- Email: [email protected]
Fast large area reflectivity scans of wafers and solar cells with high spatial resolution
Available online at www.sciencedirect.com Available online at www.sciencedirect.com
ScienceDirect ScienceDirect
Energy Procedia 00 (2017) 000–000 Available online www.sciencedirect.com Available online atatwww.sciencedirect.com Energy Procedia 00 (2017) 000–000
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Energy (2017) 000–000 166–173 EnergyProcedia Procedia124 00 (2017) www.elsevier.com/locate/procedia
7th International Conference on Silicon Photovoltaics, SiliconPV 2017 7th International Conference on Silicon Photovoltaics, SiliconPV 2017
Fast large area reflectivity scans of wafers and solar cells with high Fast large area reflectivity scans of wafers and solar cells with high spatial resolution The 15th International Symposium on District Heating and Cooling spatial resolution b c b Andreas Schüttaa,the Stefanie Wahlbb, Sylke Meyerthe Hirsch b, Jens c, Dominik Lauschb Assessing feasibility of using heat demand-outdoor Andreas Schütt , Stefanie Wahl , Sylke Meyer , Jens Hirsch , Dominik Lausch CELLOscan GbR, Gravelottestr. 5, 24116 Kiel, Germany; www.celloscan.de; [email protected]; Tel.: +49(0)1737600436 temperature function forPhotovoltaics a Germany; long-term district heat demand forecast CELLOscan GbR, Gravelottestr. 5, 24116 www.celloscan.de; [email protected]; Tel.: +49(0)1737600436 Fraunhofer Center for SiliconKiel, CSP, Otto-Eißfeldt-Str. 12, 06120 Halle, Germany a a
b b
c Fraunhofer Center for Silicon Photovoltaics CSP, 12,Germany 06120 Halle, Germany Hochschule Anhalt, Bernburger Str.Otto-Eißfeldt-Str. 55, 06366 Köthen, c a,b,c a Anhalt, Bernburger a c Hochschule Str. 55, 06366bKöthen, Germany
I. Andrić a
*, A. Pina , P. Ferrão , J. Fournier ., B. Lacarrière , O. Le Correc
IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal
b Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France Abstract c Abstract Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France A solar cell local characterization (CELLO) set-up is modified to measure reflectivity maps of any objects in a non-destructive A solar celldifferent local characterization (CELLO) is modified measure any in a non-destructive way. Four laser wavelengths (BLUEset-up 403 nm, RED 630tonm, IR 830reflectivity nm and SIRmaps 934 of nm) areobjects applicable. This paper will way. Four laser wavelengths (BLUE 403 nm, RED 630tonm, IR 830previously nm and SIR 934 nm)with are an applicable. Thissphere, paper will present thedifferent measurement principle, the reflectivity calibration samples analyzed integrating and Abstract present the measurement principle, to Compared samples previously analyzed with techniques an integrating and demonstrate the reflectivity analysistheofreflectivity wafers andcalibration solar cells. to other measurement like sphere, integrating demonstrate reflectivity analysis of wafers solar Compared to other techniques integrating spheres, this the approach has two advantages: First,and being fastcells. on large areas (15.6 cm measurement x 15.6 cm sample size; 1like million pixels, District this heating are commonly addressed in the as one of the effective solutions decreasing the spheres, approach hastotwo advantages: First, being fast literature on large areas (15.6 cmmost xresolving 15.6 cmdetails sample 1scans million pixels, measurement timenetworks 10 min 1 hour depending on the accepted noise-level) and second, insize; zoomfor with high greenhouse gas emissions theThis building sector. These systems require highsecond, investments are in returned through thehigh heat measurement time 10µm-spot min tofrom 1size). hour depending on the accepted noise-level) and resolving details zoom scans with local resolution (10 may allow optimizing processes, where reflectivity is which a key parameter like texturization, sales.resolution Due to (10 the µm-spot changedcoating climate and building renovation policies, heat demand the future decrease, local size). This may allow optimizing processes, where reflectivity isindependently a keyinparameter likecould texturization, SiN-PECVD-Antireflective orconditions wafer cutting and cleaning processes. Furthermore, measured reflectivity prolonging the investment return period. SiN-PECVD-Antireflective coating or wafer cutting and cleaning processes. Furthermore, independently measured reflectivity maps may be helpful for CELLO photo-impedance analysis in the future. The main scope of this is to assess the feasibility of using heat demand – outdoor temperature function for heat demand maps may be helpful for paper CELLO photo-impedance analysis in thethe future. forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 © 2017 Authors. Published Published by by Elsevier Elsevier Ltd. Ltd. ©buildings 2017 The Thethat Authors. vary in both construction period and typology. Three weather scenarios (low, high) and three district © 2017 The Authors. Published by Elsevier Ltd. of SiliconPV 2017 under responsibility Peer review the conference committee of PSE PSEmedium, AG. Peer review by by the scientific scientific conference committee of SiliconPVdeep). 2017 under responsibility of AG. renovation scenarios were developed (shallow, intermediate, To estimate the error, obtained heat demand values were Peer review by the scientific conference committee of SiliconPV 2017 under responsibility of PSE AG. compared with results from a dynamic heat demand model, previously developed and validated by the authors. Keywords: CELLO; reflectivity; wafer; solar cells; bathlifetime; texturization The results showed that when onlysolar weather change is considered, Keywords: CELLO; reflectivity; wafer; cells; bathlifetime; texturizationthe margin of error could be acceptable for some applications (the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). 1.scenarios, Introduction value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the 1.The Introduction decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and Reflectivity is one key parameter in the characterization of solar cells. It is well-known that texturing, renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the Reflectivitycoating is onedeposition key parameter inandthe characterization of solar cells. Itmaterial is well-known that impact texturing, antireflection of SiN,could grain orientation a major coupled scenarios). The values suggested be used to modify of themulti-crystalline function parameters for thehave scenarios considered, on and antireflection coating deposition of SiN, and grain orientation of multi-crystalline material have a major impact on the reflectance. Reflectivity losses are direct losses of the photocurrent of a certain wavelength. The front reflectivity improve the accuracy of heat demand estimations.
