This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JPHOT.2017.2731800, IEEE
Photonics Journal
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 1

Absolute electroluminescence imaging diagnosis

of GaAs thin-film solar cells
XiaoBo Hu, Tengfei Chen, Juanjuan Xue, Guoen Weng, Shaoqiang Chen, Hidefumi Akiyama, and
Ziqiang Zhu

 recently [1-13]. By measuring the spatial distribution of the EL

Abstract—A spatially-resolved absolute electroluminescence intensity from the top surface of solar cells, it is easy to find
(EL) imaging method was utilized to analyze the photovoltaic failures from the EL images such as electrode faults and cell
properties and resistive loss properties of a GaAs thin-film solar cracks which has been used in maximizing module
cell. The I-V relation was extrapolated from the absolute EL
efficiency measurements in conjunction with the
manufacturing yields such as in crystalline silicon cells [7-10].
external-quantum-efficiency (EQE) measurements, the EL Furthermore, it is possible to extract spatially resolved
extrapolated I-V relation has a merit over the conventional I-V information about the electronic material properties of solar
relation measured with a solar simulator that it could eliminate the cells such as minority-carrier diffusion length, diode
series resistance effect caused by external probe contact. Then, the performance, series resistance, shunts and local junction voltage
mapping of the internal voltage of the solar cell and the sheet from the EL images [7-24]. Recently, an absolute EL method
resistance of the window-layer of the solar cell were obtained from
the calibrated absolute EL imaging method. Finally, optic-electro [25] has been demonstrated to characterize solar cells which is
conversion losses of the solar cell including radiative loss, based on the basic reciprocity relationship [15, 16] between EL
nonradiative loss, thermalization loss, transmission loss and emission in light-emitting-diode (LED) operation and the
junction loss were quantified given by the EL and EQE external quantum efficiency (EQE) in solar-cell operation. This
measurements. method involves in obtaining measurements of the absolute EL
intensity to directly evaluate the internal current-voltage (I-V)
Index Terms—Luminiescence and fluorescence, imaging systems.
properties of solar cells, and no extra parametric adjustments or
fittings are necessary [25]. Based on this method, an absolute
EL imaging method has been developed to quantitative
I. INTRODUCTION
mapping the open-circuit voltage of Si solar cells and modules
E lectroluminescence (EL) measurement is a powerful tool to
characterize solar cells which has gained much attention
[26]. However, both of the work [25, 26] made an assumption
which considered the properties of the solar cells were spatially
uniform on the surface. However, practical solar cells have
This work was supported in part by the Recruitment Program of Global more or less spatial inhomogeneity [27] and a more accurate
Experts (1000 Talent Plan) of China, Natural National Science Foundation of method is necessary.
China (NSFC) (61604055), and Shanghai Pujiang Program (16PJ1402600) in
China. This work was also partly supported by KAKENHI No.15H03968 and In this study, an absolutely calibrated electroluminescence
No.26390075 from JSPS, the Photon Frontier Network Program of MEXT, imaging method was proposed which also considered the
OPERANDO-OIL, and NEDO in Japan. inhomogeneity over the surface of the solar cell. Based on the
Xiaobo Hu is with the Department of Electronic Engineering, East China
Normal University, Shanghai, 200241 China (e-mail: huxiaobo@
method, the mapping of the internal voltage of the solar cell, the
ee.ecnu.edu.cn). sheet resistance of the window-layer of the solar cell were
Tengfei Chen is with the Department of Electronic Engineering, East China obtained. While in conjunction with the EQE measurements, the
Normal University, Shanghai, 200241 China (e-mail: 51151213013@ I-V properties and the quantitative optic-electro losses were
stu.ecnu.edu.cn).
Juanjuan Xue is with the Department of Electronic Engineering, East China also obtained.
Normal University, Shanghai, 200241 China (e-mail: 52151213019@
stu.ecnu.edu.cn). II. EXPERIMENTAL DETAILS
Guoen Weng is with the Department of Electronic Engineering, East China
Normal University, Shanghai, 200241 China (e-mail: In this study, a mono-crystalline GaAs solar cell was used as
egweng@ee.ecnu.edu.cn). the testing sample. The total area of the solar cell is 4×2 cm2. Fig.
Shaoqiang Chen is with the Department of Electronic Engineering, East
China Normal University, Shanghai, 200241 China (phone: 021-5434-5425;
1 shows the schematic of the experimental setup for the
fax: 021-5434-5425; e-mail: sqchen@ ee.ecnu.edu.cn). absolutely calibrated EL measurements of the GaAs solar cell.
Hidefumi Akiyama is with the Institute for Solid State Physics (ISSP), the In Fig. 1(a1), a silicon charge-coupled device (CCD) camera
University of Tokyo, Chiba 2778581, Japan, and also with AIST-UTokyo
with 4800×6400 pixels was set above the solar cell to measure
Advanced Operando-Measurement Technology Open Innovation Laboratory,
Chiba, 2778589, Japan (e-mail: golgo@issp.u-tokyo.ac.jp). the EL images of the solar cell. The forward current was
Ziqiang Zhu is with East China Normal University, Shanghai, 200241 injected by a current-voltage source. The exposure time of the
China (e-mail: zqzhu@ee.ecnu.edu.cn). CCD varied from 20 ms to 2 s depending on the injection

