating base, on which a plurality of substrates are arranged around a rotation axis. The substrates can be heated by a zoning heating device, wherein the zones extend circumferentially around the rotation center. In order to optimize layer uniformity, especially layer thickness, a sensitivity matrix is formed in terms of processing parameters for feeding energy to the heating zones, and the processing parameters are changed through the sensitivity matrix.

[0004] DE 102018101173 A1 describes a process for optimizing layer properties of a layer sequence deposited on a substrate. A coefficient matrix is established, and an inverse matrix is thus established.

[0005] WO 01 / 90434A2 also describes the setting of a sensitivity matrix to optimize the deposition of layers on substrates.

[0006] WO 02 / 092876A1 describes a device for depositing a layer on a substrate. The influence of changes in processing parameters on layer performance is determined in a pre-test. A correction value is formed from the resulting inverse function.

[0007] US2016 / 0336215 A1 describes a calibration method for correcting processing parameters, wherein a sensitivity matrix is established and an inverse matrix is thus established.

[0008] DE 102019104433 A1 and DE 102020107517 A1 describe such a CVD reactor. The susceptor forming the bottom of the processing chamber symmetrically carries a plurality of substrates around its center, and process gas is introduced through an intake element arranged at the center of the processing chamber to coat the substrates. For this purpose, the susceptor is heated from below by a heating device. The heating device is a cooled RF coil. A temperature-regulating gas can be introduced between the RF coil and the bottom side of the susceptor, wherein the temperature-regulating gas is introduced such that it individually influences the net heat transfer from the heating device to the substrates. Thus, the heat flow to each substrate can be independently changed with the aid of independent parameters.

[0009] DE 102018124957 A1 also describes this type of CVD reactor. The susceptor forming the bottom of the processing chamber symmetrically carries a plurality of substrate holders around its center, and the substrate holders are placed on an air cushion and carry one or more substrates. The air cushion is generated by a temperature-regulating gas that is fed into a recess of the susceptor that houses the substrate holders. The heat conduction from the heating device to the substrates is individually influenced by the height of the air cushion or the thermal conductivity of the temperature-regulating gas. The temperature-regulating gas flow can thus be individually set and changed with the aid of independent parameters.

[0010] DE 102014104218 A1 and 102020123326 A1 describe a CVD reactor in which a purge gas can be individually fed into a preparation zone upstream of the substrates, and the flow of the purge gas can be adjusted with the aid of independent parameters.

[0011] It has been observed that a change in one of these independent parameters not only causes a change in the value of the layer properties of a layer or layer sequence deposited using the changed independent parameter, but also causes a change in the value of the layer properties of a layer or layer sequence deposited on another substrate. Layer properties refer to the properties of a layer or layer sequence deposited on a substrate. The layer properties can be, for example, the layer thickness preset by the growth rate of the layer, which can be measured at the deposited layer. Another layer property can be the layer composition. The layer can be doped, for example, or composed of more than two components, so the layer composition can be characterized by the amount of dopant embedded in the layer or the elemental ratio forming the layer. The layer can be a single crystal layer of a compound semiconductor having more than two components, such as a GaAlN layer, where the ratio of Al to Ga can depend on the independent parameter. However, for example, when manufacturing a vertical cavity surface emitting laser (VCSEL diode), the layer properties can also be the properties of the layer sequence. The Bragg reflector is here deposited separately by a large number of layers, where the layer thickness and layer composition have a significant influence on the wavelength of the VCSEL diode. It has been observed that the slightest temperature difference or the slightest difference in the growth rate on adjacent substrates in the processing chamber already causes an intolerable deviation between the wavelength and the predetermined wavelength. Changes in the cooling gas flow or the air cushion gas flow of the substrate carrier bracket have an impact on the growth of the layers on adjacent substrates, because the independently changed gas flow can cause uneven pressure in the processing chamber or the dilution effect to take effect. These cross-dependencies are regarded as disadvantageous.

[0012] However, each individual parameter assigned to a support position can also be used to influence the local variables at the support position, where such local variables can be understood as technology-related environmental variables, such as temperature, especially the substrate temperature, the flow rate of the gas, or the partial pressure of the process gas. If the local variables at the support position are purposefully changed by the independent parameter, this also causes a change in the local variables at another support position. The local variables respectively affect the deposition of the layer. Summary of the Invention

[0013] The technical problem to be solved by the present invention is to provide measures that can reduce these cross-dependencies. A method for setting a CVD reactor is provided, which can minimize the deviation in the value of the layer properties of the layers deposited on adjacent substrates. This also applies to the deviation between local variables. A method is also needed by which the target values of the layer properties or local variables at the support position can be predicted in the case of a change in one or more individual parameters. The technical problem to be solved by the present invention is to improve the method for predicting the change in values by means of a sensitivity matrix described in the above-mentioned prior art, to provide a method for setting a device for simultaneously depositing layers or layer sequences at support positions with different positions in a processing chamber based on this, and to characterize the device for this purpose.

[0014] The stated technical problem is solved by the invention as set out in the claims, where the dependent claims are not only advantageous extended designs of the invention in the parallel claims, but also independent solutions to the stated technical problem.

[0015] According to the invention, the method mentioned at the beginning is improved as follows, namely that the processing parameters consist of parameters that individually correspond to each support position. These independent parameters each have the same effect at the support position assigned to them, i.e. for example, a change in the processing parameter has the same effect on the layer properties at this support position as at each other support position. However, a change in the independent parameter also affects the value of the layer properties at another support position.

[0016] The starting point of the invention is the recognition that a change in an independent parameter is not only a change in the value of the layer properties of a layer or layer sequence deposited with the changed independent parameter, but rather that such a change in the independent parameter causes a change in the value of the layer properties of a layer or layer sequence deposited on different substrates, or based on the recognition that a change in a local variable due to a local change in an independent parameter also causes a change in the local variable at other support positions. The independent parameter can here be any processing parameter that can be used to individually influence the value of the substrate layer properties. The independent parameter can be the value of the purge gas flow, the value of the temperature regulating gas flow, the mechanical position of a body influencing the temperature in the processing chamber, the mass flow of the precursor, provided that these processing parameters can be changed independently for at least some substrates or some substrate holders each carrying at least one substrate. The mass flow of the precursor can be used to influence the layer composition, in particular the dopant incorporation. The independent parameter can in particular be any mass flow or energy flow, such as the heating power. The value of the layer properties can be the layer thickness, layer composition depending on the growth rate and growth duration as mentioned above, or the wavelength depending on the layer thickness and layer composition. The aim of the method is firstly to be able to predict to what extent a change in an independent parameter that acts mainly locally also affects the target values of the layer properties or local variables even at other support positions. Another aim is to give individual correction parameters, individual target parameters or individual correction factors, for example by which the independent parameters provided in the recipe are corrected, such that the above-mentioned cross-dependencies are largely reduced.

