DE102008044881A1 - Measuring method for a semiconductor structure - Google Patents

Abstract

The invention relates to a measuring method for a semiconductor structure having a front and a rear side, comprising the following method steps: A generating luminescence radiation in the semiconductor structure, B determining a relationship for this semiconductor structure between the material quality of the semiconductor structure, in particular the diffusion length of the minority charge carriers, and the electrical Property of at least one side of the semiconductor structure (front or back), in particular a surface recombination speed of the front or back, the relationship depending on a. an evaluation A1 of the measured intensity of the luminescence radiation with a first spectral weighting with respect to the luminescence radiation considered in the evaluation A1 and b. an evaluation A2 of the measured intensity of the luminescence radiation is determined with a second spectral weighting with respect to the luminescence radiation considered in the evaluation A2 and the first spectral weighting is different from the second spectral weighting, characterized in that in process step B additionally at least one third evaluation A3 of measured intensity of the luminescence radiation is carried out, wherein the three evaluations A1, A2 and A3 in each case with respect to the spectral weighting with respect to the considered in the respective evaluation luminescence and / or with respect to the side of the ...

Description

The The invention relates to a measuring method for a semiconductor structure with a front and a back according to the preamble of claim 1.

at Optoelectronic devices must have the optical and electrical properties of the materials used on each other be adjusted. So is important for solar cells that the generated within the penetration depth of the incident sunlight Minority carriers are able to, while their life, d. H. before they recombine, for they intended to reach contact. Her must be Diffusion length to be greater than the penetration depth of the light. The diffusion length characterizes the distribution the minority carrier in the solar cell but not alone. By recombination on the surfaces, characterized by its surface recombination rate, there is the concentration of minority carriers additionally lowers, which by pure diffusion conditional exponential distribution of the minority carriers is changed. The changed distribution is characterized by an effective diffusion length. While the diffusion length is a pure material property, the surface recombination rate is a property of the component. Both together determine the effective diffusion length and how well a solar cell performs its function can. Your experimental investigation is of great importance for the quality assurance of solar cells.

One Part of the recombination of the minority carriers is bright, so it generates photons through the surfaces be emitted. Their intensity is an absolute measure of the concentration of minority carriers. The spectrum of the emitted radiation, the so-called luminescence radiation, becomes thereby by reabsorption of the photons within the emitting Material influences. Since the absorption probability for shortwave light is usually larger than for long-wave light, is the intensity of the short-wave Light is more of a measure of the concentration of Minority carrier near the emitting surface, while long-wave Light more a measure of the totality of minority carriers is. In different spectral ranges you get different Contributions from recombining minority carriers from different distances from the emitting surface.

It It is known that by measuring the emitted radiation in 2 spectral regions, a short-wave and a longer-wave, the effective diffusion length can be determined. There are thus two evaluations of the emitted Radiation with different spectral weights with respect the considered in the respective evaluation luminescence. To what extent is the pure diffusion and the Surface recombination impact is unknown. With Such a measurement is only a link between possible Values of the true diffusion length and the surface recombination velocity produced. Because different combinations of the two sizes result in the same measurement result must be one of the two sizes be known to determine the other. Below are with Correlation of diffusion length and surface recombination speed always such combinations of possible values of the two sizes meant with the measured effective diffusion length are compatible.

The described measuring method finds in particular for characterization of solar cells or precursors in the manufacture of a solar cell application.

at Solar cells on a semiconductor such as silicon Thus, it is thus known, based on a generated in the solar cell Luminescence radiation a connection of the material quality the semiconductor structure and the surface properties to determine, in particular the relationship of the diffusion length the minority carrier of the semiconductor structure and the surface recombination rate at least one of the sides of the solar cell (front and / or back).

Of the Relationship between the diffusion length of minority carriers and the surface recombination velocity of the front and / or back of the solar cell is doing as before described on the basis of two evaluations of the measured intensity the luminescence determined. The two evaluations differ in the spectral weighting with respect to the each evaluation considered luminescence.

typically, is by means of optical filters such as band edge filter the evaluations each assigned a cut-off wavelength, so that in an evaluation essentially only luminescence radiation measured up to the cut-off wavelength and evaluated accordingly becomes. The spectral weighting is thus carried out by fixing the Cut-off wavelength.

The cutoff wavelengths for the two evaluations are selected differently, so that the relationship between the diffusion length of the luminescence radiation is determined by a comparison of the two evaluations, such as, for example, a quotient formation of the respectively measured intensities of the luminescence radiation Minority carrier and surface recombination rate can be determined. The quotient has the advantage that it eliminates all factors that leave the emitted spectrum unchanged such. B. caused by different series resistances voltage variations in the electroluminescence or inhomogeneous illumination in the photoluminescence.

