JP2006135196A - Method for measuring output of multi-junction photoelectric conversion device test cell - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for measuring an output characteristic precisely in a standard state of a group of multi-junction photoelectric conversion device by use of a dummy sunlight source at a low cost. <P>SOLUTION: In this measuring method, an output in the standard state of a test cell which is composed of a multi-junction photoelectric conversion device formed by stacking a plurality of element cells is measured as an output of a multi-junction photoelectric conversion device test cell measured under an arbitrary test light source. The method comprises the steps of obtaining the output in the standard state of a standard cell having a spectrum dependency equal to the test cell substantially, adjusting an illuminance of the test light source so as to obtain the output in the standard state of the standard cell, and measuring the output of the test cell under the test light source after adjustment of the illuminance. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an output measurement method for evaluating an output in a reference state of a photoelectric conversion element in which a plurality of element cells are stacked.

  In recent years, photoelectric conversion elements including thin film solar cells have also been diversified. For example, in addition to conventional amorphous thin film solar cells, crystalline thin film solar cells have been developed. As a method for improving the conversion efficiency of a thin film solar cell, there is a method in which two or more photoelectric conversion units are stacked to form a multi-junction type (or also called a tandem type). In this method, a front unit including a photoelectric conversion layer having a large band gap is disposed on the light incident side of the thin film solar cell, and a rear unit including a photoelectric conversion layer having a small band gap is sequentially disposed behind the unit. The photoelectric conversion is enabled over a wide wavelength range of the incident light, thereby improving the conversion efficiency of the entire solar cell. As described above, the multi-junction photoelectric conversion element has a structure in which photoelectric conversion units (hereinafter referred to as element cells) including a plurality of semiconductor junctions are stacked. The stacked element cells are formed in series connection or parallel connection.

  Examples of the multi-junction photoelectric conversion element include a solar cell, a photodiode, and a sensor. Examples of the semiconductor junction include a pn junction, a pin junction, and a MIS junction. Examples of semiconductor materials include crystalline, polycrystalline, microcrystalline, and amorphous materials. Semiconductor materials include Group 4 or compounds such as Si, SiGe, Ge, SiC, and C, Group 2-6 compounds such as GaAs, GaAlAs, and InP, CdTe, CdS, Cu2S, Cu2O, ZnO, ZnSe, and CuIn ( S, Se) 2, Cu (Ga, In) (S, Se) 2, compounds such as InGaN, organic semiconductors, or the above-mentioned compounds.

  It is very important to accurately measure the output of such a multi-junction photoelectric conversion element for the following reason. For example, in the output inspection at the time of shipment of the photoelectric conversion element, the rated output is the most important in terms of performance. For example, when the rated output is less than the standard value, it is regarded as a non-standard product. However, if the maximum output of the photoelectric conversion element to be shipped cannot be measured accurately, it cannot be clarified whether the standard value to be guaranteed is satisfied. In addition, in a situation where the output measurement error is large and the measurement result changes depending on the state of the measuring device, the yield of the photoelectric conversion element that satisfies the standard value will change even if a photoelectric conversion element of equivalent performance is manufactured. Even if stable manufacturing is performed, there is a possibility of giving the manufacturing worker a misunderstanding that the apparent yield is not stable. In other words, in such a state, the yield change due to the change in the manufacturing process tends to be insufficiently separated from the yield change due to the state of the measuring device, and the change in the output of the product and the change in the yield are likely to occur. Since it becomes ambiguous as an index of manufacturing, it is difficult to stabilize the manufacturing process.

As such a rated output, the output of the photoelectric conversion element in the reference state described below is usually used. That is, for example, output measurement (JISC 8934) of a solar cell is performed under the following conditions as a reference state.
(1) Cell temperature 25 ° C
(2) Spectral distribution AM1.5 global solar radiation standard sunlight (3) Irradiance 100mW / cm2
In the evaluation of a multijunction solar cell having spectral dependency, a method for confirming the spectral distribution is particularly important. The reference spectrum is defined by air mass and precipitable water, atmospheric turbidity, ozone content, albedo, etc., and the opportunity to obtain the reference spectrum from outdoor sunlight is very limited.

  However, it is technically difficult to accurately measure the output characteristics of a multi-junction photoelectric conversion element. The main reason is that the output current and the maximum output power of the multi-junction solar cell are limited by a photoelectric conversion unit (hereinafter simply referred to as an element cell) in which the output current is the smallest. For example, in a tandem solar cell in which two element cells are stacked and connected in series, the wavelength sensitivity band of each element cell is different. Change.

  Furthermore, in a multi-junction solar cell, measurement is difficult because not only the short-circuit current but also the fill factor and the open circuit voltage are affected at the same time depending on the degree of current limiting between the element cells. In other words, in order to obtain accurate output characteristics of multijunction solar cells under reference conditions, it is necessary to accurately reproduce the current values and current limit states that each element cell should output under the reference spectrum. There is. However, correction by spectral mismatch factor including mismatch of measurement light source spectrum with respect to reference spectrum, which was effective for obtaining accurate short-circuit current value in single junction solar cell, and mismatch of spectral sensitivity spectrum of measurement cell with respect to reference cell Therefore, it is difficult to carry out the measurement for the current of each element cell of the multijunction solar cell at the same time. Therefore, it is necessary to measure the output characteristics under a test light source having a spectral emission spectrum that matches the reference spectrum as much as possible.

  However, it is very difficult to obtain a spectral radiation spectrum that matches the reference spectrum even when outdoor sunlight or approximate solar light source is used as a test light source. The reference spectrum is defined as a spectrum in the case of atmospheric conditions in which air mass, atmospheric turbidity, precipitable water, and the like are defined, and opportunities for obtaining the reference spectrum from outdoor sunlight are very limited. Furthermore, since the spectrum of the approximate solar light source is merely approximated to the reference spectrum by correcting the spectral radiation spectrum of an available lamp with an optical filter, it is impossible to obtain the reference spectrum itself.