the losses are direct losses EQE of theand photocurrent of aquantum certain wavelength. The via frontthe reflectivity R isreflectance. linked withReflectivity the external quantum efficiency the internal efficiency IQE equation RIQE( is linked with the external quantum efficiency EQE and the internal quantum efficiency IQE via the equation )=EQE( )/(1-R( ))[1]. byStandard reflectivity measurement techniques like IQE, or global reflectance © 2017 The Authors. Published Elsevier Ltd. )=EQE( )/(1-R( ))[1]. Standard reflectivity measurement techniques like IQE, or global reflectance IQE( Peer-review under the Scientific Committee of The 15thaveraged International Symposium measurements, orresponsibility integrating of spheres give just integral values over an area on of District at leastHeating some and squaremeasurements, or integrating spheres give just integral values averaged over an area of at least some squareCooling.
1876-6102 2017demand; The Authors. Published Elsevier Ltd. Keywords:©Heat Forecast; Climatebychange 1876-6102 The Authors. Published by Elsevier Ltd. Peer review©by2017 the scientific conference committee of SiliconPV 2017 under responsibility of PSE AG. Peer review by the scientific conference committee of SiliconPV 2017 under responsibility of PSE AG.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer review by the scientific conference committee of SiliconPV 2017 under responsibility of PSE AG. 10.1016/j.egypro.2017.09.317
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173 Andreas Schütt/ Energy Procedia 00 (2017) 000–000
167
millimeters. Additionally, there are some commercially LBIC-systems which measure the local reflectivity, too. However in this paper, we will present an adapted CELLO set-up for much faster reflectivity measurements. Another advantage of this approach is that the measurement is contactless and by this not restricted to solar cells. Any object can be simply used. This paper presents the modified CELLO set-up, explains the calibration procedure and finally, gives some first examples of the technique. Furthermore, independent measured reflectivity data will also be very helpful in the future to improve the CELLO photo-impedance analysis. Nomenclature BLUE CELLO IR RED SIR
laser with 403 nm wavelength solar CELl LOcal characterization laser with 830 nm wavelength laser with 630 nm wavelength laser with 934 nm wavelength
2. Measurement principle Fig. 1 presents the modified CELLO set-up [2] for reflectance measurements. Objects, here a sample wafer, are placed on the Au coated Cu chuck. A laser beam is scanned across the sample. The reflected light is collected via a vertically arranged solar cell mini-module that is contacted via a four probe arrangement. This is the first big difference compared to the set-up for standard CELLO measurements. The laser beam is intensity modulated with a perturbation frequency f of around 6 kHz. The collected signal is analyzed via a lock-in routine resulting in an amplitude map of the sample. For reflectance measurements the sample holder room is covered with diffusively light reflecting material, imitating the function of an integrating sphere and increasing the signal yield of the reflected laser light, which is the second big modification of the standard CELLO set-up. A typical map has a size of 1000 x 1000 pixels, with 200 µm being the length of one pixel. The smallest possible pixel length is 50 µm. The measurement time for this typical map is 10 minutes to one hour, depending on how many periods of the lock-in measurement are used to average.