[0017] First, it is mainly recommended to conduct a pre-test, in which cross-reactions are quantified. In the first pre-test, a starting layer or a starting layer sequence can be simultaneously deposited on a plurality of first substrates with a first set of independent starting parameters. Each independent starting parameter preferably has the same value for all substrates or substrate racks. The value of the layer property is determined at the layer or layer sequence deposited in this way. The determination of the value of the layer property can be carried out outside the processing chamber. However, in-situ measurements can also be carried out during the deposition process, for example, by observing the light wavelength with a spectrometer. However, in a variant, instead of the layer property, local variables such as the surface temperature of the substrate or the surface temperature of the substrate holder can be measured at each support position during the deposition of the layer. In at least one second pre-test, a test layer or a test layer sequence is also respectively simultaneously deposited on a plurality of second substrates using a second set of independent parameters, i.e., test parameters. The test parameters differ from the starting parameters by at least one value. Except for one test parameter, the values of the other test parameters are preferably the same. The difference between the test parameters and the starting parameters can be that only one test parameter corresponding to a unique substrate or substrate rack is different from the starting parameters, and the remaining test parameters are the same as the starting parameters. When the substrates or substrate racks are symmetrically arranged in the processing chamber, only one second pre-test is required, in which only a single test parameter is different from the starting parameters. Otherwise, it may be necessary to conduct a second pre-test on the set of independent parameters, where each has an additional, but preferably only one, test parameter different from the starting parameters. Subsequently, if necessary, the layer or layer sequence deposited in one or more second pre-tests is measured outside the processing chamber, where the test value of the layer property is determined. However, in a variant, instead of the layer property, local variables such as the surface temperature of the substrate or the surface temperature of the substrate holder can be measured at each support position during the layer deposition process. Then, a sensitivity matrix is established from these test parameters and test values. The elements of the sensitivity matrix can be the quotient of the differences. If only one second pre-test is conducted in the case where the substrates or substrate racks are symmetrically arranged, only one second pre-test is required. This second pre-test provides the elements of one column of the sensitivity matrix. The elements of the remaining columns are generated by cyclic exchange. An element of the sensitivity matrix can be generated by taking the quotient. The quotient consists of the difference and the parameter difference, where the parameter difference is the difference between the test parameter and the starting value.

[0018] This difference can be obtained in different ways. In a preferred variant of the present invention, a test response is calculated, for which a first difference between the starting value and the average of all starting values is determined, and a second difference between the test value and the average of all test values is determined. The difference is the difference between the first difference and the second difference. However, the difference can also be the difference between the starting value and the test value, or include at least one of the two averages. The sensitivity matrix thus obtained forms the basis for predicting the change in the target value that occurs with the change of the independent parameter. Through this sensitivity matrix, for example, it is possible to predict not only the change in the target value at the support position corresponding to the local independent parameter, such as the growth rate of the layer deposited here or the temperature here. Through this sensitivity matrix, it is also possible to predict the change in the value of the layer property or the value of the local variable at each other support position, that is, how the independent parameter corresponding to the other support position affects the temperature or the growth rate of the deposited layer at the other support position.

[0019] The above method provides the basis for the following method, i.e., a method for a device for simultaneously depositing a layer or a layer sequence on substrates disposed at different support positions in a processing chamber. In a subsequent step, the sensitivity matrix is inverted. Subsequently, correction parameters, independent target parameters, or independent correction factors can be obtained through the inverted sensitivity matrix and the initial values of the layer characteristics. The independent parameters can be values of gas flow rate or heat flow rate directed to substrates disposed at different positions in the same processing chamber. However, the independent parameters can also be the positions of entities for influencing the gas flow or heat flow in the processing chamber, and multiple such entities can be provided, where these entities can individually correspond to different substrates or substrate holders and can be individually moved. According to a preferred design of the present invention, the pedestal disposed in the processing chamber has a plurality of support positions arranged symmetrically about the center for substrates or substrate holders, and the independent parameters correspond to these support positions, where these parameters affect the purge gas flow, heat flow, or process gas flow directed to the substrate holder or support position. The independent correction parameters can be obtained by multiplying the inverted sensitivity matrix by a vector having correction values. The correction values can be the difference between the starting value and the target value of the layer characteristics. For example, the target value can be a specific layer thickness, a specific layer composition, or a specific characteristic wavelength of a layer sequence (such as a Bragg reflector). However, the target value can also be a local variable, such as the temperature that can be measured at the support position, or the partial pressure or flow rate of the process gas. In a first pre-test, the independent parameters, such as the gas flow rate for generating an air cushion or the temperature-adjusting gas flow rate directed to a single substrate holder, can have the same value. The correction values are first determined by the foregoing method, in such a way that after the first pre-test, the difference is obtained between the starting value and the target value. The correction parameters are then obtained by determining the sensitivity matrix and inverting it. With these correction parameters, the starting parameters can be corrected so that the same layer characteristics can be achieved in a subsequent process of depositing a layer or a layer sequence with the starting parameters corrected in such a way. The local variables can also be set to a uniform value or adjusted individually by the said method. The independent correction factor can be the quotient of the sum of the starting parameter and the correction parameter and the starting value. This correction factor can then be applied to other independent parameters specified in the recipe, especially multiplied by them, to correct these other independent parameters, thereby reducing the tolerance of the layer characteristics. The above method is particularly applicable to operating the devices described in DE 102019104433A1 or DE 102018124957 A1. Thus, the independent parameters can be the gas flow rate or gas flow composition for generating an air cushion that supports the substrate holder, and the pedestal carrying the substrate holder is heated from below by a heating device. However, the independent parameters can also be the gas flow rate or gas flow composition for influencing the heat transfer between the heating device and the pedestal carrying the substrate holder.