Furthermore, amorphous solar cells (such as a-Si solar cells and a-SiGe solar cells) and microcrystalline silicon (μc-Si solar cells) used as element cells of multi-junction solar cells are non-linear with respect to the irradiance of the light source. Is known as a large solar cell. The spectral sensitivity of the pseudo-element cell of a multi-junction solar cell is designed based on the spectral sensitivity spectrum of each element cell under the color bias of the multi-junction solar cell, but the irradiance of monochromatic light during spectral sensitivity measurement Is often measured in the range of 5 to 100 μW / nm / cm ² . Since the solar cell requires a high illuminance of 100 mW / cm ² in the standard state, the reference cell manufactured based on the spectral sensitivity spectrum at a low illuminance is used in a solar cell with a remarkable nonlinearity with respect to the irradiance. When the output characteristics of the solar cell under the reference condition are measured using, accuracy cannot be maintained. In the spectral sensitivity measurement of a single-junction solar cell, white light with an irradiance of 100 mW / cm ² is superimposed on monochromatic light, so that a spectral sensitivity spectrum at high illuminance can be obtained. In a junction solar cell, it is difficult to obtain a spectral sensitivity spectrum at high illuminance technically. Therefore, the output evaluation method of the multi-junction solar cell using the pseudo element cell of each element cell is essentially affected by the uncertainty based on the nonlinearity of the element cell. In particular, in the case of measurement under an approximate solar light source deviated from the standard spectrum, the uncertainty is increased.

  The following techniques have been proposed as a method for accurately measuring the output characteristics of a multi-junction solar cell.

  First, a plurality of pseudo-element cells simulating the spectral sensitivity of each element cell of a multi-junction solar cell are manufactured, and the pseudo-cell corresponding to the element cell whose current is limited under a pseudo-sunlight source composed of a single light source. A method of measuring output characteristics of a multi-junction solar cell after adjusting the illuminance of the light source so as to obtain a short-circuit current value (calibration value) under the reference state of the element cell has been proposed (Non-Patent Document 1). .

  In addition, by using an approximate solar light source capable of adjusting the spectral emission spectrum, each element cell constituting the multi-junction solar cell is obtained from a plurality of pseudo-element cells in order to obtain a current value that will occur under a reference state. At the same time, a technique (Non-Patent Document 2) has been proposed that attempts to accurately measure the output characteristics of a multijunction solar cell under a reference state by adjusting the spectral radiation spectrum so that a calibration value can be obtained. That is, by adjusting the irradiance of each wavelength band of the approximate solar light source using a single-junction pseudo-element cell having a relative spectral sensitivity substantially equivalent to the element cell constituting the multi-junction solar cell, The spectral condition for simultaneously obtaining the current value generated in the reference state by each element cell is to be obtained.

  In the measurement technique of the multijunction solar cell described above, the premise is that an approximate solar light source capable of adjusting the spectral irradiance is used. In Non-Patent Document 2, in order to obtain a current value obtained from each pseudo-element cell under a reference spectrum, the irradiation light of a lamp having an auxiliary spectrum is stacked on the irradiation surface of a xenon lamp provided with an air mass filter. The spectral emission spectrum is adjusted by changing the illuminance of each lamp. Specifically, in the case of measuring a large area multi-junction solar cell, by combining a plurality of xenon lamps and a plurality of halogen lamps and changing the illuminance of each lamp, Adjust the illuminance unevenness.

  In other words, it is practical to use an approximate solar light source for evaluating the output characteristics of a multi-junction solar cell, but many approximate solar light sources use a xenon lamp, and particularly in the wavelength band of 800 nm or more, the emission line spectrum The degree of match greatly decreases due to the influence. Since the spectral sensitivity of the bottom cell of the tandem solar cell has a main sensitivity in the range of 600 nm to 1100 nm, it is easily affected by a local band mismatch of the xenon lamp. Therefore, an approximate solar light source for measuring a multi-junction solar cell is designed to improve the degree of spectral coincidence by dividing the wavelength region of 800 nm or more using a filter or mirror and stacking the spectra of halogen lamps.

Furthermore, as a method of correcting the deviation from the reference state, the deviation from the reference state is quantified by using a pseudo-element cell having a spectral sensitivity that relatively matches the spectral sensitivity of each element cell, and the output characteristics of the test cell are determined. A measurement method (Patent Document 1) that corrects a value obtained under a reference state has been proposed. However, in order to improve the accuracy of the correction value, measurement results under a plurality of shifted irradiance spectra are required.
JP 2002-1111030 A K. Emery, Co. Ro. Osterwald, T. Glatfelter, J. Burdick, G. Vishup, Solar Cells, 24 (1988) 371-380 R. Shimokawa, F. Nagamine, M. Nakata, K. Fujisawa and Y. Hamakawa: Jpn. J. Appl. Phys. 28 (1989) L845

  The present invention has been made in view of the above-described situation, and the output of the multi-junction solar cell in the reference state is the same regardless of the light receiving area. The purpose is to accurately measure and evaluate using the.

  The measurement method of the present invention measures the output in a reference state of a test cell composed of a multi-junction photoelectric conversion element in which a plurality of element cells are stacked as the output of the multi-junction photoelectric conversion element test cell measured under an arbitrary test light source. A measurement method for obtaining an output in a reference state of a reference cell having a spectral dependence substantially equivalent to that of the test cell, the test light source so as to obtain an output in the reference state of the reference cell The method of measuring the output of the test cell under the test light source after adjusting the illuminance. The output in the reference state can be measured and evaluated accurately at low cost regardless of the light receiving area.