1943-0655 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JPHOT.2017.2731800, IEEE
Photonics Journal
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 2

current. Then in Fig. 1(b1), a mask was put closely above the be seen from Fig. 2(b) that the maximum difference of the
cell which only allow part of the EL emitted from the cell absolute EL intensity is about 1 order and many small local
surface to pass through and measured by the CCD camera. The defects could be seen which indicate the inhomogeneity of the
shape of the transparent part of the mask was designed to be solar cell.
square with a side length of 0.5 cm. Then, a calibrated Si
photodiode was put above the transparent part of the mask in
Fig. 1(b1) in the face-to-face geometry as Fig. 1(c1) shows to
measure the absolute EL power. The shape of the effective area
of the Si photodiode is a circle with a diameter of 0.85 cm which
could totally cover the transparent part of the mask. The
absolute EL power was measured and denoted by φ (photon flux)
as Fig. 1(c2) shows. The EL intensity measured by CCD is a
relative quantity, the EL intensity of each pixel measured by
CCD is denoted by aj as Fig. 1(b2) shows. Then a calibrated
constant C was defined which could calibrate the EL intensity
measured by CCD to an absolute intensity, then we have
C   / a j
as Fig. 1(b2) shows. Finally, the absolute EL
intensity of each pixel Φi (photon flux) could be calibrated to be
Φi=C· Ai, where Ai is the EL intensity of each pixel measured by Fig. 2. EL image measured by CCD camera before calibration with relative EL
CCD before calibration. intensity (a); and after calibration with absolute EL emission rate (b).
The I-V curve of the GaAs solar cell was measured under
AM1.5G, 1-sun illumination condition using a solar simulator at Fig. 3 (a) shows the image of the absolute EL intensity of the
room temperature. The short-circuit current density (Jsc) was GaAs solar cell under various injection current density, it is
25.3 mA/cm2, the open-circuit voltage (Voc) was 0.996 V, and obviously to see that higher injection current density leads to
the energy conversion efficiency (η) was 19.4%. The EQE of the higher EL emission intensity. Fig. 3(b) shows the absolute EL
solar cell was measured by a quantum efficiency measurement intensity across the cell along the x-direction (as the dash arrow
system (ORIEL IQE200, Newport Corp.) where the excitation shows in the inset) for three different forward injection current
spot radius of the EQE measurement was 0.8 mm, and the spot densities. It could be seen that the EL intensity gradually
was positioned at the cell center to ensure that the cell electrodes decreased along the x-direction for all current densities. It is
were avoided. The EL spectra was measured by a fiber probe easy to understand this point since as the distance from the left
connected to a spectrometer which is composed of a current injection electrode gradually increased along the
monochromator and a cooled silicon charge-coupled device x-direction, the voltage V(x) gradually decreased due to the
(CCD). resistance effect of the top p-GaAs layer and thus leading the EL
intensity also decreased. G. T. Koishiyev [28] proposed a lateral
current model which included the sheet resistance of the top
layer ρ to illustrate the decreasing of the EL intensity along the
x-direction as follows,
A
 ( x)  I  L x 
 1  ( )[1  (1  )2 ]
 (0)  2 W L  (1)
where  is the EL intensity, I is the total injection current
(mA), β=q/AkT and A is the diode quality factor. L and W
are the cell length (x-direction) and width respectively. The
red solid line shows the fitting result by using Eq. (1) which
considered a sheet resistance ρ of the p-GaAs top layer
Fig. 1. The schematic of the experimental setup for the absolutely calibrated EL
measurements of the GaAs solar cell. with the value of (0.1±0.02) Ω/□ and the diode quality
factor A with the value of (1.8±0.05).