[0020] In embodiments of the present invention, it may be stipulated that a change in an independent parameter not only changes one value, but simultaneously changes two values. For example, adjusting the temperature of a gas stream may affect the substrate temperature and the growth rate, or simultaneously affect the growth rate and the layer composition. This also applies to the gas stream that forms an air cushion for a substrate holder driven in rotation. The magnitude of the gas stream can affect the temperature, the growth rate, or the layer composition. There can also be multiple independent parameters that affect different values, such as temperature, growth rate, or layer composition. By successively applying the above method multiple times, correction parameters can be obtained for multiple independent parameters. However, it is also possible to start from the initial parameter set and conduct preliminary tests with different test parameter sets, where different independent parameters are qualitatively changed in different parameter sets. Thus, starting from the operating point of the system defined by the initial values, a sensitivity matrix can be generated respectively, which indicates the changes that occur in one or more values when the said or another independent parameter is changed.

[0021] The present invention also relates to a device for depositing a layer or a layer sequence on a plurality of substrates in a processing chamber. The device may include valves and mass flow controllers, which in turn correspond to a gas mixing device. A processing chamber may be arranged in a reactor housing. In the processing chamber, the base may have multiple support positions for a plurality of substrates. Process gases provided by the gas mixing device can be fed into the processing chamber through a gas inlet element according to a recipe stored in a control device. The control device is also arranged to independently direct the gas flow or the heat flow to different substrates or to one or more substrate holders carrying substrates respectively according to independent parameters specified by the recipe. The control device should have correction factors for correcting the independent parameters. These correction factors can be stored in the memory of the control device. The control device is also arranged to correct the independent parameters provided by the recipe with the correction factors. The control device may have a microcomputer or a microprocessor, which can be programmed through a program. The present invention also relates to such a program. The correction factors can also be elements of a matrix, but the matrix usually only has diagonal elements. The independent parameters can be multiplied by this correction matrix.

[0022] The present invention also relates to a method for depositing a layer or a layer sequence on a plurality of substrates, wherein the correction factors are determined and stored in the control device in the above manner. When performing the said method, the independent parameters that have been changed by the correction factors before are used.

[0023] A system for depositing a layer on a substrate uses independent parameters q i to operate, and changes in these independent parameters q i result in changes in the value λ i During the implementation of the method according to the present invention, the initial parameters can be interpreted as vectors.

[0024]

[0025] With these starting parameters In a first pretest with starting values The layer properties are determined, and these starting values can also be understood as vectors.

[0026]

[0027] Starting value Is different from the target value For example, the target value refers to a preset such as the wavelength that the Bragg reflector mirror should have as a characteristic. All target values Generally have the same value. However, the target value can also be a local variable at the support position where the substrate is supported in the processing chamber. For example, it can be temperature. However, the target value is generally also broadly understood as the value of the layer property set when adopting a specific set of starting parameters (vector) or the value of the local variable at the support position.

[0028] The following describes an example of the method to illustrate how to determine the correction value Δλ i .

[0029] The correction value Δλ is calculated according to these target values and starting values by the following formula i .

[0030]

[0031] The purpose of this method is to determine the target parameter Through these target parameters, a layer or a layer sequence can be deposited, and its layer property, that is, for example, the wavelength reaches the target value An intermediate goal of this method is to determine the correction parameter Δq i , which is used to calculate the target parameter according to the starting parameters For example, according to the following formula

[0032]

[0033] However, the method of the present invention also includes a related pre-stage, in which initially only the change of the value λ i Can be predicted.

[0034] In these second pretests, test parameters are used Which can also be understood as vectors.

[0035]

[0036] Here, all elements of the vector, except for one element in the vector, remain at the same value of the uniform parameter q0. Only one element of the vector differs from all other elements of the vector by a parameter difference Δq from the uniform parameter q0.

[0037] The uniform parameter preferably has the value of the starting parameter

[0038]

[0039] By measuring the layer or layer sequence deposited during the second preliminary test, the test value is then determined. These test values can also be understood as vectors. Similarly, local variables can also be measured in the second preliminary test.

[0040]

[0041] If the support positions for the substrate of the device are arranged asymmetrically, it may be necessary to conduct second preliminary tests corresponding to the number of support positions, in which different elements of the vector of test parameters differ from the uniform parameter. However, in the case of symmetrical arrangement, only one column of the test value matrix needs to be determined. The other columns are obtained by conjunction (Konjunktion) of the test values conducted in the above manner, i.e., by cyclic permutation.

[0042]

[0043] The sensitivity matrix S can be created from the test value matrix in various ways i,j , where the sensitivity matrix S i,j Each element of represents the change in the test value when an independent parameter changes.

[0044] In the first alternative, the average value of the starting values and the average value of the test values are used to create the sensitivity matrix S i,j .

[0045]

[0046] From these two average values the test response is obtained as follows

[0047]

[0048] The sensitivity matrix S can be calculated as follows from these test responses i,j :

[0049]

[0050] Alternatively, the sensitivity matrix S can also be calculated as follows i,j

[0051]

[0052] Or the sensitivity matrix S can be calculated as follows i,j

[0053]

[0054] Test parameters Preferably, they respectively correspond to the unified parameter q0

[0055] With the sensitivity matrix S created in this way i,j , the value λ can be predicted as follows, for example i

[0056] λ i ←S i,j Δq i (15)

[0057] Sensitivity matrix S i,j The correction value Δλ i and the correction parameter Δq i are associated as follows

[0058] Δλ j =S i，j Δq i (16)

[0059] By inverting the sensitivity matrix S i,j the inverse sensitivity matrix is obtained

[0060]

[0061] The correction parameter Δq i can be directly calculated from the correction value Δλ i Calculated

[0062]

[0063] From this, the correction factor k can be calculated as follows i

[0064]

[0065] With the correction factor k i the target parameter can be calculated as follows Wherein is the starting parameter preset by the formula

[0066]

[0067] These correction factors k can be multiplied with independent parameters preset by the formula during subsequent production processes i for the purpose of correcting them.