  The reference cell is an output current that can be generated under the reference state of one element cell with respect to two sets of element cells in which the output characteristic changes most greatly with respect to the spectrum change constituting the multi-junction photoelectric conversion element test cell. Measure the ratio of the current mismatch, which is the output characteristic obtained under the irradiation of the light source with the changed spectral conditions, so that the output current of the other element cell decreases by the same ratio when it is increased by a certain ratio. The value of the current mismatch ratio that maximizes the output is preferably equal to that of the multi-junction photoelectric conversion element test cell, so that the output of the multi-junction photoelectric conversion element test cell can be measured and evaluated more accurately.

  In particular, the reference cell may be selected from a test cell composed of a plurality of the multi-junction photoelectric conversion elements that are actual measurement objects as representative of their spectral dependence.

At that time, the selection measures the output characteristics of the test cell composed of the plurality of multi-junction photoelectric conversion elements under two sets of spectral conditions of a current mismatch ratio different from the current mismatch ratio under a reference state, It is possible to perform measurement with less error if it is performed by a method of selecting a multi-junction solar cell whose gradient ratio with respect to the current mismatch ratio of the output obtained under two sets of spectral conditions is closest to the average of them. Preferred In other words, the present invention comprises the following configuration as one means for achieving the above object.

  The measurement method according to the present invention is a method for evaluating the maximum output of a multi-junction photoelectric conversion element having a plurality of semiconductor junctions, and has substantially the same spectral dependence under irradiation of a light source having different spectral conditions. Using the reference cell, the output characteristic of the photoelectric conversion element is measured, and the maximum output in the reference state of the photoelectric conversion element is obtained from the output characteristic of the reference cell under the reference state. Even if the measurement is performed under a low-cost light source with a single lamp that cannot obtain the measured voltage, the measured photoelectric conversion element has the same spectral dependence as the reference cell, so the measured characteristics can be accurately measured using the output characteristics of the reference cell. The maximum output of the cell can be measured.

  In the reference cell according to the present invention, the dependence of the output characteristics on the change in current mismatch between the element cells when the output current that can be generated under the reference state of each element cell is changed is determined by the photoelectric conversion to be measured. Since it is substantially equivalent to that of the element and is selected as representative of the spectral dependence of a group of manufactured photoelectric conversion elements, output characteristics are accurately measured for a large number of manufactured photoelectric conversion elements. Characterized by being able to do it.

  The spectral dependence according to the present invention is very small from the output current that can be generated under the reference state for two sets of element cells in which the output characteristics change most greatly among the multi-junction photoelectric conversion elements having a plurality of semiconductor junctions. This is an output characteristic that is acquired under irradiation of a light source whose spectral conditions have been changed so that when one is changed by the amount, the other changes by the same amount of current with the opposite sign, and is acquired under the reference state. Therefore, it is possible to verify that the reference cell has substantially the same dependency as the measured cell under the spectrum conditions that can be assumed.

  The method of selecting a reference cell according to the present invention is a method of selecting a cell in which a current mismatch ratio between the two sets of element cells from which a peak position at the maximum spectrum-dependent output is obtained is an average of the group of photoelectric conversion elements. In addition, the output characteristics of the group of photoelectric conversion elements are measured under two sets of spectral conditions that give a current mismatch different from the current mismatch ratio under a reference state. This is a method of selecting a multi-junction solar cell whose gradient ratio with respect to the current mismatch ratio of the maximum output is closest to the average of a group. The method of selecting a multi-junction solar cell that is close to the average of the gradient ratios is to analyze the output characteristics of a general approximate solar light source by two different spectral emission spectra without a spectrum variable approximate solar light source. It is characterized by enabling selection of reference cells that represent the spectral dependence of a group of multi-junction solar cells.

  In addition, the measurement method of the present invention is substantially the same as a low-cost large-area single light source using a large-cost multi-area light source, without checking the reference spectrum with a pseudo-element cell. The output characteristics of a large-area multi-junction solar cell module can be accurately measured using a reference cell having the same spectral dependence.

  In addition, by the measurement method of the present invention, it is possible to accurately measure the output characteristics of a multi-junction solar cell using a reference cell having substantially the same spectral dependence under outdoor sunlight that is converted every moment. .

In addition, since it is confirmed that the spectrum dependence of the output characteristics of the reference cell of the present invention is substantially equal to that of the cell to be measured, it is possible to correct by using the output characteristics under the reference state of the reference cell. The output characteristics can be accurately measured under various spectral conditions.
In addition, in the reference cell selection method of the present invention, it is necessary to produce a pseudo element cell every time in the prior art, because of the difference in spectral dependence between products that occurs in the manufacturing process. It is possible to select a reference cell having dependency, and by using the selected reference cell, it is possible to accurately measure the output characteristics even for a group of multijunction solar cells having slightly different spectral dependencies. I can do it.

  Furthermore, in the present invention, there is provided a measuring method using a spectrum-dependent multijunction solar cell representative of a group of multijunction solar cells produced in large quantities in the manufacturing process as a reference cell, and a shunt component of each element cell. Because the reference cell that reflects the distribution of the diode factor and series resistance component is used, even if the spectral emission spectrum of the approximate solar light source changes, the conditions for measuring the maximum output under the reference state of the test cell It reproduces accurately. In addition, since the maximum output of the reference cell is obtained under a reference state with an irradiance of 100 mW / cm 2, the measurement result does not include uncertainty based on the nonlinearity of the element cell such as the above-described pseudo element cell. , Which enables accurate measurement.