C
R
CELLO unit with DAB W
S
Si-Mini-module
laser optic
Sample wafer Au coated Cu chuck
measurement feedback
software
Fig. 1. Schematic of a modified CELLO set-up for reflectivity measurements. The reflected laser beam is collected via a vertical placed solar cell mini-module that is connected via a 4 probe set-up (C: counter electrode; R: reference electrode; W: working electrode; S: sense electrode) to the CELLO unit with the data acquisition board (DAB).
Andreas Schütt/ Energy Procedia 00 (2017) 000–000
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173
168
3. Calibration For calibration, a solar cell is used, that has been previously characterized via an integrating sphere [1], cf. “reference solar cell” in Fig. 2. The global integral reflectance values in % at the four different wavelengths are used to calibrate the CELLO amplitude maps, cf. Table 1. The solar cell as calibration reference has some drawbacks: it is quite big and brittle. Thus, for easier handling in further measurements, a surface structured Al-block is used as a reference sample for calibration, instead of the reference cell, that can be simply placed next to the chuck. Thus the calibrated reflectivity values of the Al-block are used as reference for calibrating the reflectivity maps, cf. Table 1.
Paper: white paint A
Paper: white paint B
Al foil dI1b Data Sample : File : Mean Val. :
Paper: white paint C
rf1165_amada_mitt 18.10 %
Al block Reference solar cell -1.00 %
1 cm
54.00 %
Fig. 2. BLUE laser (403 nm) reflectivity map after calibration of various objects. Table 1. For calibration routine the maps are multiplied with the scaling factor. Reflectivity values of the Al block and the paints are given as examples. LASER
Blue RED IR SIR
Amplitude Reference solar cell (µA)
Reflectivity Reference solar cell [2] (%)
Scaling factor map
Reflectivity Al block
Reflectivity Paint A
Reflectivity Paint B
Reflectivity Paint C
(%/µA)
(%)
(%)
(%)
(%)
39.65 6.41 14.12 14.57
20 1.25 3.33 5.41
0.5044 0.1950 0.2358 0.3713
26.84 23.79 20.61 19.44
73.71 64.71 62.13 74.31
66.67 60.21 58.48 69.76
60.24 52.67 53.08 63.77
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173 Andreas Schütt/ Energy Procedia 00 (2017) 000–000
169
4. Examples of reflectivity scans 4.1. Object and paint selection of the sample room Fig. 2 and Table 1 present a calibrated reflectivity scan of various objects, including the reference solar cell, the reference Al-block, a piece of Al-foil and several pieces of paper painted with white paint of different compositions. Here, white latex paint is used mixed with various compositions of BaS, the standard paint for integrating spheres. Paint A shows the highest reflectance and is selected for the room walls as diffused light reflector. The slight inhomogeneity has only a negligible effect on the global diffusively reflected light. The optimal white paint increased the signal of the back reflected light and thus increases the measurement accuracy and reduces noise. 4.2. Alkaline textured solar cell
Fig. 3. Blue laser (401 nm) reflectivity map of an alkaline textured solar cell. Zoom picture (black frame) reveals details due to isotropically etched grains.
Fig. 3 presents the blue reflectance map of an alkaline textured solar cell. The zoom map reveals details of the isotropically etched grains, resulting in reflectivity differences. This information can be very useful in the future to optimize the texturization processes. In general, such reflectance maps of solar cells give precious additional information for the interpretation of standard CELLO measurements: The CELLO photocurrent maps are typically partially affected by reflectivity inhomogeneity. In the analysis of CELLO photo-impedance measurements via a sophisticated fit model, also the spectral reflectivity maps can be calculated. However, with the new reflectivity measurement approach presented in this paper we have access to these maps, independently measured. Using this independently measured reflectivity maps as input parameter for the impedance analysis might result in more
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173 Andreas Schütt/ Energy Procedia 00 (2017) 000–000
170
convincing photo-impedance results. We will publish a direct comparison of the two differently measured reflectivity maps and their impact to the CELLO photo-impedance analysis in the future elsewhere. 4.3. Black silicon surface texturization Black Silicon samples were produced by a maskless ICP etching process with SF6 and O2 at 5 ° C [3]. For this, a plasma etch chamber from Oxford Instruments (Plasmalab System 100 ICP 65) was used with 13.56 MHz frequency and 600 Watt ICP power. The plasma etching time was varied between 0 minutes (reference) and 10 minutes. An overview CELLO reflectivity scan with the Blue laser is presented in Fig. 4. Within this magnification, a strongly reduced reflectivity across the complete area can be observed. Fig. 5 presents reflectivity zoom maps of various wavelengths. The maps reveal various inhomogeneities and spots having higher reflectivity. These spots can be attributed to atmospheric particles (dirt) since the samples were present in air for a few days or even particles from the aluminum chamber wall, which gets etched by the fluorine radicals as well. Scratches due to the manual handling are also very likely. Global reflectivity measurements were performed using the UV/VIS/NIR spectrometer Lambda 1050 from Perkin Elmer, cf. Fig. 6. Table 2 summarizes the results. In general, there is a good correlation between the integral reflectance values and the plasma textured regions with the CELLO reflectivity measurements. However, the good positions of the CELLO high resolution reflectivity maps are a few percentage points better than the integral reflectivity values, cf. Table 2. One possible explanation for this might be, that the “bad” spots shift the mean value of the integral maps. Thus the CELLO maps high resolution maps helps to better reveal the full potential of the black Si material. Further, it is obvious that lateral inhomogeneity’s of the plasma texturing process can be revealed excellently by using the CELLO reflectivity technique.