III. RESULTS AND DISCUSSIONS

Fig. 2(a) shows the EL image with forward injection current
of 15 mA/cm2 measured by CCD camera without calibration
with EL intensity of arbitrary unit; Fig. 2(b) shows the
calibrated image of the absolute EL emission rate with unit of
photons/(s·cm2) based on the steps illustrated in Fig. 1. It can

1943-0655 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JPHOT.2017.2731800, IEEE
Photonics Journal
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 3

distributed grid line electrodes. However, a relatively big valley

was seen in Fig. 4(c) along the y-direction as the dash line
rectangle shows, this is due to the broken of the grid electrode at
the corresponding position on the surface of the cell where a
line-shape region with low intensity along the x-direction could
be seen obviously in the left images as the dash-line rectangle
shows [27]. The voltage drop across the electrode broken region
decreased with decreasing current density, which indicated the
local series but not shunt resistance increased, since the effect of
shunt resistance on the voltage is more pronounced for smaller
injection current densities [29].

Fig. 3. (a) Images of the absolute EL intensity of GaAs solar cell under different
forward injection current density. (b) Horizontal line scans of the absolute EL
intensity (blue circle lines) across the cell (x-direction) as shown by the inset,
the red solid lines are the the fitting results.

According to the reciprocity theorem established by U. Rau

[15], the EL intensity emi (photons/(s·cm2)) of a solar cell
emitted at any position “i” (the CCD camera pixel) from the
solar cell’s surface is given by
qVi
emi ( E )  Qei ( E )bb ( E )exp( )
(2) kT
where Qei(E) is the EQE of the solar cell, E is the photon energy, Fig. 4. (a) The internal voltage images of GaAs solar cell under forward
q is the elementary charge, k is the Boltzmann constant and T is injection current density of 7.5, 15 and 37.5 mA/cm2,respectively; (b)
the Kelvin temperature, Vi is the internal junction voltage and internal voltage across the cell in the x-direction as the dash arrow in the
inset shows; (c) internal voltage across the cell in the y-direction as the
bb is the spectral photon density of a black body, which dash arrow in the inset shows.
depends on Planck’s constant h and the vacuum speed c of light,
which is given by U. Rau [15] stated that a solar cell that has the
theoretical maximum power conversion efficiency will
2 E 2 E
bb ( E )  3 2 exp( ) also act as an LED with the maximum possible
hc kT (3) luminescence efficiency, the solar-cell EQE and the
Thus, the internal voltage Vi of each pixel could be obtained external EL quantum efficiency (ηemLED) in LED operation
based on Eq. (1) as following shows, were directly linked by a reciprocity relation. The EL
kT emi quantum efficiency (ηemLED) is defined as the ratio between
Vi  ln 
q  Qei  EL Eg bb ( E )dE the number of emitted photons from the surface and the
(4) total injection electrons as follows,
where <Qei>EL presents the average value of the EQE over the em
EL emission spectra. Fig. 4(a) shows three images of the emLED 
internal voltage distribution over the solar cell’s surface with
J inj q
(5)
forward injection current density of 7.5, 15, 37.5 mA/cm2 
according to Eq. (4). Here we assume the EQE is spatially where em (photons/(s·cm2)) is the total emitted photon density
uniform. Fig. 4(b) shows the internal voltage across the cell in from the top surface of the solar cell, Jinj is the injection current
the x-direction (as the dash arrow shows in the inset) under density. Fig. 5 shows the measured EL quantum efficiency as a
different forward injection current density. It is obviously to see function of the injection current density Jinj, the EL quantum
that as the injection current density gradually increases, the efficiency was found to increase gradually with increasing
voltage drop across the cell also increases, the voltage drop was injection current density, which indicates an increased radiative
considered mainly due to the sheet resistance of the top layer. recombination rate at increased carrier densities [30]. In Fig. 5,
Fig. 4(c) shows the internal voltage across the cell in the both the results measured by Si-photodiode and CCD camera
y-direction (as the dash arrow shows in the inset) under different are presented, the EL intensity of CCD imaging was obtained by
forward injection current density. There is no obvious voltage spatial integration of the CCD images and then the absolute
drop across the y-direction for all the injection current densities. value was calibrated by the Si-photodiode, then the EL quantum
The small periodical waving of the voltage across the efficiency was calculated and the result coincided well with that
y-direction is due to the shading of light by the periodically measured by Si-photodiode which verified the reliability of the