  In other words, the measurement method of the present invention uses a multi-lamp type light source capable of spectral adjustment, and is substantially spectrum-dependent from a single light source having a different degree of spectrum matching without checking the reference spectrum in the pseudo-element cell. It is possible to accurately measure the output characteristics of a multi-junction solar cell using a reference cell having the same characteristics.

As a result of applying the above-described conventional technique to the measurement of a multi-junction photoelectric conversion element having a large area, the present inventor has found that there is a problem described below, and has devised the present invention. That is, (1) In the spectrum variable type approximate solar light source using a plurality of lamps as described above, a xenon lamp, a halogen lamp, or the like is used as a lamp type in order to sufficiently match the spectral emission spectrum with the reference spectrum. There is a need. Moreover, in order to obtain 1000 W / m ² that is the irradiance in the reference state at the irradiation surface position, it is necessary to combine a lamp with a high rated output, and when measuring a large-area multi-junction solar cell module, It is necessary to arrange a large number of lamps without uneven illuminance, and for example, it is necessary to provide individual power sources for the respective lamps and precisely adjust them. For this reason, such a large-area spectrum-variable approximate solar light source has a complicated structure and a large increase in manufacturing cost compared to a conventional single light source. Furthermore, because the rated output life of halogen lamps is about 100 hours, advanced control technology such as frequent lamp replacement and adjustment of illuminance each time is necessary to maintain the standard spectrum and predetermined illuminance unevenness. In addition, skill is required.

  The spectrally variable approximate solar light source can accurately measure the output characteristics of multijunction solar cells, but for the above reasons, it can be used for small-area multijunction solar cells developed in laboratories. As a main light source for a large-area solar cell module, it becomes very expensive.

  (2) In addition, in order to obtain a reference spectrum for measuring a multi-junction solar cell using outdoor sunlight, the above-described pseudo-element cell is used and simultaneously obtained from any pseudo-element cell under the reference spectrum. A method of waiting for a spectral condition for reproducing the current value is conceivable.

  However, the solar irradiance spectrum changes under the influence of air mass (atmospheric pressure and solar altitude, altitude) and atmospheric turbidity, falling water amount, ozone, albedo and the like. Even in the solar irradiance spectrum range of the day, the reference spectral condition is obtained about twice, and the measurement needs to be completed within a few minutes. Furthermore, depending on the season, the reference spectral conditions may not be obtained throughout the day. Therefore, it is very difficult to accurately obtain the output characteristics of the solar cell in the reference state using the sunlight spectrum.

(3) Furthermore, the output characteristics of a multi-junction solar cell must consider not only the short-circuit state of the element cells, but also the shunt component and series resistance component, diode factor, etc. that are affected by each element cell under a bias condition. In particular, the multijunction solar cell has a constant spectral dependence because the factors that affect the output characteristics of the multijunction solar cell described above change due to fluctuations in conditions such as the film forming process and the processing process during manufacture. Should result in a distribution. Therefore, a measurement that quantifies the deviation from the reference state using a pseudo-element cell that has a spectral sensitivity that relatively matches the spectral sensitivity of each element cell, and corrects the output characteristics of the test cell to a value obtained under the reference state. In the method (Patent Document 1), it is necessary to measure the deviation every time for all the multi-junction solar cells produced in large quantities in the manufacturing process, which is complicated and problematic in practicality. In other words, the method for obtaining the output characteristics under the reference mode by the method disclosed in Patent Document 1 is generally used for evaluating the output characteristics of multi-junction solar cells produced in large quantities on the production line. One solar cell to improve the measurement accuracy and correction accuracy based on the fact that a single pseudo solar light source cannot be used and the influence of the distribution of the pseudo light source deviation and the spectrum dependence of the multi-junction solar cell However, the outline of the measurement method of the present invention will be briefly described here. The method of the present invention sequentially obtains the short-circuit current (calibration value) in the reference state of the pseudo-element cell, which in turn consists of:
(2) The illuminance of the two-lamp type light source is adjusted with the calibration value of the pseudo element cell to obtain a reference state, and an output of the reference cell candidate in the reference state is obtained. In addition, from this state, a reference cell is selected from reference cell candidates among a plurality of test cells by performing measurement using a two-lamp light source while changing the spectrum condition (current mismatch ratio).
(3) For example, the illuminance of the test light source of the one-lamp type light source is adjusted so that the output P _MAX (measured by the two-lamp type light source) in the reference state of the reference cell is obtained.
(5) By measuring the test cell with a test light source of, for example, a one-lamp type light source after adjustment, an output (P _MAX , that is, maximum output) in the reference state of the test cell can be obtained.

  Hereinafter, the present invention will be described in more detail with reference to a solar cell as an example, showing the best mode of implementation.

  First, the spectral dependence of a multijunction solar cell will be described. When the element cells are connected in series, the output characteristics of the multi-junction solar cell are limited by the element cell that limits the current. Fig. 1 shows the spectral dependence of a tandem solar cell module in which an amorphous silicon solar cell as the top layer and a microcrystalline silicon solar cell as the bottom layer are connected in series, and a 100-stage integrated structure is formed on a glass substrate of 910 mm x 455 mm size. Showing sex. Under the approximate solar light source consisting of a xenon lamp, a halogen lamp, and an optical filter, using a pseudo element cell having a spectral sensitivity that relatively matches the spectral sensitivity of the top layer and the bottom layer of the element cell of the tandem solar cell, The irradiance of each lamp is changed. The pseudo element cell for the top provides a single-junction amorphous silicon solar cell with a plurality of colored glass filters, and substantially matches the relative spectral sensitivity of the top layer of the tandem solar cell. Furthermore, the pseudo element cell for the bottom: the single cell gives a color glass filter to the single-junction microcrystalline silicon solar cell, and substantially matches the relative spectral sensitivity of the bottom layer of the tandem solar cell. Each of It and Ib on the horizontal axis in FIG. 1 is obtained by dividing the short-circuit current value output by each pseudo-element cell under the approximate solar light source by the short-circuit current ground (calibration value) under the reference state. Ib / It corresponds to the current mismatch ratio of the tandem solar cell in which the photocurrents of the respective element cells are balanced in the reference state, and indicates a deviation from the reference state of the light source. For a tandem solar cell in which the photocurrent of each element cell is balanced in the reference state, due to the difference from the reference state of the light source, the top layer current is excessive when it is less than 1, and the bottom current is excessive when it is greater than 1. It is a state reflecting the state.