Zoom dI1b Data Sample : File : Mean Val. :
1 cm
0.00 %
rf1234_csp_black_ 6.12 %
15.00 %
Fig. 4. BLUE laser (403 nm) reflectivity map of a black Si sample. The sample is cleaved. The red area in the center is the measured reflection of the Cu chuck.
Andreas Schütt/ Energy Procedia 00 (2017) 000–000
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173
171
Table 2. Comparison of global reflection measurements and CELLO-reflectivity measurements. LASER
Wavelength (nm)
Black Si sample GRM integral value reflectivity (%)
CELLO Bad spot reflectivity (%)
CELLO normal spot reflectivity (%)
Blue RED IR SIR
401 650 830 934
7.92 3.95 3.23 2.95
15 32 15 10
5.3 1.9 1.2 1.0
RED
Blue dI1a Data Sample : File : Mean Val. :
15%
dI2a Data Sample : File : Mean Val. :
rf1235_csp_black_ 5.54 %
32%
5.3% 3.50 %
1 mm
7.60 %
1.9% 1.00 %
1 mm
IR dI3a Data Sample : File : Mean Val. :
15% 1 mm
dI4a Data Sample : File : Mean Val. :
rf1235_csp_black_ 1.16 %
10%
1.2% 0.30 %
1.90 %
rf1235_csp_black_ 1.97 %
2.90 %
SIR rf1235_csp_black_ 1.09 %
1%
1 mm
Fig. 5. Zoom map of Fig. 4 of the black Si sample with various wavelengths.
0.20 %
2.00 %
Andreas Schütt/ Energy Procedia 00 (2017) 000–000
172
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173
Fig. 6. Global reflection measurement (GRM) of the black Si wafer (red line). Reference (black line): as cut wafer with no texture and no antireflection coating.
4.4. Silicon wafer sawn by diamond wire Fig. 7 presents the IR reflectivity map of a diamond wire sawn silicon wafer. Visible are thin horizontal lines in slicing direction, which are also visible with naked eye. However the vertical stripes, not visible with naked eye, are unexpected. Their cause is unknown. They are visible on both wafer sides. These patterns might have a negative effect on the efficiency of the final processed solar cell, since similar patterns have been previously be observed on defected solar cells. In the future, systematic CELLO reflectivity measurements of wafers in comparison with CELLO efficiency analysis measurements might clarify this issue.
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173 Andreas Schütt/ Energy Procedia 00 (2017) 000–000
173
IR dI2b Data Sample : File : Mean Val. :
6.00 %
1 cm
rf1253_csp_wafer_ 10.28 %
14.00 %
Fig. 7. IR laser (830 nm) reflectivity map of a diamond sawn silicon wafer.
5. Importance and future deployment Spatially resolved reflectance maps of solar wafers and cells are very helpful for the optimization of various processes that influence the reflectivity, like surface texturization and antireflection coatings. The method may be very interesting for the optimization of bath life times of the texturization process. A correlation of texturization bath life times with reflectance maps may help to find the optimal time to replace the bath, thus reduce processing and disposal costs. Additionally, reflectance maps might become an interesting independent measurement parameter map as input for the CELLO photoimpedance analysis that is capable of measuring local diffusion lengths, bulk lifetime, and back surface recombination velocity [2, 4]. 6. References [1] Montesdeoca Santana A. Texturization process of monocrystalline silicon with Na2CO3/NaHCO3 solution for solar cells. Dissertation, Basica de la Universidad de La Laguna, 2012. [2] Schütt A. Ortsaufgelöste Charakterisierung in der Photovoltaik mit der CELLO-Technik (Solar Cell Local Characteriszation), Dissertation, CAU Kiel, 2015. [3] Hirsch J, Gaudig M, Bernhard N, Lausch D.Optoelectronic properties of Black-Silico generated through inductively coupled plasma (ICP) processing for crystalline silicon solar cells, Applied Surface Science 374, 252 (2016). [4] Wagner JM, Schütt A, Carstensen J, Föll H. Series resistance contribution of majority carriers in CELLO impedance analysis: Influence of wafer thickness variation, Solar Energy Mater. Solar Cells 146, 129 (2016).