[0068] The present invention also relates to a method for predicting changes in two different target values wherein two or more target values can be, respectively, layer properties of a substrate at a plurality of different support positions arranged in a processing chamber, or local physical variables affecting layer growth, such as environmental properties at the support positions. It is also specified that the first target value can be a layer property, such as layer thickness or the wavelength of a Bragg mirror. The second target value can be a local variable at the support position, such as temperature, e.g., substrate temperature. To implement this variant of the method, the above method can be successively implemented multiple times. In the first variant, it can be specified that in a first pre-test, starting values of both the first value of the layer property or local variable and the second value of the layer property or local variable are determined by a set of starting parameters In other pre-tests, the independent parameter q is then changed in the above manner to obtain a first sensitivity matrix S i through which the change in the first value λ i,j when the first independent parameter q i changes can be predicted, and a second sensitivity matrix S' i is obtained through which the change in the second value λ' i,j when the second independent parameter q i changes can be predicted. Through this variant, for example, the change in layer thickness or layer growth rate with respect to the thickness of the air cushion, and the change in substrate surface temperature with respect to the temperature-regulating gas flow can be determined. i

[0069] It is also specified that two (or more) different parameters q i and q i simultaneously affect two or more different target values to different extents. These parameters can be, for example, the gas for rotating the substrate holder (rotating gas flow) and the temperature-regulating gas, where these two parameters can have different intensities of influence on layer thickness and temperature. The purpose of this method is also to consider, if necessary, only the influence of multiple parameters on one target value, such as the influence on layer thickness. For this purpose, a matrix S i,j is accordingly determined for the gas generating rotation, and a matrix S' i,jThereafter, for example, by a method (not the content of the method), the parameters of the temperature-regulating gas are determined / regulated such that a desired substrate temperature profile is achieved given an initial temperature deviation. Thereafter, using the matrix S′ i,j to predict the influence of changes in the parameters of the temperature-regulating gas on the target value of the layer thickness. Thereafter, when determining the correction value for the gas generating rotation using the matrix in Equation 18 this initially unintentional and undesired secondary influence of the temperature-regulating gas on the target value of the layer thickness can be taken into account by adding it to the correction value Δλ i and minimized by the correction parameter of the rotating gas flow.

[0070] Thus, it can also be provided that, in order to predict changes in layer properties or local variable values, in one or more other third preliminary tests, a second test layer or a sequence of second test layers is deposited simultaneously on a plurality of third substrates with a second set of test parameters corresponding to another independent parameter and different from the first test parameters. In the second preliminary test, for example, the mass flow rate of the gas generating the air cushion, which is located on the substrate support, can be changed, or the substrate support can be rotated. This value can be the layer thickness and / or the temperature. When depositing the second test layer, the mass flow rate of the temperature-regulating gas can be changed. It is also possible to determine the layer thickness and / or the temperature as values here. Then, second test values of the same layer properties, such as layer thickness or layer composition, or second local variables, such as the temperature of the substrate, are measured on the second test layer or the sequence of test layers. Then, a second sensitivity matrix S′ i,j is formed based on the second test values of the layer properties or local variables. By this method, the degree of influence of two different parameters, such as the mass flow rate of the gas generating the air cushion or the mass flow rate of the temperature-regulating gas, on the same layer properties or the same local variables can be predicted if necessary.

[0071] However, the above method can also be executed continuously multiple times, and the starting values are determined separately in the first preliminary test after each method step is completed. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Embodiments of the present invention will be described below with reference to the accompanying drawings.

[0073] In the drawings:

[0074] Figure 1 a top view of the base 3 of the CVD reactor is shown,

[0075] Figure 2 a cross-sectional view taken along line II-II according to Figure 1 is shown. DETAILED DESCRIPTION

[0076] An embodiment of an apparatus for performing a method for coating a semiconductor layer, in particular on a semiconductor substrate, has a reactor housing 1, which can be evacuated and which has a processing chamber 2 therein, and the housing can be made of refined steel.

[0077] Below the upper wall of the housing 1 there is a processing chamber top 14, which can be cooled in this embodiment, and for this purpose a cooling device 15 is formed by cooling channels. In the center of the processing chamber 2 there is a gas inlet element 9, which has a gas outlet for discharging process gases. The process gases can be hydrides of elements of Group V and organometallic compounds of elements of Group III. These process gases together with a carrier gas are fed into the processing chamber 2 from the central gas inlet element 9, and the carrier gas can be, for example, hydrogen. The process gases and the carrier gas flow radially from the inside out through the processing chamber 2. The gas outlet element 10 extends around the outer edge of the base 3. The exhaust gas can be pumped out of the processing chamber 2 through this gas outlet element 10 by a vacuum pump (not shown).

[0078] The bottom of the processing chamber 2 opposite the processing chamber top 14 is formed by the upper side 3' of the base 3.

[0079] On the upper side 3' of the base 3 there are a plurality of support positions 5' for the substrates respectively, and among them, the support positions 5' are arranged symmetrically around the center of the base 3. It is also feasible to arrange a plurality of substrates on each support position 5'.

[0080] In the embodiment shown in the figure there are a plurality of recesses 4, which have bottoms, and input lines 8 lead into these bottoms. In each recess 4 there is a substrate holder 5 for holding the substrate 7. By feeding gas into the input line 8, an air cushion 6 is formed between the bottom side of the substrate holder 5 and the bottom of the recess 4, and this air cushion causes the substrate holder 5 to be kept suspended and additionally drives it to rotate around an axis. In other embodiments of the present invention, the substrate 7 can also be placed directly on the base 3, so that the base 3 only has a plurality of support positions for the substrate 7.

[0081] A sealing plate 12 extends below the bottom side 3” of the base 3. A gap 13 is formed between the bottom side 3” and the sealing plate 12. The input lines 16, 17 lead into the gap 13 at different radial positions. The inlet portion 16' is located radially inside an arc line around the rotational axis 20 of the base 3, and the arc line extends through the radially inner edge of the recess 4. The second inlet portion 17' of the input line 17 is located below the recess 4. In other embodiments of the present invention, these inlet portions 16', 17 are not required, or only one of the inlet portions 16', 17' is required.

[0082] Below the sealing plate 12, there is a spiral coil which constitutes the heating device 11. An RF field can be generated through the coil, which generates eddy currents in the base 3, thereby heating the base 3. The coil of the heating device 11 is hollow. A coolant can flow through the cavity of the heating device 11. However, in other embodiments of the present invention, the heating device can also be a resistance heating device for heating the base 3, or a radiant heater, through which the base 3 is heated by thermal radiation.

[0083] The input pipelines 8, 16, 17 are connected to a gas mixing system, which has a mass flow controller 18 and a valve 19. Among them, the valve 19 and the mass flow controller 18 are controlled by a control device 22. The control device 22 can have a microcontroller or a microprocessor, in which a memory is arranged, and the memory contains a program. Through this program, the mass flow controller 18 and the valve 19 are controlled according to the program also stored in the memory.