  FIG. 1 represents the spectral dependence of the test cell. , It and Ib are quantities reflecting the current value of the pseudo element cell. This represents the state of the light source. After adjusting the lamp illuminance so that the following (Equation 1) is satisfied, IV measurement of the tandem solar cell module is performed, and the change in the output characteristics is described.

ここで、ＩｔｍとＩｂｍは基準状態に近い但し青赤比があっていない状態の近似太陽光光源下でのトップ擬似要素セルとボトム擬似要素セルの短絡電流値である。Ｉｔ０とＩｂ０は基準状態に近く青赤比があっている状態のトップ擬似要素セルとボトム擬似要素セルの短絡電流値であって一般に校正値と呼ばれる。δは基準状態からの摂動を表す。

Here, Itm and Ibm are the short circuit current values of the top pseudo-element cell and the bottom pseudo-element cell under an approximate solar light source that is close to the reference state but has no blue-red ratio. It0 and Ib0 are short-circuit current values of the top pseudo-element cell and the bottom pseudo-element cell in a state where the blue-red ratio is close to the reference state and are generally called calibration values. δ represents perturbation from the reference state.

  The spectral dependence evaluation method based on the above equation is a technique for evaluating the spectral dependence of output characteristics as a perturbation from the reference state of a multijunction solar cell. is there.

  The test cell of FIG. 1 is a tandem solar cell, and a cell having the maximum short-circuit current (Isc) under the reference condition (Ib / It = 1) is selected and illustrated. Therefore, the top cell and the bottom cell are so-called current matching under the reference condition.

  In addition, for multi-junction solar cells having three or more layers of semiconductor junctions, even if three pseudo element cells are used, the current mismatch ratio is determined by using two pseudo element cells that have the largest spectral dependence. Spectral dependence can be evaluated with high accuracy and good reproducibility.

  Next, the reference cell will be described. In the present invention, the reference cell has substantially the same spectral dependence as the test cell. The relative spectral sensitivities of each element cell of the target multi-junction solar cell are substantially the same, and there are shunt components, diode factors, and series resistance components that affect the current-voltage characteristics when bias voltage is applied to each element cell. It is desirable that they are substantially the same and the linearity with respect to irradiance is substantially equivalent. The reference cell may be a combination of a group of photoelectric conversion elements that reproduce the characteristics of the element cells in an equivalent circuit. More preferably, it may be divided and separated from a test cell itself having a spectrum dependency representing a group of multi-junction solar cells as a target.

  The output characteristics of the reference cell under the reference state must be obtained by measurement under the condition in which the reference state is confirmed. The irradiance of the pseudo solar light source in the previous period needs to be adjusted so that one or more of Pmax, Isc, FF, and Voc under the reference state of the reference cell is reproduced.

  Further, a reference cell selection method will be described. The reference cell sorting method in the present invention is a method for sorting multijunction solar cells having an average spectral dependence based on the spectral dependence distribution of a group of manufactured multijunction solar cells.

  In other words, the reference cell sorting method according to the present invention selects a group of multijunction solar cells by selecting, as a reference cell, a multijunction solar cell that is closest to the average value of the gradient rate described later of the test cells composed of the group of multijunction solar cells. This is a method of selecting a reference cell that represents the spectral dependence of the battery. The difference in spectral sensitivity between the element cells and the current mismatch ratio under the bias voltage of the element cells are accurate. Provide specific methods for effective and effective selection.

  Specifically, when selecting an average for the spectral dependence of the maximum output, the current mismatch ratio corresponding to the peak position in the spectral dependence of the maximum output in FIG. 1 corresponds to the average value of a group of solar cells. Select a multi-junction solar cell as the reference cell. Preferably, a group of solar cells reflecting the population statistics of the multi-junction solar cells is randomly extracted, the previous spectrum dependence of the group of multi-junction solar cells is evaluated, and a peak current is obtained. A multi-junction solar cell having a maximum output peak at a position that matches the average mismatch ratio is selected as a reference cell.

  By using the sorting method according to the present invention, it is possible to sort a reference cell representing the spectrum dependence of a test cell with an inexpensive single-light-source pseudo-sunlight source. Pmax is measured using an approximate solar light source 1 having an irradiance distribution in which the current mismatch ratio is less than 1 and an approximate solar light source 2 having an irradiance distribution in which the current mismatch ratio is greater than 1, and the following ( Using Equation (2), (Equation 3), and (Equation 4), the gradient rate R in the vicinity of the reference spectrum condition (Ib / It = 1) is obtained.