1943-0655 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JPHOT.2017.2731800, IEEE
Photonics Journal
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 4

measurements. The inset shows the measured absolute EL with a solar simulator (the blue open circles). The difference
spectra of the solar cell with various injection current density, between the I-V curves measured by the two methods was
the peak energy is around 1.42 eV which is estimated as the considered mainly due to the effect of series resistance (Rs),
bandgap energy of the GaAs cell [25]. The EL quantum which may originate from the cell itself including the resistance
efficiency is obtained through the integrations of the EL peak in of the base bulk, the base contact and the external probe contact.
the absolute EL spectra. The I-V curve measured by the EL method after considering a
lump resistance RL (the red dash dot line) is very close to the I-V
curve measured with a solar simulator, here the value of RL is
estimated as 1 Ω cm2. Also, the ideal I-V curve in the radiative
recombination limit condition is shown (the green dot line), the
difference between the I-V curve measured by the EL method
and the I-V curve in the radiative limit is due to non-radiative
recombination as following shows [14],
kT
V  Vrad  VEL   lnemLED
q (8)
where the voltage loss ΔV between the voltage in the radiative
limit (Vrad) and the voltage estimated from the EL measurement
(VEL) is related to the EL quantum efficiency (ηemLED).

Fig. 5. EL quantum efficiency of the GaAs solar cell measured under LED
operation as a function of the injection current density, both the results
measured by Si-photodiode and CCD camera are presented. The inset is the
absolute EL spectra divided by injection electron number for various injection
current densities of the solar cell.

The carrier-balance equation describing the optoelectronic

processes in the solar cell irradiated with sunlight could be
expressed as [25]

sun  J / q  em  nr  (1 / em LED )em (6)

sun   Qe ( E )S AM 1.5 ( E)dE
where is the absorption rate of
sunlight, SAM1.5 is the standard AM1.5G solar spectrum; J is the
current density (+ for forward injection current and – for Fig. 6. I-V curves evaluated from EL measurement (the red solid line) and

photo-generated current); em is the external radiative

from I-V measurement with a solar simulator (the blue open circles). I-V curve
measured by EL method after considering a lump resistance RL (the red dash
emission rate as above mentioned and was experimentally dot line).

obtained using absolute EL measurments; nr is the It is noted that the bending of the I-V curve near the
nonradiative recombination rate and ηemLED is the EL quantum maximum power point (Jm, Vm) is softer than that in the
efficiency as above mentioned. According to the reciprocity radiative limit, which lowers the filling factor, the soft bending
relation between EL and EQE of a solar cell [15], the absolute in the EL I-V curve is considered to originate from the soft
EL emission rate em is given by Eq. (2). The absolute EL bending of the EL quantum efficiency curve near Jm as the inset
emission rate em as a function of the forward injection current
in Fig. 6 shows. In other words, the soft bending is caused by the
lowered luminescence quantum efficiency or radiative
density is experimentally measured. Therefore, the I-V curve of
recombination efficiency at low carrier density and not by series
the solar cell could be predicted according to the equation:
resistance. Table 1 shows the basic parameters of the GaAs
kT em ( I )
V (I )  ln  solar cell evaluated from the I-V curves obtained from I-V
q  Qe  EL  Rb ( E )dE measurement with solar simulator, by EL method and from
Eg
(7)
radiative limit estimation, respectively.
Based on the I-V curve extrapolated from the EL
measurement, the essential photovoltaic parameters such as Voc,
FF, η and the maximum output voltage and current (Vm, Jm)
could be obtained.
Fig. 6 shows the I-V curves evaluated from the absolute EL
measurement (the red solid line) and from I-V measurement