Such a method is for example in Würfel, P. et al., "Diffusion lengths of silicon solar cells from luminescence images", Journal of Applied Physics, 2007. 101 (123110): p. 1-10 described. Furthermore, such a method is in PCT / AU2007 / 001050 disclosed.

With This method thus allows the relationship between Diffusion length of the minority carriers and determine surface recombination rate. Now one of the sizes is known or can it can be determined by different measurements, so may determined relationship to the remaining size getting closed.

typically, can in such measurements, the surface recombination speed due from experience or it is due other measurements known, so with the method described the diffusion length of minority carriers in the semiconductor structure.

Of these, The present invention is based on the object, to simplify and improve the known measuring method and to create a wider application possibility.

Especially should an evaluation be possible without one of the two physical quantities (diffusion length the minority carrier or surface recombination rate) have to be.

Solved these are tasks by a measuring method for a semiconductor structure according to claim 1. Advantageous embodiments of the measuring method according to the invention can be found in claims 2 to 14.

With the inventive method is additional at least one third evaluation, for example the intensity from a third spectral range used by surface recombination caused deviation of the charge carrier distribution of to determine the distribution expected for pure diffusion and from this the size of the surface recombination rate. Only through this at least one additional evaluation is the determination of true diffusion length and surface recombination speed as separate sizes possible.

The inventive method for the determination of Diffusion length and surface recombination rate with the help of luminescent radiation can be applied to all types of semiconductors or Semiconductor structures are applied, ie both with those direct as well as indirect optical transitions as well as inorganic and organic semiconductors.

Of the Term "semiconductor structure" designates a structure based on a semiconductor and other components may have, such. B. further semiconductor layers, electrically passivating layers on the surfaces and / or layers to reduce the reflection of optical radiation. Likewise are Dopings in partial areas to form a pn junction possible and metallizations on the surfaces, for adding or removing charge carriers.

The Measuring method according to the invention for a Semiconductor structure with a front and a back comprises the following process steps: In one process step A, luminescence radiation is generated in the semiconductor structure.

Subsequently is in a step B a connection for this semiconductor structure between the electrical material quality of Semiconductor structure and the electrical property of at least one Side of the semiconductor structure (i.e., the front and / or the back) certainly. "Context" means here and below, that when specifying one of the two sizes the other size can be determined. in this connection it is within the scope of the invention that for determining the context other parameters are taken into account, such as the basic doping of the semiconductor structure or characteristics the measuring equipment used and / or optical filters.

advantageously, the material quality of the semiconductor structure by the diffusion length of the minority carriers described and the electrical property of at least one side over the surface recombination rate at this Page. It is also within the scope of the invention, other physical Sizes, such as the electrical material quality over the lifetime of the minority carriers too describe.

The aforementioned relationship is dependent on a first evaluation A1 of the measured intensity of the luminescence radiation with a first spectral weighting with respect to the Lumineszenzstrahlung considered determined depending on a second evaluation A2 of the measured intensity of the luminescence with a second spectral weighting with respect to the considered in the evaluation A2 luminescence.

at The evaluations A1 and A2 are only the one to one page the semiconductor structure (front or back) radiated Considered luminescence radiation.

The spectral weighting of the first evaluation A1 and the second evaluation A2 are different.

Essential is that in process step B additionally at least a third evaluation A3 of the measured intensity of Luminescence radiation is made. The three evaluations A1, A2 and A3 differ in terms of spectral weighting with regard to the considered in the respective evaluation Luminescence radiation and / or with respect to the side of the semiconductor structure, whose radiated luminescence radiation is taken into account in the evaluation becomes.

Two any of the three evaluations thus differ either in terms of spectral weighting, or that in a Evaluation of the emitted to the front luminescence and in the other evaluation, the ones emitted to the back Luminescence radiation is taken into account or distinguish them Both in terms of spectral weighting and the in the evaluation considered side of the semiconductor structure.

Dependent Of the at least three evaluations A1, A2 and A3, the material quality becomes the semiconductor structure and / or the electrical property at least a surface of the semiconductor structure determined. The term "dependent of "here and hereinafter means that the said Evaluations essential parameters for the determination of the mentioned Result sizes are, such as the diffusion length the minority carrier as a measure of the electrical material quality of the semiconductor structure. It is within the scope of the invention that the determination of depends on other parameters, in particular of physical Material parameters of the investigated semiconductor structure or characteristics the measuring equipment used or of optical filters used.