ここで、Ｐｍａｘ１とＰｍａｘ２は、それぞれ近似太陽光光源１と近似太陽光光源２で測定した最大出力を表す。（Ｉｂ／Ｉｔ）１と（Ｉｂ／Ｉｔ）２はそれぞれ近似太陽光光源１と近似太陽光光源２の電流ミスマッチ比を表す。ＲａｖｅはＮ個の多接合太陽電池の勾配率の平均値、ｉは平均の勾配率に最も近い勾配率を有する多接合太陽電池の番号に対応する。勾配率Ｒは図２に破線で示した多接合太陽電池の最大出力のスペクトル依存性の基準スペクトル条件近傍の勾配率に相当する。

Here, Pmax1 and Pmax2 represent the maximum outputs measured by the approximate solar light source 1 and the approximate solar light source 2, respectively. (Ib / It) 1 and (Ib / It) 2 represent current mismatch ratios of the approximate solar light source 1 and the approximate solar light source 2, respectively. Rave corresponds to the average value of the gradient rates of the N multijunction solar cells, and i corresponds to the number of the multijunction solar cell having the gradient rate closest to the average gradient rate. The gradient rate R corresponds to the gradient rate in the vicinity of the reference spectral condition of the spectral dependence of the maximum output of the multijunction solar cell shown by the broken line in FIG.

  Different irradiance spectra using a multi-junction solar cell whose conversion efficiency in STC is known and the conversion efficiency of the multi-junction solar cell obtained by confirming the reference state (STC) using a conventional pseudo-element cell as a reference cell The effectiveness of this measurement method was confirmed by verifying the error from the conversion efficiency obtained below.

Example 1
A method of measuring the maximum output of the multijunction solar cell by using the multijunction solar cell of Example 1 as a reference cell will be described with reference to FIG.

  First, a pseudo element cell was produced. That is, a tandem solar cell having a laminated structure of a top layer (amorphous silicon cell) and a bottom layer (thin film microcrystalline silicon cell) on a glass substrate was produced. The obtained tandem solar cell was irradiated with monochromatic light while paying attention to the color bias light, thereby obtaining the relative spectral sensitivity characteristics of the top layer and the bottom layer shown in FIG.

  In order to produce a pseudo-element cell that relatively matches the spectral sensitivity of the top layer, a single-junction amorphous silicon solar battery was produced on a glass substrate, and a glass filter of CAW500 (made by HOYA) was attached to the upper surface of the cell. In order to improve the degree of coincidence of the relative spectral sensitivities, the film thickness of the amorphous silicon i layer was changed around 3000A.

  In addition, in order to produce a pseudo element cell that relatively matches the spectral sensitivity of the bottom layer, a single-junction thin-film microcrystalline silicon solar cell is produced on a glass substrate, and A71 (manufactured by ATG) and M30 (manufactured by HOYA). A glass filter was attached to the upper surface of the cell. In order to improve the degree of coincidence of the relative spectral sensitivity, the film thickness of the thin film microcrystalline silicon i layer was changed around 2 μm.

  The pseudo-element cells of the top layer and the bottom layer were stabilized by light irradiation, and then the short-circuit current was calibrated as a calibration value in the reference state by the Japan Quality Organization (JQA).

  Under the two-lamp type light source composed of a xenon lamp and a halogen lamp and having the spectral radiation distribution shown in FIG. 5, a calibration value priced by JQA is obtained from the top layer pseudo element cell and the bottom layer pseudo element cell. The illuminance of the xenon lamp and the halogen lamp was adjusted as shown.

  The tandem solar cell having the spectral sensitivity shown in FIG. 4 was obtained using the two-lamp type light source with the illuminance and the spectrum adjusted in this way while cooling the cell temperature to 25 ° C. with IV characteristics. When measured, a conversion efficiency of 10.56% was obtained.

  In order to evaluate the spectral dependence of the IV characteristics of the reference cell, the illuminance of the xenon lamp and the halogen lamp of the two-lamp light source was changed, and the tandem solar cell was measured. Specifically, a short-circuit current 1.1 times the calibration value from the top layer pseudo-element cell is set to a calibration value from the bottom layer pseudo-element cell so that the irradiance condition has a current mismatch ratio of 0.818. It adjusted so that 0.9 times of short circuit current might be obtained simultaneously, and measured the IV characteristic of the tandem type solar cell.

  By changing the spectral conditions of the two-lamp type light source in the current mismatch ratio range of 0.8 to 1.2, the spectral dependence shown by the solid line in FIG. 3 was obtained. Here, the range of the current mismatch ratio is based on the premise that the range of change in the current value of each element cell is about ± 10%. In the range where the current mismatch ratio is less than 0.8 or greater than 1.2, the output current of the element cell changes by 10% or more with respect to the current value obtained in the reference state. The reason is that it is no longer guaranteed. The current mismatch ratio at which the maximum output peak of each element cell was obtained was 1.0.

  Using the tandem solar cell as a reference cell, the maximum output of two types of tandem solar cells was measured. The test cell 1 having a different film thickness from the test cell 2 of the tandem solar cell having the same film thickness configuration as that of the reference cell substantially matches the relative spectral sensitivities of the high approximate two-lamp light source and each element cell. When the IV measurement was performed in the reference state (STC) using the pseudo-element cell and the calibration value, the conversion efficiencies of the test cell 1 were 10.53% and the test cell 2 was 10.56%.

Next, while intentionally changing the irradiance spectrum of the two-lamp type light source so that the current mismatch ratio becomes 0.8 to 1.2 using the value of the pseudo-element cell, Only the light intensity was adjusted so that a conversion efficiency of 10.56% was obtained from the reference cell.

Then, IV measurement was performed with respect to the test cell 1 test cell 2, and the result of having measured conversion efficiency is shown in Table 1.

試験セル２では、全てのスペクトル条件下でＳＴＣで得られる変換効率と同一の測定結果が得られた。また、試験セル１では、電流ミスマッチ比が１となる条件では、ＳＴＣと同一の変換効率が得られたが、異なるスペクトル条件下でも−１％から＋１．４％の範囲の誤差での変換効率の測定結果となった。

In test cell 2, measurement results identical to the conversion efficiency obtained with STC were obtained under all spectral conditions. Moreover, in the test cell 1, the same conversion efficiency as that of the STC was obtained under the condition that the current mismatch ratio was 1, but the conversion efficiency with an error in the range of −1% to + 1.4% even under different spectral conditions. The measurement results were as follows.