1943-0655 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JPHOT.2017.2731800, IEEE
Photonics Journal
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 5

Table 1. Parameters of the GaAs solar cell evaluated from the I-V curves obtained from I-V
measurement with solar simulator, from EL measurement and from radiative limit estimation.
Voc Jsc Vm Jm FF η
(V) (mA/cm2) (V) (mA/cm2)
Solar simulator 0.996 25.3 0.845 22.96 0.767 19.4%
EL method 1.01 25.3 0.849 24.05 0.798 20.4%
Radiative limit 1.13 25.3 1.04 24.71 0.899 25.7%

Actually, EL method has a merit over the traditional I-V

measurement that it avoids the influence of the external parasitic
resistance effect such as the series resistance due to the probe
contact, which may affect the evaluation of the real parameters
of the solar cell. For traditional I-V measurement, contact
probes of high quality should be required. For common I-V
measurement system especially in small laboratories without
accurate calibration conditions, enormous difference due to
different probe contact condition in different measurement
system may occur. In Fig. 7, the IV curves from test 1 to test 3 is
obtained from different IV measurement system in different
small labs (with different probe contact condition) for the same
solar cell sample, it is obviously to see the great inaccuracy due
to the series resistance effect from external contact. To study the
Fig. 7. I-V curves of the same GaAs solar cell measured under different
effect of series resistance on the I-V measurement by the EL condition and method, including with solar simulator system under different
method, a resistor of 10 Ω was artificial series-connected to the probe contact condition from test 1 to test 3 and with EL method without and
EL measurement system, that is, in addition to the contact probe. with an artificial connected series resistor.
It can be seen that the series resistance has little influence on the
I-V evaluation from the EL method, which also means that the The present measurements of EL efficiency and solar cell
series resistance caused by the probe contact could be ignored. EQE also provided all the terms of emission loss, nonradiative
The I-V measurement from EL method is especially reliable for loss processes in addition to the I-V curves. We evaluated all the
parameters evaluation of solar cells with small internal series loss/output rates in the solar cell under AM1.5G 1-sun
resistance since the series resistance caused by external probe irradiation under the working condition of maximum output
contact could be neglected. power [25]. Table 2 shows the calculated parameters of the
GaAs solar cell working at the maximum-output-power
condition, including the energy gain and loss in the cell, and all
the values are given in ratio. 19.4% of the solar energy is
converted into electric energy, while the remainder goes into
0.17% radiative emission (EM) loss, 6.2% nonradiative
recombination (NR) loss, 14.8% thermalization (TH) loss,
46.2% transmission (TR) loss and 13.2% junction (JN) loss.
The contribution of radiative emission loss is negligibly small in
the maximum-out-put-power condition. The calculated ratio of
the output power 19.4% is identical with the measured cell
efficiency of 19.4% as previous section shows, indicating the
accuracy of the calculation.

Table 2. Parameters of the GaAs solar cell working at the maximum-output-power condition.
All values are given in ratio.
Input Loss
AM1.5G EM NR TR TH JN Power
output
GaAs cell 1.00 1.70E-3 0.062 0.462 0.148 0.132 0.194

absolute EL measurements. Then, the I-V relation from the

IV. CONCLUSIONS combination of the absolute EL and EQE measurements was
In summary, an absolutely calibrated electroluminescence extrapolated and was found to have a merit over the
(EL) imaging method was utilized to investigate the properties conventional I-V relation measured with a solar simulator that
of a GaAs thin-film solar cell. The distribution of the internal the series resistance effect caused by external probe contact
voltage of the solar cell and the sheet resistance of the could be eliminated. Finally, the optic-electro conversion losses
window-layer of the solar cell were determined from the of the solar cell including radiative loss, nonradiative loss,

1943-0655 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JPHOT.2017.2731800, IEEE
Photonics Journal
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 6

thermalization loss, transmission loss and junction loss were 2009.