With The method according to the invention thus makes it possible to the material quality and / or the surface property to determine without any of these two sizes must be known in advance. In particular, with the inventive Method the determination of both the diffusion length of Minority carriers, as well as the surface recombination rate one side of the semiconductor structure possible.

The Invention is based on the Applicant's finding that the at least an additional evaluation A3 compared to the previously known measuring methods in that they relate to the spectral weighting and / or the considered side of the semiconductor structure of the two other evaluations, the necessary additional information delivers to the relationship between material quality the semiconductor structure and surface electrical property at least one of the sides of the semiconductor structure exactly one pair of values for the material quality and the electrical surface property to determine.

The Measuring method according to the invention is suitable for Application in inorganic and organic semiconductors and indeed with both direct and indirect optical transitions, especially in the case of silicon and solar cells produced from silicon, or their precursors in the production.

The determination of the relationship between diffusion length and surface recombination velocity is basically known and, for example, in WO 2008/014537 , Page 20, line 21 to page 29, line 28 described. This text passage is explicitly incorporated by reference into this specification as belonging to the invention.

advantageously, In step B, a first pair of evaluations becomes a first one Relationship between the diffusion length of the semiconductor structure and the surface recombination rate at least one side of the semiconductor structure. Analog becomes from a second pair of evaluations a second relationship between the diffusion length of the semiconductor structure and the surface recombination speed at least one side of the semiconductor structure, wherein in the first and in the second context the same page is used. The first and second pairs of evaluations differ in at least one evaluation and a value pair (eg. a combination of minority carrier diffusion length and surface recombination rate), which corresponds to both the first and the second context.

If three evaluations A1, A2 and A3 are used as the basis for the method according to the invention, in this advantageous embodiment, for example, a first relationship based on the evaluation A1 and A2 and a second relationship based on the evaluation A1 and A3 will be agrees and for these two connections a single value pair of diffusion length and surface recombination speed, which corresponds to both contexts.

The spectral weighting on evaluation of the measured luminescence radiation is characterized by having different wavelengths or different wavelength ranges a different one Influence on the evaluated intensity of the luminescence radiation have.

Advantageously, the spectral weighting is realized in that the first evaluation A1 is assigned a first cutoff wavelength λ ₁ such that essentially only radiation with a wavelength less than or equal to λ _{1 is} taken into account in the evaluation A1. This assignment of a cut-off wavelength thus determines the spectral weighting of the evaluation A1 in this advantageous embodiment. In the same way, the evaluation A2 is assigned a second limit wavelength λ ₂ and the evaluation A3 is assigned a third limit wavelength λ ₃ , and the limit wavelengths are chosen such that λ ₁ <λ ₂ <λ ₃ .

The Absorption of radiation is usually over Absorption coefficient described or on the reciprocal, as a measure of the penetration depth of the radiation into the absorbing medium. The definition of cut-off wavelengths can thus also be defined as a penetration depth for the considered in the respective evaluation Interpret luminescence radiation.

Advantageously, the cutoff wavelengths λ ₁ , λ ₂ and λ _{3 are} therefore determined as a function of the penetration depth for radiation in the semiconductor structure and the thickness of the semiconductor structure, ie in particular depending on the absorption coefficient for the corresponding material of the semiconductor structure.

Applicant's investigations have shown that λ _{1 is} advantageously selected such that the penetration depth is less than 50% of the thickness, λ _{2 is selected} such that the penetration depth is between 50% and 150% of the thickness and λ _{3 is} chosen such is that the penetration depth is> 100% of the thickness.

A further improvement of the measurement is achieved if the abovementioned penetration depths for λ _{1 are} less than 30% of the thickness, for λ ₂ between 80% and 120% of the thickness and for λ ₃ more than 150% of the thickness. In particular, it is advantageous that the cut-off wavelengths are selected such that the penetration depth for λ _{1 is} less than 10%, for λ ₂ the penetration depth is approximately 100% and for λ ₃ the penetration depth is more than 300% of the thickness.

The different penetration depths of the selected cut-off wavelengths are a measure of what influence the surface of the semiconductor based on the cutoff wavelength Has evaluation. The advantageous choice of the as described above Boundary wavelengths therefore enables a good separation the influence of surface properties and electrical Material properties of the semiconductor structure.

In the aforementioned advantageous embodiment, the first evaluation A1 is thus associated with the lowest penetration depth with respect to the evaluated luminescence radiation. Investigations by the Applicant have shown that good evaluability is given in particular when, in the aforementioned advantageous embodiment, the evaluation A1 with the cut-off wavelength λ _{1 is used} to determine both the first and the second relationship, in particular that the first relationship depends on the evaluation A1 and A2 and the second relationship is determined depending on the evaluation A1 and A3.