ScienceDirect ScienceDirect
Energy Procedia 00 (2017) 000–000 Available online www.sciencedirect.com Available online atatwww.sciencedirect.com Energy Procedia 00 (2017) 000–000
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Energy (2017) 000–000 166–173 EnergyProcedia Procedia124 00 (2017) www.elsevier.com/locate/procedia
7th International Conference on Silicon Photovoltaics, SiliconPV 2017 7th International Conference on Silicon Photovoltaics, SiliconPV 2017
Fast large area reflectivity scans of wafers and solar cells with high Fast large area reflectivity scans of wafers and solar cells with high spatial resolution The 15th International Symposium on District Heating and Cooling spatial resolution b c b Andreas Schüttaa,the Stefanie Wahlbb, Sylke Meyerthe Hirsch b, Jens c, Dominik Lauschb Assessing feasibility of using heat demand-outdoor Andreas Schütt , Stefanie Wahl , Sylke Meyer , Jens Hirsch , Dominik Lausch CELLOscan GbR, Gravelottestr. 5, 24116 Kiel, Germany; www.celloscan.de; [email protected]; Tel.: +49(0)1737600436 temperature function forPhotovoltaics a Germany; long-term district heat demand forecast CELLOscan GbR, Gravelottestr. 5, 24116 www.celloscan.de; [email protected]; Tel.: +49(0)1737600436 Fraunhofer Center for SiliconKiel, CSP, Otto-Eißfeldt-Str. 12, 06120 Halle, Germany a a
b b
c Fraunhofer Center for Silicon Photovoltaics CSP, 12,Germany 06120 Halle, Germany Hochschule Anhalt, Bernburger Str.Otto-Eißfeldt-Str. 55, 06366 Köthen, c a,b,c a Anhalt, Bernburger a c Hochschule Str. 55, 06366bKöthen, Germany
I. Andrić a
*, A. Pina , P. Ferrão , J. Fournier ., B. Lacarrière , O. Le Correc
IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal
b Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France Abstract c Abstract Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France A solar cell local characterization (CELLO) set-up is modified to measure reflectivity maps of any objects in a non-destructive A solar celldifferent local characterization (CELLO) is modified measure any in a non-destructive way. Four laser wavelengths (BLUEset-up 403 nm, RED 630tonm, IR 830reflectivity nm and SIRmaps 934 of nm) areobjects applicable. This paper will way. Four laser wavelengths (BLUE 403 nm, RED 630tonm, IR 830previously nm and SIR 934 nm)with are an applicable. Thissphere, paper will present thedifferent measurement principle, the reflectivity calibration samples analyzed integrating and Abstract present the measurement principle, to Compared samples previously analyzed with techniques an integrating and demonstrate the reflectivity analysistheofreflectivity wafers andcalibration solar cells. to other measurement like sphere, integrating demonstrate reflectivity analysis of wafers solar Compared to other techniques integrating spheres, this the approach has two advantages: First,and being fastcells. on large areas (15.6 cm measurement x 15.6 cm sample size; 1like million pixels, District this heating are commonly addressed in the as one of the effective solutions decreasing the spheres, approach hastotwo advantages: First, being fast literature on large areas (15.6 cmmost xresolving 15.6 cmdetails sample 1scans million pixels, measurement timenetworks 10 min 1 hour depending on the accepted noise-level) and second, insize; zoomfor with high greenhouse gas emissions theThis building sector. These systems require highsecond, investments are in returned through thehigh heat measurement time 10µm-spot min tofrom 1size). hour depending on the accepted noise-level) and resolving details zoom scans with local resolution (10 may allow optimizing processes, where reflectivity is which a key parameter like texturization, sales.resolution Due to (10 the µm-spot changedcoating climate and building renovation policies, heat demand the future decrease, local size). This may allow optimizing processes, where reflectivity isindependently a keyinparameter likecould texturization, SiN-PECVD-Antireflective orconditions wafer cutting and cleaning processes. Furthermore, measured reflectivity prolonging the investment return period. SiN-PECVD-Antireflective coating or wafer cutting and cleaning processes. Furthermore, independently measured reflectivity maps may be helpful for CELLO photo-impedance analysis in the future. The main scope of this is to assess the feasibility of using heat demand – outdoor temperature function for heat demand maps may be helpful for paper CELLO photo-impedance analysis in thethe future. forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 © 2017 Authors. Published Published by by Elsevier Elsevier Ltd. Ltd. ©buildings 2017 The Thethat Authors. vary in both construction period and typology. Three weather scenarios (low, high) and three district © 2017 The Authors. Published by Elsevier Ltd. of SiliconPV 2017 under responsibility Peer review the conference committee of PSE PSEmedium, AG. Peer review by by the scientific scientific conference committee of SiliconPVdeep). 2017 under responsibility of AG. renovation scenarios were developed (shallow, intermediate, To estimate the error, obtained heat demand values were Peer review by the scientific conference committee of SiliconPV 2017 under responsibility of PSE AG. compared with results from a dynamic heat demand model, previously developed and validated by the authors. Keywords: CELLO; reflectivity; wafer; solar cells; bathlifetime; texturization The results showed that when onlysolar weather change is considered, Keywords: CELLO; reflectivity; wafer; cells; bathlifetime; texturizationthe margin of error could be acceptable for some applications (the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). 1.scenarios, Introduction value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the 1.The Introduction decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and Reflectivity is one key parameter in the characterization of solar cells. It is well-known that texturing, renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the Reflectivitycoating is onedeposition key parameter inandthe characterization of solar cells. Itmaterial is well-known that impact texturing, antireflection of SiN,could grain orientation a major coupled scenarios). The values suggested be used to modify of themulti-crystalline function parameters for thehave scenarios considered, on and antireflection coating deposition of SiN, and grain orientation of multi-crystalline material have a major impact on the reflectance. Reflectivity losses are direct losses of the photocurrent of a certain wavelength. The front reflectivity improve the accuracy of heat demand estimations.