[0084] The air flow can be fed into each of the total five recesses in the embodiment separately and independently through the independent input pipeline 8. The recesses respectively support the substrate holder 5 on the air cushion 6. Therefore, all the air cushions 6 of the substrate holders 5 can be adjusted individually through the control device 22. The height of the air cushion 6 and thus the distance between the substrate holder 5 and the bottom of the recess 4 can be adjusted individually by the mass flow rate of the gas forming the air cushion 6. The mass flow rate of this gas not only affects the temperature of the surface of the substrate 7 carried by the substrate holder 5. The change in the mass flow rate also causes the process gas above the substrate 7 to be diluted, because the process gas fed into the recess 4 flows into the processing chamber 2 through the edge-side gap between the substrate holder 5 and the wall of the recess 4. This affects the growth rate of the layer deposited on the substrate 7. This results in different thicknesses of the layers deposited on the substrate.

[0085] The layer composition of the ternary or quaternary semiconductor layer deposited on the substrate may change due to the change in the temperature of the substrate 7.

[0086] In the said embodiment, at least one input pipeline 16, 17 is introduced below each substrate holder 5, and gas can enter the gap 13 through this pipeline. During operation, if the base 3 rotates around its axis 20, the gas can be fed through the feeding parts 16' and 17' synchronously with the rotation of the base 3. The air flow through the gap 13 can be changed independently, so that the heat conduction between the heated base 3 and the cooler coil 11 is changed. The temperature of the substrate 7 is changed by the changed heat flow.

[0087] Via valve 19 and mass flow controller 18, a mixed gas with different thermal conductivity characteristics can be fed into the input pipelines 8, 16, 17, such as an adjustable mixture composed of nitrogen and hydrogen. By selecting the mixing ratio, the heat transfer can be adjusted separately by conduction towards the substrate support 5 or away from the base 3 under one of the substrates 7.

[0088] Via an optional input pipeline 23 that is introduced at the upper side 3' of the base 3 on the radially inner side of the substrate 7 through an introduction part 23', other temperature-regulating gases can be fed, which can also be a mixture composed of two gases with different thermal conductivity characteristics. It is also stipulated here that an introduction part 23' is independently assigned to each substrate 7, and an independent gas mixture or an independent gas flow can be fed into the processing chamber 2 by this introduction part.

[0089] In a variant of the device or method, reactive gases can also be fed into the introduction part 23'.

[0090] The present invention is applicable to both devices with the above four input pipelines 8, 16, 17, 23 and devices with only one or less than four input pipelines 8, 16, 17, 23. In order to implement the method, in principle, only one set of these input pipelines needs to be realized.

[0091] For the manufacture of VCSEL diodes, it is crucial to precisely control the growth rate and the material composition of each individual substrate 7. These variables have a direct impact on the wavelength of the light emitted by the diode. The layer sequence deposited on the substrate 7 forms a Bragg mirror. Here, the material composition determined by temperature is a key variable. In addition, the individual layer thickness determined by the growth rate is also a key variable here. During the production process, even the smallest deviation between individual wafers may lead to incorrect results. However, the method is not limited to the manufacture of layer sequences for forming Bragg reflectors, but also involves the manufacture of layers or layer sequences for other components.

[0092] It has been found that changes in the gas flow rates through the input pipelines 8, 16, 17, 23 not only affect the material composition or growth rate of the layers or layer sequences deposited on the respectively corresponding substrates 7, but also affect the material composition or growth rate of other substrates 7 due to cross effects.

[0093] These gas flows or gas mixing ratios are referred to as independent parameters in this disclosure document. In this disclosure, the material composition or growth rate is referred to as a value.

[0094] One of the technical problems to be solved by the present invention is to find the target parameters for the preset independent parameters such as gas flow rate or gas mixing ratio To achieve the target values for the material composition, layer thickness, or wavelength in the case of VCSEL diodes

[0095] In a method for a CVD reactor used to set a layer sequence for depositing a Bragg mirror that forms a wavelength as uniform as possible on a substrate 7, in a first preliminary test, an uncorrected flow rate of a temperature-regulating gas is fed through at least one of the aforementioned input pipelines 8, 17, 16, 23. For example, an air flow with the following uniform parameter q0 is set.

[0096] For example, an air flow with the following uniform parameter λ i is set:

[0097]

[0098] Then, five substrates in the aforementioned embodiment are measured, and the wavelength of the Bragg reflector is determined. The following starting values of the wavelength

[0099]

[0100] In a second preliminary test, the layer sequence is coated on five other substrates. However, the gas flow rates are not the same. At least one gas flow rate is reduced by a parameter difference Δq. For example, the following test parameter set is used.

[0101]

[0102] These substrates are also measured in terms of the Bragg reflection wavelength, and the following test values

[0103]

[0104] Then, the average value of the starting values and the average value of the test values are determined using Equations 9 and 10. After that, the test response is determined according to Formula 11 by means of these average values and the measured wavelength

[0105] When depositing the layer sequence in the second preliminary test, a gas flow rate of 400 sccm is changed to 150 sccm with Δq. A column of the sensitivity matrix S i,j has the following form

[0106]

[0107] In the embodiment, the substrate support 5 and the introduction parts 16', 17', 8', 23' are arranged symmetrically with respect to the rotation axis 20. Therefore, it can be considered that the change in the independent parameter corresponding to one of the substrates 7 is the same as the change in the independent parameter corresponding to each substrate 7. Subsequently, the sensitivity matrix S can be established by cyclic exchange as shown in the following table i,j for the other columns of

[0108]

[0109] By forming the inverse sensitivity matrix the correction parameter Δq can be calculated using the above method and the correction value Δλ given in Equation 19 i i ,

[0110]

[0111] Subsequently, the target parameter can be adjusted according to the relationship given in Equations 19 and 20 using these correction parameters Δq i in the form of the corrected flow rates leading to the introduction parts 8', 16', 17', 23'.

[0112] These correction values Δλ i and the target parameter can be stored in the memory of the control device 22.

[0113] For example, in a variant of the method, a second starting value of the vector implemented with the parameter vector given in Equation 21 can be established after a first preliminary test, and the first value λ i is a Bragg reflection, and the second value λ' i can be temperature. After a second preliminary test implemented with the test parameters given in Equation 23, for example, the second test value can be determined. Then, a second sensitivity matrix S' i,j can be established in a similar manner to the above method. After forming the second inverse sensitivity matrix, other correction values, correction parameters, etc. can be calculated using the above method.