In It is the aforementioned advantageous embodiments particularly advantageous that the evaluations A1 to A3 consider luminescence radiation, which is emitted from the same side of the semiconductor structure, d. H. that a measurement page is defined, which the front or the back of the semiconductor structure is and the evaluations A1 to A3 in each case only those radiated on the measuring side, take into account measured luminescence radiation.

As previously described, it is also within the scope of the invention that alternatively or additionally, the evaluations by distinguish the side of the semiconductor structure whose radiated Luminescence radiation considered in the respective evaluation becomes.

In an advantageous embodiment of the method according to the invention, therefore, a first and a second measurement side is defined, wherein the first measurement side is the front or back side of the semiconductor structure and the second measurement side is the side of the semiconductor structure opposite the first measurement side. Furthermore, in step B at least a fourth evaluation A4 of the measured intensity of the luminescence radiation is additionally performed. The evaluations A1 and A3 is a cutoff wavelength λ _i allocated such that substantially only radiation with a wavelength equal to λ _i becomes smaller taken into account in the evaluation A1 and A3. In the same way, the evaluations A2 and A4 are assigned a second limit wavelength λ _ii .

Similarly, the evaluations A1 and A2 is the assigned to the first measurement side, so that in these evaluations only the measured luminescence radiation emitted on the first measurement side is taken into account. In the same way, the evaluations A3 and A4 are assigned the second measurement page.

at This advantageous embodiment will thus be two Cutoff wavelengths and two measurement sides crosswise at four Evaluations varied. This allows a particularly accurate Determination of the diffusion length of minority carriers and / or the surface recombination rate.

Advantageously, in the embodiment described above, the cut-off wavelength λ _{i is selected} according to the above-mentioned conditions for λ ₁ and the cut-off wavelength λ _{ii in} accordance with the conditions given above for λ ₃ . Due to the significantly different penetration depths of the cutoff wavelengths λ ₁ and λ ₃ , the accuracy of the evaluation is further improved.

The Penetration depth is in each case from the side of the semiconductor structure starting from the minority carrier be injected by electroluminescence by applying a voltage or which is illuminated in the photoluminescence.

The inventive method is particularly suitable for measuring a solar cell or a precursor during manufacture a solar cell. It is advantageous that the luminescence radiation generated by a voltage applied to the contacts of the solar cell is, d. H. is an electroluminescent radiation.

advantageously, For this purpose, a voltage is selected that is approximately corresponds to the voltage of the solar cell at the operating point. The working point refers to the point on the characteristic of the solar cell, at the product of current and voltage gives the maximum power.

In A further advantageous embodiment is in step A the luminescence radiation by illuminating the semiconductor structure generated with an excitation radiation, d. H. the luminescence radiation is photoluminescence radiation.

In order to avoid a falsification of the measurement of the luminescence radiation by the excitation radiation used, it is advantageous for the spectrum of the excitation radiation essentially to comprise only wavelengths smaller than a limiting wavelength λ _A. λ _A is chosen such that the penetration depth of the excitation radiation is less than 10%, in particular less than 5% of the thickness of the semiconductor structure.

alternative or in addition it is advantageous that the lighting the semiconductor structure is made from a lighting side, which is the front or the back of the semiconductor structure and that only luminescence radiation is taken into account in the evaluations which is opposite to that of the lighting side Side of the semiconductor structure is emitted. In this advantageous Embodiment thus additionally serves the semiconductor structure as an optical filter in order to influence the measurement of the luminescence radiation to avoid by the excitation radiation.

In a further advantageous embodiment of the invention Method of generating the luminescent radiation by illumination The semiconductor structure is distinguished by at least two evaluations in that in one evaluation the semiconductor structure of the front side is illuminated and in the other evaluation the semiconductor structure illuminated from the back. This will be different Measurement conditions for the two evaluations achieved. These Embodiment has the advantage that, for example only on one side of the semiconductor structure in the measuring device a detector must be arranged and only on two sides in each case a radiation source is arranged to act on the front and the back with excitation radiation. Typically, radiation sources are much cheaper, compared to detectors, especially spatially resolving Detectors, so that a measuring device with two sources of radiation and a detector compared with a measuring device with two Detectors and a radiation source significantly cheaper is.

Advantageously, in the method according to the invention, the luminescence radiation is measured by means of a camera, such as a CCD camera. The measurement signal of the camera is thus a measure of the intensity of the luminescence radiation. In particular, it is advantageous to carry out the method according to the invention with spatial resolution, for example by using a spatially resolving camera. For this purpose, methods are already known and, for example, in Dice, P. et al, supra. described. In this embodiment, the method according to the invention thus determines the diffusion length of the minority charge carriers and / or the surface recombination velocity in a spatially resolved manner for a multiplicity of location points of the semiconductor structure, ie a so-called "mapping" of the respective physical quantities is performed.