the losses are direct losses EQE of theand photocurrent of aquantum certain wavelength. The via frontthe reflectivity R isreflectance. linked withReflectivity the external quantum efficiency the internal efficiency IQE equation RIQE( is linked with the external quantum efficiency EQE and the internal quantum efficiency IQE via the equation )=EQE( )/(1-R( ))[1]. byStandard reflectivity measurement techniques like IQE, or global reflectance © 2017 The Authors. Published Elsevier Ltd. )=EQE( )/(1-R( ))[1]. Standard reflectivity measurement techniques like IQE, or global reflectance IQE( Peer-review under the Scientific Committee of The 15thaveraged International Symposium measurements, orresponsibility integrating of spheres give just integral values over an area on of District at leastHeating some and squaremeasurements, or integrating spheres give just integral values averaged over an area of at least some squareCooling.
1876-6102 2017demand; The Authors. Published Elsevier Ltd. Keywords:©Heat Forecast; Climatebychange 1876-6102 The Authors. Published by Elsevier Ltd. Peer review©by2017 the scientific conference committee of SiliconPV 2017 under responsibility of PSE AG. Peer review by the scientific conference committee of SiliconPV 2017 under responsibility of PSE AG.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer review by the scientific conference committee of SiliconPV 2017 under responsibility of PSE AG. 10.1016/j.egypro.2017.09.317
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173 Andreas Schütt/ Energy Procedia 00 (2017) 000–000
167
millimeters. Additionally, there are some commercially LBIC-systems which measure the local reflectivity, too. However in this paper, we will present an adapted CELLO set-up for much faster reflectivity measurements. Another advantage of this approach is that the measurement is contactless and by this not restricted to solar cells. Any object can be simply used. This paper presents the modified CELLO set-up, explains the calibration procedure and finally, gives some first examples of the technique. Furthermore, independent measured reflectivity data will also be very helpful in the future to improve the CELLO photo-impedance analysis. Nomenclature BLUE CELLO IR RED SIR
laser with 403 nm wavelength solar CELl LOcal characterization laser with 830 nm wavelength laser with 630 nm wavelength laser with 934 nm wavelength
2. Measurement principle Fig. 1 presents the modified CELLO set-up [2] for reflectance measurements. Objects, here a sample wafer, are placed on the Au coated Cu chuck. A laser beam is scanned across the sample. The reflected light is collected via a vertically arranged solar cell mini-module that is contacted via a four probe arrangement. This is the first big difference compared to the set-up for standard CELLO measurements. The laser beam is intensity modulated with a perturbation frequency f of around 6 kHz. The collected signal is analyzed via a lock-in routine resulting in an amplitude map of the sample. For reflectance measurements the sample holder room is covered with diffusively light reflecting material, imitating the function of an integrating sphere and increasing the signal yield of the reflected laser light, which is the second big modification of the standard CELLO set-up. A typical map has a size of 1000 x 1000 pixels, with 200 µm being the length of one pixel. The smallest possible pixel length is 50 µm. The measurement time for this typical map is 10 minutes to one hour, depending on how many periods of the lock-in measurement are used to average.
C
R
CELLO unit with DAB W
S
Si-Mini-module
laser optic
Sample wafer Au coated Cu chuck
measurement feedback
software
Fig. 1. Schematic of a modified CELLO set-up for reflectivity measurements. The reflected laser beam is collected via a vertical placed solar cell mini-module that is connected via a 4 probe set-up (C: counter electrode; R: reference electrode; W: working electrode; S: sense electrode) to the CELLO unit with the data acquisition board (DAB).