[0114] In an embodiment of the present invention, it can also be specified that the measured value λ i in the method, such as Bragg reflection or layer thickness or temperature, is affected by two different independent parameters λ i . These values λ i can be affected, for example, by both the mass flow rate forming the air cushion 6 and the mass flow rate of the temperature-regulating gas flowing through the gap 13. Different individual parameters q i affect the value λ i ​​Have different degrees of influence. In this deformation scheme, the second test value is also determined And the second sensitivity matrix S′ is determined in a similar manner i,j . Thus, different independent parameters q can be predicted i On the value λ i What kind of possible different influences will be produced

[0115] In addition, by forming one or two inverse matrices Parameter optimization can also be achieved

[0116] The above embodiments are used to illustrate all the inventions included in the present application. The present invention makes creative improvements to the prior art through at least the following technical features respectively. Among them, the combination of two, multiple or all technical features can also be combined, that is:

[0117] A method, characterized in that in the first pre-test, respectively through a first set of independent starting parameters Deposit the starting layer or starting layer sequence on a plurality of first substrates 7 simultaneously, and determine the starting value of the layer characteristics at the starting layer or starting layer sequence Or measure the starting value of the local variable during the deposition of the starting layer or starting layer sequence In one or more second pre-tests, respectively through a second set of independent test parameters Deposit the test layer or test layer sequence on a plurality of second substrates 7 simultaneously, and determine the test value of the layer characteristics at the test layer or test layer sequence Or measure the test value of the local variable during the deposition of the test layer or test layer sequence And at least from the test values of the layer characteristics or local quantities Form at least one sensitivity matrix S i,j , the elements of which respectively represent each independent parameter q i The change of has an impact on each value λ of the layer characteristics or local variables i Influence

[0118] A method for setting a device for simultaneously depositing a layer or a layer sequence on a substrate 7 at different support positions 5' arranged in a processing chamber 2, wherein, according to the method described in claim 1, form at least one sensitivity matrix S i,j , by inverting the sensitivity matrix S i,j Form an inverse sensitivity matrix And through the inverse sensitivity matrix Obtain the independent correction parameter Δq i Independent target parameters Or an independent correction factor for correcting the independent parameter q i To achieve the value λi preset target value

[0119] A method, characterized in that an independent parameter q i is the value of an air flow or a heat flow, which air flows or heat flows are directed at a substrate 7 arranged in the same processing chamber 2 at different support positions 5'.

[0120] A method, characterized in that the device has a plurality of identically constructed substrate supports 5, each of which carries one or more substrates 7, and the independent parameter q i influences the purge air flow, heat flow or process gas flow directed at the substrate support 5.

[0121] A method, characterized in that, when performing a first pre-test, the independent starting parameters have the same value.

[0122] A method, characterized in that, in one or more second pre-tests, only one of the independent test parameters differs from all the other independent test parameters by a parameter difference Δq.

[0123] A method, characterized in that the substrate 7 or the substrate holder 5 carrying the substrate 7 or the support position supporting the substrate 7 is symmetrically arranged in the processing chamber 2, and only one second pre-test is performed, in which only one of the independent test parameters differs from the other independent test parameters identical to each other by a parameter difference Δq, and in which a sensitivity matrix S is established from the test values of the layer properties thus obtained i,j for the first column of the sensitivity matrix S i,j and the remaining columns of the sensitivity matrix S are formed by cyclic exchange of the test values.

[0124] A method, characterized in that the elements of the sensitivity matrix S are formed by the quotient of the difference and the parameter difference Δq, where the difference is the difference between the test value i,j and the average value of the starting values and the starting value or the difference between the test value and the starting value and the starting value or the difference between the difference between the test value and the average value of the test values and the difference between the starting value and the average value of the starting values and the starting value and the average value of the starting values

[0125] ​A method, characterized in that an independent correction parameter Δq i is formed by multiplying an inverse sensitivity matrix by a vector containing a correction value Δλ i , and / or an independent correction value Δλ i is the difference between a starting value and a target value , and / or an independent correction factor k i is the quotient of the sum of a starting parameter and a correction parameter Δq i divided by a starting value .

[0126] A method, characterized in that an independent parameter q i is an air flow or an air flow component, through which an air cushion 6 is generated, and the air cushion 6 bears a substrate support 5, and the substrate support 5 is heated from below by heating a base 3 with a heating device 11, and / or an independent parameter q i is an air flow or an air flow component, through which the heat transfer between a temperature control device 11 and a base 3 can be affected, and / or an independent parameter q i is an air flow of a process gas, and the process gas contains elements for forming a deposited layer or a layer sequence, or elements doped into the deposited layer or the layer sequence.

[0127] A method, characterized in that by successively applying the method according to one of the preceding claims multiple times, for the same or different target values and different independent parameters q i a plurality of sensitivity matrices S i,j are formed.

[0128] A method, characterized in that in order to predict the change of a second value λ′ of layer characteristics of a layer or a layer sequence deposited on a plurality of substrates 7 at different support positions 5' arranged in a processing chamber 2, or to predict local variables affecting layer growth, an initial value i obtained by a first set of separate initial parameters q S is used, and in one or more other third pre-tests, on a plurality of third substrates 7, a second test layer or a second test layer sequence is simultaneously deposited with a third set of second test parameters corresponding to and different from the first test parameter i and separate parameters q , and a second test value of layer characteristics is determined at the second test layer or the second test layer sequence or a second test value of local variables is measured during the deposition of the second test layer or the second test layer sequence ​and a second test value of a layer characteristic or a local variable constitute at least one second sensitivity matrix S′ i,j , the elements of the second sensitivity matrix respectively indicating the change of each of the additional individual parameters q i in the second target value of the layer characteristic or the local variable for each of the effects.

[0129] A method, characterized in that, in a first application of the method, the independent parameter q' i is the gas flow rate for generating the air cushion for carrying the local substrate support 5, and the target value is the layer thickness of the layer or the layer thickness of at least one layer in the layer sequence. In a subsequent second application of the method, the independent parameter q i is the gas flow rate affecting the heat transfer from the heating device 11 to the substrate 7, and the target value is the temperature of the substrate 7 during the layer deposition process, or may also be the layer thickness of the layer or the layer thickness of at least one layer in the layer sequence.