In addition to the determination of the diffusion length, the minority carrier and the surface recombination rate are the same with the method according to the invention described above the determination of local voltage variations in a semiconductor structure by measuring the electroluminescence radiation possible:
In an advantageous embodiment of the measuring method according to the invention, this therefore comprises the following method steps:
In a step i. the diffusion length and the surface recombination velocity of at least one surface of the semiconductor structure is determined by means of the method according to the invention. This determination is carried out in a spatially resolved manner so that in each case the diffusion length and the surface recombination velocity for this region are determined for different regions of the semiconductor structure. Typically, a CCD camera is used for such a spatially resolved determination, so that a diffusion length and a surface recombination velocity for the associated region of the semiconductor structure can be determined for each pixel of the image determined by the CCD camera.

In a step ii. is for each location in step i. carried out the determination theoretically for a location-independent voltage expected luminescence radiation calculated. This calculation is dependent on the for the respective location point in step i. certain diffusion length and Surface recombination.

In a step iii. the quotient is calculated for each pixel the one in step i. measured luminescence radiation and in step ii. calculated luminescence radiation is formed. Would all be in step i. observed intensity variations only by Variation of the diffusion length and / or the surface recombination speed caused, then the quotient formed for all pixels have the same value. After the quotient still existing Intensity variations are due to variations in the lateral stress distribution, d. H. the stress distribution parallel to the front or back attributed to the semiconductor structure. The voltage variations can from the step iii. be generated quantitatively, since there is a change in the voltage by kT / e = 0.026 V changes the luminescence intensity by a factor of 2.72.

Further Features and advantageous embodiments of the invention Method will be explained below with reference to the figures. Showing:

1 a measuring device for carrying out an advantageous embodiment of the measuring method according to the invention, in which only luminescence radiation radiated from the front side of the semiconductor structure is evaluated, and

2 a measuring device for carrying out a further embodiment of the measuring method according to the invention, in which both from the front and from the back of the semiconductor structure radiated luminescence radiation is evaluated.

In the 1 illustrated measuring device is used to measure a semiconductor structure 1 with a front side 1a and a back 1b , The measuring device comprises a camera designed as a CCD camera 2 with a lens 2a , The objective 2a is designed in such a way and the semiconductor structure is arranged such that from the front 1a the semiconductor structure 1 emitted luminescence radiation through the lens 2a on a not shown CCD chip in the camera 2 is shown. This advantageously comprises a square grid of 1024 × 1024 pixels, ie a signal is evaluated separately for each of these pixels. The perpendicular to the plane in 1 standing front 1a is thus measured spatially resolved at 1024 × 1024 location points.

Between semiconductor structure 1 and camera 2 is an optical filter device 3 arranged from the in 1 three lenticular filters are shown by way of example.

The filter device 3 is designed such that depending on supplied control signals different filters in the beam path between the semiconductor structure 1 and lens 2a the camera 2 be introduced.

The luminescence radiation in the semiconductor structure 1 will be at the in 1 illustrated device via a light source 4 produced, it is thus photoluminescence. The light source 4 is designed as a laser that generates light radiation with a wavelength of about 700 nm and includes a lens assembly, by means of which the laser radiation over the entire surface and in approximately homogeneous on the back 1b the semiconductor structure 1 is shown.

The measuring device in 1 further comprises a control and evaluation unit, not shown, which is designed as a computer. The computer is both with the light source 4 , the filter device 3 , as well as the camera 2 connected, for the control of these elements as well as for the storage and evaluation of the measuring signals of the camera 2 ,

An embodiment of the measuring method according to the invention, applied to silicon as a semiconductor material is described with the in 1 illustrated measuring device as follows:
First, luminescence radiation in the semiconductor structure 1 generated. This is the back 1b over the entire surface and approximately homogeneously with radiation of wavelength 700 nm of the light source 4 applied. The radiation penetrates into the semiconductor structure 1 one and generates there electron-hole pairs. The corresponding recombination of electron-hole pairs generates luminescence radiation which, inter alia, over the front 1a the semiconductor structure 1 is emitted.

The filter device 3 has three different shortpass filters.

In a first step, the filter device 3 controlled by the control unit such that a short-pass filter in the beam path between the semiconductor structure 1 and camera 2 is pivoted, which essentially passes only radiation having a wavelength less than or equal to 900 nm. In this measurement step is thus only from the front 1a radiated luminescence radiation with a wavelength of less than or equal to 900 nm from the camera 2 detected spatially resolved and the corresponding measurement signals of the CCD camera are stored in the evaluation.