Andreas Schütt/ Energy Procedia 00 (2017) 000–000
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173
168
3. Calibration For calibration, a solar cell is used, that has been previously characterized via an integrating sphere [1], cf. “reference solar cell” in Fig. 2. The global integral reflectance values in % at the four different wavelengths are used to calibrate the CELLO amplitude maps, cf. Table 1. The solar cell as calibration reference has some drawbacks: it is quite big and brittle. Thus, for easier handling in further measurements, a surface structured Al-block is used as a reference sample for calibration, instead of the reference cell, that can be simply placed next to the chuck. Thus the calibrated reflectivity values of the Al-block are used as reference for calibrating the reflectivity maps, cf. Table 1.
Paper: white paint A
Paper: white paint B
Al foil dI1b Data Sample : File : Mean Val. :
Paper: white paint C
rf1165_amada_mitt 18.10 %
Al block Reference solar cell -1.00 %
1 cm
54.00 %
Fig. 2. BLUE laser (403 nm) reflectivity map after calibration of various objects. Table 1. For calibration routine the maps are multiplied with the scaling factor. Reflectivity values of the Al block and the paints are given as examples. LASER
Blue RED IR SIR
Amplitude Reference solar cell (µA)
Reflectivity Reference solar cell [2] (%)
Scaling factor map
Reflectivity Al block
Reflectivity Paint A
Reflectivity Paint B
Reflectivity Paint C
(%/µA)
(%)
(%)
(%)
(%)
39.65 6.41 14.12 14.57
20 1.25 3.33 5.41
0.5044 0.1950 0.2358 0.3713
26.84 23.79 20.61 19.44
73.71 64.71 62.13 74.31
66.67 60.21 58.48 69.76
60.24 52.67 53.08 63.77
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173 Andreas Schütt/ Energy Procedia 00 (2017) 000–000
169
4. Examples of reflectivity scans 4.1. Object and paint selection of the sample room Fig. 2 and Table 1 present a calibrated reflectivity scan of various objects, including the reference solar cell, the reference Al-block, a piece of Al-foil and several pieces of paper painted with white paint of different compositions. Here, white latex paint is used mixed with various compositions of BaS, the standard paint for integrating spheres. Paint A shows the highest reflectance and is selected for the room walls as diffused light reflector. The slight inhomogeneity has only a negligible effect on the global diffusively reflected light. The optimal white paint increased the signal of the back reflected light and thus increases the measurement accuracy and reduces noise. 4.2. Alkaline textured solar cell
Fig. 3. Blue laser (401 nm) reflectivity map of an alkaline textured solar cell. Zoom picture (black frame) reveals details due to isotropically etched grains.
Fig. 3 presents the blue reflectance map of an alkaline textured solar cell. The zoom map reveals details of the isotropically etched grains, resulting in reflectivity differences. This information can be very useful in the future to optimize the texturization processes. In general, such reflectance maps of solar cells give precious additional information for the interpretation of standard CELLO measurements: The CELLO photocurrent maps are typically partially affected by reflectivity inhomogeneity. In the analysis of CELLO photo-impedance measurements via a sophisticated fit model, also the spectral reflectivity maps can be calculated. However, with the new reflectivity measurement approach presented in this paper we have access to these maps, independently measured. Using this independently measured reflectivity maps as input parameter for the impedance analysis might result in more
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173 Andreas Schütt/ Energy Procedia 00 (2017) 000–000
170
convincing photo-impedance results. We will publish a direct comparison of the two differently measured reflectivity maps and their impact to the CELLO photo-impedance analysis in the future elsewhere. 4.3. Black silicon surface texturization Black Silicon samples were produced by a maskless ICP etching process with SF6 and O2 at 5 ° C [3]. For this, a plasma etch chamber from Oxford Instruments (Plasmalab System 100 ICP 65) was used with 13.56 MHz frequency and 600 Watt ICP power. The plasma etching time was varied between 0 minutes (reference) and 10 minutes. An overview CELLO reflectivity scan with the Blue laser is presented in Fig. 4. Within this magnification, a strongly reduced reflectivity across the complete area can be observed. Fig. 5 presents reflectivity zoom maps of various wavelengths. The maps reveal various inhomogeneities and spots having higher reflectivity. These spots can be attributed to atmospheric particles (dirt) since the samples were present in air for a few days or even particles from the aluminum chamber wall, which gets etched by the fluorine radicals as well. Scratches due to the manual handling are also very likely. Global reflectivity measurements were performed using the UV/VIS/NIR spectrometer Lambda 1050 from Perkin Elmer, cf. Fig. 6. Table 2 summarizes the results. In general, there is a good correlation between the integral reflectance values and the plasma textured regions with the CELLO reflectivity measurements. However, the good positions of the CELLO high resolution reflectivity maps are a few percentage points better than the integral reflectivity values, cf. Table 2. One possible explanation for this might be, that the “bad” spots shift the mean value of the integral maps. Thus the CELLO maps high resolution maps helps to better reveal the full potential of the black Si material. Further, it is obvious that lateral inhomogeneity’s of the plasma texturing process can be revealed excellently by using the CELLO reflectivity technique.