  By the way, when the spectral dependence of the test cell 1 and the test cell 2 was evaluated, the result of the black circle and the white circle shown in FIG. 3 was obtained. The test cell 1 has a current mismatch ratio at which a conversion efficiency peak is obtained of 0.96, which is 0.04 lower than that of the reference cell. From FIG. 3, the conversion efficiency of sample 1 is higher than that of the reference cell under spectral conditions where the current mismatch ratio is less than 0.96, while the conversion efficiency of sample 1 is low under spectral conditions where the current mismatch ratio is greater than 1. It turns out that it becomes. However, it can be confirmed that the difference is less than ± 2%. On the other hand, since test cell 2 has the same spectral dependence as the reference cell, it showed the same conversion efficiency under all spectral conditions.

  From the above, by using the measurement method according to the present invention, it was possible to measure the exact conversion efficiency under the spectral conditions close to the reference spectrum by using the reference cell having substantially the same spectral dependence as the sample.

(Comparative Example 1)
By the conventional illuminance adjustment method shown in Non-Patent Document 1, the irradiance of the simulated solar light source is adjusted using a pseudo-element cell corresponding to the element cell whose current is limited under the spectral irradiance of the given light source. Later, the conversion efficiency of the multi-junction solar cell was measured.

  Table 2 shows the results of measuring the conversion efficiency of sample 2 under the irradiance of a pseudo solar light source with current mismatch ratios of 0.8, 0.9, 1.0, 1.1, and 1.2. Yes.

電流ミスマッチ比として０．８、０．９、１．０が得られる分光放射スペクトル条件下では、試料２はボトム層で電流制限されるため、ボトム用擬似要素セルを基準セルとして用い、その校正値が得られるように放射照度を調整した。その後に試料２の出力特性を測定すると、１１．１４％、１０．９７％、１０．５６％の変換効率となった。

Under the spectral emission spectrum conditions where current mismatch ratios of 0.8, 0.9, and 1.0 are obtained, sample 2 is current limited at the bottom layer, so the bottom pseudo-element cell is used as the reference cell and its calibration is performed. Irradiance was adjusted to obtain a value. Thereafter, when the output characteristics of the sample 2 were measured, the conversion efficiencies were 11.14%, 10.97%, and 10.56%.

  In addition, the sample 2 is current-limited at the top layer under the spectral emission spectrum conditions in which current mismatch ratios of 1.0, 1.1, and 1.2 are obtained, so the top pseudo-element cell is used as the reference cell. When the irradiance was adjusted so that the calibration value was obtained and the output characteristics were measured, the conversion efficiencies were 10.56%, 10.73%, and 11.06%.

  Under the spectral emission spectral condition of the light source that obtains a current mismatch ratio of 1.0, it substantially matches the reference spectral condition, so it matches the conversion efficiency in STC, but deviates from the reference spectral condition. Under the conditions, there is a large error in the measured conversion efficiency. This is because, in the rate limiting element cell short-circuit current matching method, the FF value of each element cell of the tandem solar cell differs for each light source spectrum condition, which causes an error in the conversion efficiency measurement result. Show. In particular, in a multi-junction solar cell including element cells having excellent device quality, the spectral dependence of the FF value is increased, so that it is considered that the measurement error of conversion efficiency tends to increase.

(Example 2)
In Example 2, 20 samples are randomly extracted as reference cell candidates from a tandem solar cell module product including a top layer amorphous silicon cell and a bottom layer thin film microcrystalline silicon cell, and representative spectra are obtained from these reference cell candidates. Two other products with similar spectral dependence were measured using two with dependence as reference cells.

  FIG. 5 shows the results of the spectrum dependence of the 20 reference cell candidates extracted. Table 3 shows the current mismatch ratio at which the maximum output of these reference cell candidates has a peak. An average of 50 current mismatch ratios of 0.950 was obtained.

これらの基準セル候補２０個から、最も平均値に近い基準セルとして、ＨＭＡ１１４２２４２２１、及びＨＭＡ１１４２２４０２２をさらに選別し基準セルとしたこれらは図５で、３と４で示したスペクトル依存性を有し、以下では基準セル３、及び基準セル４で参照する。基準セル３および基準セル４の基準状態での出力特性をＪＱＡで測定した結果、それぞれ最大出力として４１．２Ｗおよび４０．５Ｗが得られた。

From these 20 reference cell candidates, HMA114224221 and HMA114224422 are further selected as reference cells that are closest to the average value, and have the spectrum dependence shown in FIGS. 3 and 4 in FIG. Then, the reference cell 3 and the reference cell 4 are referred to. As a result of measuring the output characteristics of the reference cell 3 and the reference cell 4 in the reference state by JQA, 41.2 W and 40.5 W were obtained as the maximum outputs, respectively.

  The output of 450 tandem solar cell products was measured using a xenon pulse light source that cannot change the spectrum of a single lamp. Incidentally, the current mismatch ratio of the xenon pulse type light source was 0.90.

  Specifically, first, the illuminance of the xenon pulse type light source was adjusted so that the output of the reference cell 3 was 41.2W. Next, the output of 450 tandem solar cell products was measured under the xenon pulse type light source in that state. This measurement result is taken as the result of the measurement method of the present invention.

  For comparison, the maximum output of the same 450 tandem solar cell products was measured under a large-area two-lamp light source in which the illuminance of the xenon lamp and halogen lamp was adjusted so that the current mismatch ratio was 1.00. did. This measurement result is considered as a result of measurement in a substantially standard state.