Corresponding In a second measuring step, a second short-pass filter in pivoted the beam path, which only radiation less than or equal 1000 nm passes and that for this measurement step Corresponding measuring signals of the CCD camera are provided by the evaluation unit stored separately.

After all In a third measuring step, a short-pass filter is pivoted in, which only radiation with a wavelength smaller passes through 1050 nm and the associated Measuring signals of the camera are also separated in the evaluation unit stored.

As well it is within the scope of the invention, the different measurements at the same time, for example, by three separate filter arrangements with each assigned camera.

Essential is that, depending on the three measurements described above three evaluations A1 to A3 are made. These evaluations Assign each of the measurement signals a number, which is a measure of the intensity of the measured at the respective measurement Luminescence radiation is.

In this case, recourse may be had to measuring methods which are known per se, for example the measuring signal for each pixel of the camera may be available for a given integration time 2 be integrated separately.

Essential is that in all three evaluations an identical evaluation is done, so eventually for each pixel there are three values which correspond to the intensities of the luminescence radiation for the evaluations A1 to A3 correspond.

Subsequently, the evaluation unit determines a first relationship between the diffusion length of the semiconductor structure 1 and the surface recombination velocity of one side of the semiconductor structure 1 certainly. In the present example, the semiconductor structure represents 1 a precursor of a solar cell, which consists of a p-doped silicon wafer, on the front side 1a an n-doped emitter was diffused. The surface recombination rate at the front 1a is therefore irrelevant and for the evaluation is only the surface recombination speed at the back 1b relevant.

The evaluation unit now determines a first relationship between the diffusion length of the minority charge carriers in the semiconductor structure 1 and the surface recombination speed of the backside 1b depending on the numerical values determined in evaluations A1 and A2. It is particularly advantageous here if the quotient of the determined numerical values from the evaluations A1 and A2 is formed, since in this case effects which have a multiplicative effect on the measurement result have no influence on the evaluation.

The quotient formed from the evaluations A1 and A2 allows the determination of a first relationship between the mentioned diffusion length and surface recombination speed via a predetermined theoretical formula. This relationship and the following relationships are advantageously as in WO 2008/014537 , Page 20, line 21 to page 29, line 28 determined.

Essential Now, that is dependent on a second context is formed by the evaluations A1 and A3 in the same way, d. H. advantageously also via the formation of a Quotient.

The The evaluation unit now determines the value pair from the diffusion length and surface recombination rate, both corresponds to the first, as well as the second context. by virtue of the choice of cut-off wavelengths of the aforementioned short-pass filter differ the measurement conditions, so that only one pair of values from diffusion length and surface recombination speed exists, which fulfills both connections.

In this way, an unambiguous determination of both the diffusion length and the surface recombination velocity is possible without any of these two quantities being previously known or would have to be estimated.

In 2 is a development of the measuring device in 1 represented, with which a further advantageous embodiment of the measuring method according to the invention can be carried out:
In the 2 and 1 identical reference signs denote identical objects.

The measuring device in 2 additionally has a second camera designed as a CCD camera 2 ' with a lens 2 ' and a second filter device 3 ' on.

The two cameras 2 and 2 ' as well as filter devices 3 and 3 ' are on opposite sides of the semiconductor structure to be measured 1 arranged.

Because of this, the light source is 4 which also in 2 designed as a laser with an output radiation in the range of 700 nm, slightly laterally arranged, the optics of the light source 4 designed such that the back 1b the semiconductor structure 1 over the entire surface and substantially homogeneously charged with the laser radiation.

With the device according to 2 can thus by means of a filter device 3 , and camera 2 from the front 1a radiated luminescence and according to the camera 2 and the filter device 3 ' from the back 1b radiated luminescence radiation can be measured.

It is essential here that the filter device 3 ' additionally has a long-pass filter, which only transmits radiation having the wavelength greater than 700 nm. This is necessary because the radiation of the light source 4 partly at the back 1b is reflected and the measurement by means of the camera 2 ' would affect. By the aforementioned long-pass filter is a disturbance of the measurement due to the radiation of the light source 4 avoided.

With the device according to 2 an embodiment of a method according to the invention is carried out, which basically in the case of 1 corresponds to the measuring method described.

In this case, however, a total of 4 evaluations are carried out, described here for silicon as semiconductor material:
First, a first cutoff wavelength of 900 nm and a second cutoff wavelength of 1050 nm are defined. The filter devices 3 and 3 ' are designed such that optionally corresponding short-pass filter in the beam path between the semiconductor structure 1 and camera 2 or camera 2 ' can be swiveled.