Zoom dI1b Data Sample : File : Mean Val. :
1 cm
0.00 %
rf1234_csp_black_ 6.12 %
15.00 %
Fig. 4. BLUE laser (403 nm) reflectivity map of a black Si sample. The sample is cleaved. The red area in the center is the measured reflection of the Cu chuck.
Andreas Schütt/ Energy Procedia 00 (2017) 000–000
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173
171
Table 2. Comparison of global reflection measurements and CELLO-reflectivity measurements. LASER
Wavelength (nm)
Black Si sample GRM integral value reflectivity (%)
CELLO Bad spot reflectivity (%)
CELLO normal spot reflectivity (%)
Blue RED IR SIR
401 650 830 934
7.92 3.95 3.23 2.95
15 32 15 10
5.3 1.9 1.2 1.0
RED
Blue dI1a Data Sample : File : Mean Val. :
15%
dI2a Data Sample : File : Mean Val. :
rf1235_csp_black_ 5.54 %
32%
5.3% 3.50 %
1 mm
7.60 %
1.9% 1.00 %
1 mm
IR dI3a Data Sample : File : Mean Val. :
15% 1 mm
dI4a Data Sample : File : Mean Val. :
rf1235_csp_black_ 1.16 %
10%
1.2% 0.30 %
1.90 %
rf1235_csp_black_ 1.97 %
2.90 %
SIR rf1235_csp_black_ 1.09 %
1%
1 mm
Fig. 5. Zoom map of Fig. 4 of the black Si sample with various wavelengths.
0.20 %
2.00 %
Andreas Schütt/ Energy Procedia 00 (2017) 000–000
172
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173
Fig. 6. Global reflection measurement (GRM) of the black Si wafer (red line). Reference (black line): as cut wafer with no texture and no antireflection coating.
4.4. Silicon wafer sawn by diamond wire Fig. 7 presents the IR reflectivity map of a diamond wire sawn silicon wafer. Visible are thin horizontal lines in slicing direction, which are also visible with naked eye. However the vertical stripes, not visible with naked eye, are unexpected. Their cause is unknown. They are visible on both wafer sides. These patterns might have a negative effect on the efficiency of the final processed solar cell, since similar patterns have been previously be observed on defected solar cells. In the future, systematic CELLO reflectivity measurements of wafers in comparison with CELLO efficiency analysis measurements might clarify this issue.
Andreas Schütt et al. / Energy Procedia 124 (2017) 166–173 Andreas Schütt/ Energy Procedia 00 (2017) 000–000
173
IR dI2b Data Sample : File : Mean Val. :
6.00 %
1 cm
rf1253_csp_wafer_ 10.28 %
14.00 %
Fig. 7. IR laser (830 nm) reflectivity map of a diamond sawn silicon wafer.
5. Importance and future deployment Spatially resolved reflectance maps of solar wafers and cells are very helpful for the optimization of various processes that influence the reflectivity, like surface texturization and antireflection coatings. The method may be very interesting for the optimization of bath life times of the texturization process. A correlation of texturization bath life times with reflectance maps may help to find the optimal time to replace the bath, thus reduce processing and disposal costs. Additionally, reflectance maps might become an interesting independent measurement parameter map as input for the CELLO photoimpedance analysis that is capable of measuring local diffusion lengths, bulk lifetime, and back surface recombination velocity [2, 4]. 6. References [1] Montesdeoca Santana A. Texturization process of monocrystalline silicon with Na2CO3/NaHCO3 solution for solar cells. Dissertation, Basica de la Universidad de La Laguna, 2012. [2] Schütt A. Ortsaufgelöste Charakterisierung in der Photovoltaik mit der CELLO-Technik (Solar Cell Local Characteriszation), Dissertation, CAU Kiel, 2015. [3] Hirsch J, Gaudig M, Bernhard N, Lausch D.Optoelectronic properties of Black-Silico generated through inductively coupled plasma (ICP) processing for crystalline silicon solar cells, Applied Surface Science 374, 252 (2016). [4] Wagner JM, Schütt A, Carstensen J, Föll H. Series resistance contribution of majority carriers in CELLO impedance analysis: Influence of wafer thickness variation, Solar Energy Mater. Solar Cells 146, 129 (2016).