[0130] A device, characterized in that correction factors k i for correcting the independent parameter q i are stored in the control device 22, and the independent parameter q provided by the formula is corrected by these correction factors i .

[0131] A device, characterized in that, in order to determine the correction factor k i , the sensitivity matrix S according to one of the preceding claims is constituted and used i,j or the inverse sensitivity matrix

[0132] A method, characterized in that the independent parameter q is corrected by the correction factor k i , and the correction factor k i is determined especially according to one of claims 1 to 12. i

[0133] A control program for controlling the valve 19 and the mass flow controller 18 of the device according to any one of claims 13 or 14 to execute the method according to claim 15.

[0134] ​All the publicly disclosed technical features (either alone or in different combinations) are important for the present invention. The disclosure of the affiliated / priority document (copy of the prior application) is also fully incorporated in the disclosure document of the present application, and for this purpose, the features of this document are also included in the claims of the present application. The dependent claims, with their features, can represent unique and creative improvement solutions to the prior art even without the technical features of the cited claims, especially for the purpose of filing divisional applications based on these technical features. The invention described in each claim may also have one or more features that are particularly provided with reference numerals in the foregoing description and / or given in the list of reference numerals. The present invention also relates to various design forms, in which individual features mentioned in the above description are not implemented, especially when they are considered irrelevant to the corresponding intended use or can be replaced by other means with the same technical effect.

[0135] List of Reference Numerals

[0136] 1 Reactor housing

[0137] 2 Processing chamber

[0138] 3 Base

[0139] 3' Upper side

[0140] 3” Bottom side

[0141] 4 Recess

[0142] 5' Support position

[0143] 5 Substrate holder

[0144] 6 Air cushion

[0145] 7 Substrate

[0146] 8 Input pipeline

[0147] 8 Inlet part

[0148] 9 Gas inlet element

[0149] 10 Gas outlet element

[0150] 11 Heating device

[0151] 12 Sealing plate

[0152] 13 Gap

[0153] 14 Top of the processing chamber

[0154] 15 Cooling device

[0155] 16 Input pipeline

[0156] 16' Inlet part

[0157] 17 Input pipeline

[0158] 17' Inlet part

[0159] 18 Mass flow controller

[0160] 19 Valve

[0161] 20 Axis of rotation

[0162] 21 Axis of rotation

[0163] 22 Control device

[0164] λ i Value

[0165] λ′ i Second value

[0166] Starting value

[0167] Second starting value

[0168] Target value

[0169] Second target value

[0170] Test response

[0171] Δλ i Correction value

[0172] Test value

[0173] Second test value

[0174] Average value

[0175] Average value of the starting value

[0176] Average value of the test value

[0177] q i Independent parameter

[0178] Starting parameter

[0179] Test parameter

[0180] Second test parameter

[0181] Target parameter

[0182] Δq i Correction parameter

[0183] q0 Unified parameter

[0184] Δq Parameter difference

[0185] S i,j Sensitivity matrix

[0186] S′ i,j Second sensitivity matrix

[0187] Inverse sensitivity matrix

[0188] k i Correction factor

Claims

1. A method for predicting a change in a value (λ i ) of a layer property of a layer or a layer sequence deposited on a plurality of substrates (7) at different support positions (5') arranged in a processing chamber (2) or for predicting a change in a value of a local variable that affects layer growth at the support position (5'), wherein, Feeding a gas and / or regulating the temperature in the processing chamber (2) with process parameters that can be varied, in order to influence the layer properties of the deposited layer or the value of a local variable (λ i ), wherein in a preliminary test at least one sensitivity matrix (S i,j ) is formed, the elements of which respectively represent the influence of the change of each process parameter on each value of the layer properties (λ i ) or each value of the local variable, characterized in that the process parameters are composed of parameters (q i ) that individually correspond to each support position (5'), wherein the independent parameters (q i ) have the same effect on the layer properties or the local variable there at the corresponding support position (5'), wherein a change in one of the independent parameters (q i ) also affects the values (λ i ) at other support positions (5').

2. The method according to claim 1, wherein In a first preliminary test, with a first set of independent starting parameters Deposit a starting layer or a starting layer sequence simultaneously on a plurality of first substrates (7), and determine starting values of layer properties at the starting layer or the starting layer sequence Or measure starting values of local variables during the deposition of the starting layer or the starting layer sequence In a second preliminary test, with a second set of independent test parameters Deposit a test layer or a test layer sequence simultaneously on a plurality of second substrates (7), and determine test values of layer properties on the test layer or the test layer sequence Or measure test values of local variables during the deposition of the test layer or the test layer sequence And form the sensitivity matrix (S ) at least according to the test values of the layer properties or the local variables i,j ).

3. A method for setting up an apparatus for simultaneously depositing a layer or a layer sequence on a substrate (7) at different support positions (5') arranged in a processing chamber (2), wherein, forming at least one sensitivity matrix (S i,j ) according to the method of claim 1 or 2, forming an inverse sensitivity matrix by inverting the sensitivity matrix (S i,j ) and obtaining independent correction parameters (Δq i ), independent target parameters or independent correction factors for correcting independent parameters (q i i ) to achieve a preset target value of a value (λ i ) 4. The method according to one of the preceding claims, characterized in that, The independent parameter (q i ) is a value of an air flow or a heat flow directed to a substrate (7) at different support positions (5') arranged in the same processing chamber (2).

5. The method according to one of the preceding claims, characterized in that, The device has a plurality of substrate supports (5) of identical construction, each of which supports one or more substrates (7), and the individual parameter (q i ) influences the purge gas flow, heat flow or process gas flow flowing to the respective substrate support (5).

6. The method according to any one of claims 2 to 5, characterized in that, During the first pretest, the independent starting parameters have the same value.

7. The method according to any one of claims 2 to 6, characterized in that In one or more second pre-trials, only one of the independent test parameters differs from all the other independent test parameters by a parameter difference (Δq).

8. The method according to any one of claims 2 to 7, characterized in that The substrate (7) or the substrate support (5) carrying the substrate (7) or the substrate support (5) carrying the substrate (7) or the support position supporting the substrate (7) is symmetrically arranged in the processing chamber (2), and only one second pre-test is performed, wherein only one of the independent test parameters differs from the other independent test parameters identical to each other by a parameter difference (Δq), wherein a test value of the layer properties obtained thereby establishes the first column of the sensitivity matrix (S i,j ), and the remaining columns of the sensitivity matrix (S i,j ) are formed by cyclic exchange of the test values .