The Control and evaluation unit is with both filter devices and connected to both cameras.

In the measurement, a first evaluation A1 with the first cut-off wavelength with respect to that of the front side 1a radiated luminescence, a second evaluation A2 of the second cut-off wavelength with respect to the emitted from the front luminescence by means of the camera 2 and the filter device 3 performed. Accordingly, a third evaluation A3 with the first cut-off wavelength of the backside 1b radiated luminescence radiation and a fourth evaluation A4 with the second cutoff wavelength with respect to that of the back 1b radiated luminescence by means of the camera 2 ' and the filter device 3 ' performed.

at this embodiment of the invention Measuring method are thus four numerical values, each one Measure of the intensity of the luminescence radiation represent, available for evaluation.

The evaluation unit now determines from the evaluations A1 and A2 a first relationship and from the evaluations A3 and A4 a second relationship between the diffusion length of the minority charge carriers and the surface recombination speed of the rear side 1b the semiconductor structure 1 ,

Also here exists only due to the described measurement conditions a value pair of diffusion length and surface recombination speed, which applies to both the first and the second context applies and accordingly from the evaluation of this pair of values determined.

The measuring method, which with the in 2 shown measuring device, has with respect to the measuring method of the measuring device in 1 a higher accuracy, since by the use of four evaluations and measurement of the luminescence radiation from two different sides of the semiconductor structure 1 greater accuracy is achieved.

QUOTES INCLUDE IN THE DESCRIPTION

This list The documents listed by the applicant have been automated generated and is solely for better information recorded by the reader. The list is not part of the German Patent or utility model application. The DPMA takes over no liability for any errors or omissions.

Cited patent literature

• - AU 2007/001050 [0010]
• - WO 2008/014537 [0031, 0078]

Cited non-patent literature

• Cube, P. et al., "Diffusion lengths of silicon solar cells from luminescence images", Journal of Applied Physics, 2007. 101 (123110): p. 1-10 [0010]
• Dice, P. et al., Supra . [0055]

Claims (15)