  When the results of the measurement method of the present invention were compared with the results measured in the almost standard state, they matched with an error within 1%. In other words, it was found that the output of the multi-junction photoelectric conversion element can be measured accurately and inferior to an expensive two-lamp light source by using the measurement method of the present invention even if an inexpensive one-lamp light source is used. .

  The illuminance adjustment was performed using a tandem solar cell deviating from the average current mismatch ratio indicated by 5 in FIG. 5 as a reference cell, and the maximum output measurement results of 450 modules were obtained. When compared with the results measured in the state, it was confirmed that there was an error of 5% or more. In other words, by using the reference cell selected by the reference cell selection method according to the present invention, even if an inexpensive one-lamp type light source that cannot perform spectrum adjustment is used, only the illuminance adjustment is performed. The maximum output of the battery could be measured accurately.

(Example 3)
In Example 3, a general pseudo-solar light source is used when the spectral dependence of a tandem solar cell randomly selected at the time of selecting a reference cell and the current mismatch ratio at which the peak of output characteristics can be obtained cannot be measured. A method for selecting a reference cell representative of the spectral dependence of a group of tandem solar cells is illustrated.

  For the 20 tandem solar cells randomly extracted in Example 2, the maximum output was measured using the xenon pulse light source used in the manufacturing process. The current mismatch ratio of the light source at the time of measurement was 0.96. It was 1.07 when the air mass filter of the said light source was replaced | exchanged, the spectral radiation spectrum of the previous light source was changed, and the current mismatch ratio was measured. The maximum output of the 20 tandem solar cells was measured using the light source after air mass filter replacement, and the gradient rate R shown in the following (Equation 2) was evaluated.

表４は、２０個のタンデム型太陽電池の勾配率Ｒを示したものである。勾配率Ｒの平均値として−０．１６が得られ、平均値に近いタンデム型太陽電池としてＨＭＡ１１４２２４２２１、及びＨＭＡ１１４２２４０２２が選別された。これらのタンデム型太陽電池は実施例２のスペクトル依存性の評価結果から得られたものと同一である。本発明による基準セルの選別法により、一般的な擬似太陽光光源を用いて一群のタンデム型太陽電池のスペクトル依存性を代表するものが選別する事ができた。

Table 4 shows the gradient rate R of 20 tandem solar cells. -0.16 was obtained as the average value of the gradient rate R, and HMA114224221 and HMA114224222 were selected as tandem solar cells close to the average value. These tandem solar cells are the same as those obtained from the spectral dependence evaluation results of Example 2. According to the reference cell selection method of the present invention, a typical pseudo-solar light source can be used to select a representative of the spectrum dependence of a group of tandem solar cells.

It is an example of the spectrum dependence of a tandem type solar cell. It is the schematic regarding the reference | standard cell selection method of this invention. It is a figure which shows the measurement error at the time of using the reference cell of a different spectrum dependence. It is an example of the relative spectral sensitivity of a tandem type solar cell. It is an example of the spectral dependence of the maximum output of a group of tandem solar cells.

Explanation of symbols

1 Spectral dependence of a tandem solar cell having a film thickness configuration different from that of the reference cell 2 Spectral dependence of a tandem solar cell having the same film thickness configuration as that of the reference cell 3 Maximum output current mismatch ratio of a group of tandem solar cells Spectral dependence of solar cells with spectral dependence close to the average value (thick solid line)
4 Spectral dependence of a solar cell having a spectral dependence close to the average value of the maximum output current mismatch ratio of a group of tandem solar cells (thick dashed line)
5 Spectral dependence of solar cells with spectral dependence deviating from the average value of the maximum output current mismatch ratio of a group of tandem solar cells

Claims (4)

A measurement method for measuring an output in a reference state of a test cell composed of a multi-junction photoelectric conversion element in which a plurality of element cells are stacked as an output of a multi-junction photoelectric conversion element test cell measured under an arbitrary test light source,
Obtaining a reference cell output of a reference cell having substantially the same spectral dependence as the test cell;
Adjusting the illuminance of the test light source so as to obtain an output in a reference state of the reference cell;
A method for measuring an output of a multi-junction photoelectric conversion element test cell, comprising the step of measuring the output of the test cell under the test light source after adjusting the illuminance.   The reference cell is an output current that can be generated under the reference state of one element cell with respect to two sets of element cells in which the output characteristic changes most greatly with respect to the spectrum change constituting the multi-junction photoelectric conversion element test cell. Measure the ratio of the current mismatch, which is the output characteristic obtained under the irradiation of the light source with the changed spectral conditions, so that the output current of the other element cell decreases by the same ratio when it is increased by a certain ratio. The method of measuring an output of a multi-junction photoelectric conversion element test cell according to claim 1, wherein a value of the current mismatch ratio at which the output becomes maximum is equivalent to that of the multi-junction photoelectric conversion element test cell.   3. The reference cell according to claim 1, wherein the reference cell is selected from a test cell composed of a plurality of the multi-junction photoelectric conversion elements as a representative of their spectral dependence. Output measurement method of multi-junction photoelectric conversion element test cell.   The screening is performed by measuring output characteristics of the test cell including the plurality of the multi-junction photoelectric conversion elements under two spectral conditions of a current mismatch ratio different from the current mismatch ratio under a reference state. The multi-junction photoelectric conversion device test according to claim 3, wherein the multi-junction photoelectric conversion element test according to claim 3 is performed by selecting a multi-junction solar cell having a gradient ratio with respect to a current mismatch ratio of an output obtained under a spectral condition that is closest to an average thereof. Cell output measurement method.

Images