9. The method according to any one of claims 2 to 8, characterized in that , form the elements of the sensitivity matrix (S i,j ) with the quotient of the difference and the parameter difference (Δq), where the difference is the average value of the test value and the starting value , the difference between the test value and the starting value , or the difference between the test value and the test value , and the difference between the average value of the test value and the starting value and the difference between the average value of the starting value and the starting value . .

10. The method according to any one of claims 3 to 9, characterized in that Independent correction parameter (Δq i ) is formed by multiplying the inverse sensitivity matrix by a vector containing the correction value (Δλ i ), and / or the independent correction value (Δλ i ) is the difference between the starting value and the target value , and / or the independent correction factor (k i ) is the quotient of the sum of the starting parameter and the correction parameter (Δq i ) divided by the starting value .

11. The method according to one of the preceding claims, characterized in that, The independent parameter (q i ) is an air current or an air current component, and an air cushion (6) is generated through the air current or the air current component. The air cushion (6) bears a substrate support (5), and the base (3) is heated by a heating device (11) to heat the substrate support from below.

12. The method according to one of the preceding claims, characterized in that, The independent parameter (q i ) is an air flow or air flow component used to affect heat transfer between the temperature control device (11) and the base (3).

13. The method according to one of the preceding claims, characterized in that, The independent parameter (q i ) is the gas flow of the process gas, which contains elements for forming a deposited layer or layer sequence or elements doped into the deposited layer or layer sequence.

14. An apparatus for setting up for simultaneously depositing a layer or a layer sequence on a substrate (7) at different support positions (5') arranged in a processing chamber (2), or a method for setting local variables influencing layer growth at the support positions (5'), characterized in that, By successively applying the method according to one of claims 3 to 14 a plurality of times, a plurality of sensitivity matrices (S i，j ) are formed for the same or different target values and different independent parameters (q ). and different independent parameters (q i ) of the sensitivity matrix (S i，j ).

15. The method according to any one of claims 3 to 14, characterized in that To predict the change of the second value (λ') of the layer properties of the layer or layer sequence deposited on a plurality of substrates (7) at different support positions (5') arranged in the processing chamber (2) or of the local variables influencing the layer growth, the initial values obtained by a first set of independent initial parameters (q i ) are used S and in one or more other third preliminary tests, on a plurality of third substrates (7) simultaneously, a second test parameter corresponding to a third set of additional independent parameters (q ) and different from the first test parameter i is used to deposit a second test layer or a second test layer sequence respectively, wherein a second test value of the layer properties is determined at the second test layer or the second test layer sequence or a second test value of the local variable is measured during the deposition of the second test layer or the second test layer sequence and at least one second sensitivity matrix (S' ) is formed from the second test values of the layer properties or the local variables , the elements of the second sensitivity matrix respectively indicating the influence of the change of each of the additional independent parameters (q ) on each of the second target values of the layer properties or the local variables i,j . i ​​ 16. The method according to any one of claims 3 to 15, characterized in that, To predict the variation of the value of the layer property (λ i ), in one or more other third preliminary tests, on a plurality of third substrates (7), a second test layer or a second test layer sequence is deposited simultaneously in a third set corresponding to another separate parameter (q i ), which is different from the first test parameter . The second test value of the same layer property is determined at the second test layer or the second test layer sequence or the second test value of the same local variable is measured when depositing the second test layer or the second test layer sequence and at least one second sensitivity matrix (S' i,j ) is constituted by the second test value of the layer property or the same local variable. The elements of the second sensitivity matrix respectively indicate the influence of the variation of each of the other separate parameters (q i ) on each of the second target values of the layer property or the local variable . ​ 17. The method according to claim 14, 15 or 16, characterized in that, In a first application of the method, the independent parameter (q i ) is the gas flow rate for generating the air cushion that bears the local substrate support (5), and the target value is the layer thickness of the layer or the layer thickness of at least one layer in the layer sequence. In a subsequent second application of the method, the independent parameter (q' i ) is the gas flow rate that affects the heat transfer from the heating device (11) to the substrate (7), and the target value is the temperature of the substrate (7) during the layer deposition process, or can also be the layer thickness of the layer or the layer thickness of at least one layer in the layer sequence.

18. An apparatus for depositing a layer or a layer sequence on a plurality of substrates (7) arranged in a processing chamber (2), the apparatus having a gas mixing device with a valve (19) and a mass flow controller (18), a processing chamber (2) arranged in a reactor housing (1), a susceptor (3) arranged in the processing chamber (2) having support positions for the plurality of substrates (7), a gas inlet element (9) for introducing process gases provided in the gas mixing device according to a recipe stored in a control device (22), wherein, The control device (22) is arranged to direct the gas flow or the heat flow to different substrates (7) or one or more substrate carriers (5) carrying the substrates (7) independently of each other according to the independently preset parameters (q i ), characterized in that correction factors (k i ) for correcting the independently preset parameters (q i ) are stored in the control device (22), and the independently preset parameters (q i ) provided by the recipe are corrected by the correction factors, and the correction factors are determined by the method according to one of claims 1 to 16.

19. A method for separately and simultaneously depositing a layer or a layer sequence on a plurality of substrates (7) in a processing chamber (2) by means of the apparatus according to claim 18, wherein, Feed a gas into the processing chamber (2) and / or adjust the temperature in the processing chamber (2) according to preset processing parameters, wherein the processing parameters include different independent parameters (q i ) that independently correspond to the substrate (7), and the independent parameters can be changed independently. A change in one of the independent parameters (q i ) not only causes a change in the value (λ i ) of the layer properties of the layer or layer sequence deposited with the changed independent parameter (q i ), but also causes a change in the value (λ i ) of the layer properties of the layer or layer sequence deposited on another substrate (7). Or a change in one of the independent parameters (q i ) not only causes a change in the value (λ i ) of the local variable that affects layer growth at the support position (5') of the substrate, but also causes a change in the local variable at another support position (5'). It is characterized in that the independent parameter (q i ) is corrected by a correction factor (k i ) determined especially according to one of claims 1 to 16.

20. A control program for controlling a valve (19) and a mass flow controller (18) of the device according to claim 18 to perform the method according to claim 19.

21. A method, apparatus, or control program, characterized in that Having one or more specific features of any one of the foregoing claims.

Images