Measuring method for a semiconductor structure ( 1 ) with a front and a back ( 1a . 1b ), the following method steps comprising: A generating luminescence radiation in the semiconductor structure ( 1 ), B determining a relationship for this semiconductor structure ( 1 ) between the electrical material quality of the semiconductor structure ( 1 ) and the electrical property of the front and / or back ( 1a . 1b ) of the semiconductor structure ( 1 ), and wherein at least two evaluations of the measured intensity are made and the relationship depends on a. a first evaluation A1 of the front or back side ( 1a . 1b ) of the semiconductor structure ( 1 ) radiated, measured intensity of the luminescence radiation with a first spectral weighting with respect to the considered in the evaluation A1 luminescence and b. a second evaluation A2 of the front or back ( 1a . 1b ) of the semiconductor structure ( 1 ), the measured intensity of the luminescence radiation is determined with a second spectral weighting with respect to the luminescence radiation considered in the evaluation A2 and the first spectral weighting is different from the second spectral weighting, characterized in that in process step B additionally at least one third evaluation A3 of the front or back side ( 1a . 1b ) of the semiconductor structure ( 1 ), the three evaluations A1, A2 and A3 in each case with regard to the spectral weighting with respect to the luminescence radiation considered in the respective evaluation and / or with respect to the side of the semiconductor structure ( 1 ), whose radiated luminescence radiation is taken into account in the evaluation, differ and that, depending on the at least three evaluations A1, A2 and A3, the material quality of the semiconductor structure ( 1 ) and / or the electrical property of at least one surface of the semiconductor structure ( 1 ) is determined. A method according to claim 1, characterized in that in step B from a first pair of evaluations, a first relationship between the diffusion length of the semiconductor structure ( 1 ) and the surface recombination velocity of at least one side of the semiconductor structure ( 1 ) (Front or back side ( 1a . 1b )) and from a second pair of evaluations a second relationship between the diffusion length of the semiconductor structure ( 1 ) and the surface recombination velocity of at least one side of the semiconductor structure ( 1 ) (Front or back side ( 1a . 1b )), wherein the first pair and the second pair of evaluations differ in at least one evaluation, and that a value pair (diffusion length, surface recombination velocity) is determined which corresponds to both the first and the second context. Method according to at least one of the preceding claims, characterized in that the first evaluation A1 is associated with a first cut-off wavelength λ ₁ , such that substantially only radiation with a wavelength less than or equal to λ _{1 is} taken into account in the evaluation A1 and in the same way the evaluation A2 a second cutoff wavelength λ ₂ and the evaluation A3 is assigned a third cutoff wavelength λ ₃ and wherein λ _{1 is} smaller than λ ₂ and λ _{2 is} smaller than λ ₃ . Method according to Claim 3, characterized in that the cut-off wavelengths λ ₁ , λ ₂ and λ ₃ depend on the penetration depth for radiation in the semiconductor structure ( 1 ) and the thickness of the semiconductor structure ( 1 ), wherein λ _{1 is selected} such that the penetration depth is less than 50%, in particular less than 30%, in particular less than 10% of the thickness, λ _{2 is} chosen such that the penetration depth is between 50% and 150%, in particular between 80% and 120%, most preferably in about 100% of the thickness and λ _{3 is selected} such that the penetration depth is greater than 100%, in particular greater than 150%, most preferably greater than 300% of the thickness. Method according to claim 2 and at least one of claims 3 and 4, characterized in that the evaluation A1 with the cut-off wavelength λ _{1 is used} to determine both the first and the second relationship, in particular that the first relationship depends on the evaluations A1 and A2 and the second relationship is determined depending on the evaluations A1 and A3. Method according to at least one of the preceding claims 3 to 5, characterized in that a measurement page is defined which the front or the back ( 1a . 1b ) of the semiconductor structure ( 1 ) and that in the evaluations A1 to A3 in each case the measured luminescence radiation emitted on the measurement side is taken into account. Method according to at least one of claims 1 or 2, characterized in that a first and a second measurement side is defined, wherein the first measurement side is the front or rear side ( 1a . 1b ) of the semiconductor structure ( 1 ) and the second measuring side is the side of the semiconductor structure opposite the first measuring side ( 1 ) and that in step B additionally at least a fourth evaluation A4 of the measured intensity of the luminescence radiation is made, wherein the evaluations A1 and A3, a cut-off wavelength λ _{i is} assigned, such that substantially only radiation having a wavelength less than or equal to λ _i in the Evaluation A1 and A3 is taken into account and in the same way the evaluations A2 and A4 is assigned a second cutoff wavelength λ _ii and the evaluations A1 and A2 are assigned to the first measurement page, such that in the evaluations A1 and A2 only the ones on the first measurement side radiated, measured luminescence radiation is taken into account and according to the evaluations A3 and A4, the second measurement page is assigned. A method according to claim 7, characterized in that λ _{i is selected} according to the conditions specified in claim 4 for λ ₁ and λ _ii according to the conditions specified in claim 4 for λ ₃ . Method according to at least one of the preceding claims, characterized in that the semiconductor structure to be measured ( 1 ) is a solar cell or a precursor in the manufacture of a solar cell and that in step A the luminescent radiation is electroluminescent radiation which is generated by applying an external voltage to the solar cell. Method according to claim 9, characterized in that that the voltage applied to the solar cell is about the voltage corresponds to the solar cell at the operating point. Method according to at least one of the preceding claims, characterized in that in step A the luminescence radiation is photoluminescence radiation which is produced by illuminating the semiconductor structure ( 1 ) is generated with an excitation radiation. A method according to claim 11, characterized in that the spectrum of the excitation radiation substantially only wavelengths smaller than a cut-off wavelength λ _A , wherein λ _{A is selected} such that the penetration depth is less than 10%, in particular less than 5% of the thickness of the semiconductor structure ( 1 ). Method according to at least one of Claims 11 and 12, characterized in that the illumination of the semiconductor structure ( 1 ) from a lighting side which covers the front or the back ( 1a . 1b ) of the semiconductor structure ( 1 ) and that in the evaluations only luminescence radiation is taken into account which differs from the side of the semiconductor structure opposite the illumination side ( 1 ) is radiated. Method according to at least one of Claims 11 and 12, characterized in that at least two evaluations differ in that in one evaluation the semiconductor structure ( 1 ) from the front ( 1a ) and in the other evaluation the semiconductor structure ( 1 ) from the back ( 1b ) is illuminated. Measuring method for determining local voltage variations in a semiconductor structure ( 1 ), comprising the following method steps: i. Spatially resolved determination of the diffusion length and the surface recombination velocity of at least one surface of the semiconductor structure ( 1 ), wherein the determination by a method according to any one of claims 1 to 13, preferably in accordance with at least claim 9, ii. Calculation of the theoretically expected for a location-independent voltage luminescence radiation for each location of the step i. determined determination depending on the respective location in step i. determined diffusion length and surface recombination rate and iii. Forming the quotient of the measured in step i and in step ii. calculated luminescence radiation and calculation of the voltage distribution along the surface of the semiconductor structure ( 1 ), where an intensity variation by a factor of 2.72 corresponds to a voltage variation of 0.026V.