[22] H. Kampwerth, T. Trupke, J. W. Weber and Y. Augarten, “Advanced
quantitatively determined from the EL and EQE measurements. luminescence based effective series resistance imaging of silicon solar
cells,” Appl. Phys. Lett., vol. 93, 202102, 2008.
[23] M. Glatthaar, J. Giesecke, M. Kasemann, J. Haunschild, M. The, W. Warta
REFERENCES and S. Rein, “Spatially resolved determination of the dark saturation
current of silicon solar cells from electroluminescence images,” J. Appl.
[1] T. Fuyuki, H. Kondo, T. Yamazaki, Y. Takahaschi, and Y. Uraoka, Phys., vol. 105, 113110, 2009.
“Photographic surveying of minority carrier diffusion length in [24] M. Yoshita, L. Zhu, C. Kim, T. Mochizuki, T. Nakamura, M. Imaizumi, S.
polycrystalline silicon solar cells by electroluminescence,” Appl. Phys. Q. Chen, H. Kubota, Y. Kanemitsu, H. Akiyama, “Calibration standards
Lett., vol. 86, 262108K, 2005. and measurement accuracy of absolute electroluminescence and internal
[2] F. Fruehauf, and M.Turek, “Quantification of Electroluminescence properties in multi-junction and arrayed solar cells”, in Proc. SPIE 9743,
Measurements on Modules,” Energy Procedia, vol. 77, pp. 63-68, 2015. Physics, Simulation, and Photonic Engineering of Photovoltaic Devices
[3] G. W. Shu, J. Y. Lin, H. T. Jian, J. L. Shen, S. C. Wang, C. L. Chou, W. C. V, 97430D, 2016.
Chou, C. H. Wu, C. H. Chiu, and H. C. Kuo, “Optical coupling from [25] S. Q. Chen, L. Zhu, M. Yoshita, T. Mochizuki, C. Kim, H. Akiyama, M.
InGaAs subcell to InGaP subcell in InGaP/InGaAs/Ge multi-junction solar Imaizumi and Y. Kanemitsu, “Thorough subcells diagnosis in a
cells,” Opt. express, vol. 21, pp. A123-A130, 2013. multi-junction solar cell via absolute electroluminescence-efficiency
[4] M. Seeland, R. Rösch, H. Hoppe, “Quantitative analysis of measurements,” Sci. Rep., vol. 5, 7836, 2015.
electroluminescence images from polymer solar cells,” J. Appl. Phys., vol. [26] T. Mochizuki, C. Kim, M. Yoshita, J. Mitchell, L. Zhu, S. Q. Chen, H.
111, 024505, 2012. Takato, Y. Kanemitsu, and H. Akiyama, “Solar-cell radiance standard for
[5] Z. Hameiri, A. M. Soufiani, M. K. Juhl, L. Jiang, F. Huang, Y. B. Cheng, absolute electroluminescence measurements and open-circuit voltage
and T. Trupke, “Photoluminescence and electroluminescence imaging of mapping of silicon solar modules,” J. Appl. Phys., vol.119, 034501,
perovskite solar cells.,” Prog. Photovoltaics, vol. 23, pp. 1697-1705, 2015. 2016.
[6] M. Bokalič, J. Raguse, J. R. Sites, and M. Topič, “Analysis of [27] T. Fuyuki, and A. Kitiyanan, “Photographic diagnosis of crystalline
electroluminescence images in small-area circular CdTe solar cells,” J. silicon solar cells utilizing electroluminescence,” App. Phys. A, vol. 96,
Appl. Phys., vol. 114, 123102, 2013. pp. 189-196, 2009.
[7] B. Li, A. Stokes, and J. Doble, “Evaluation of two-dimensional electrical [28] G.T. Koishiyev, and J. R. Sites, “Impact of sheet resistance on 2-D
properties of photovoltaic modules using bias-dependent modeling of thin-film solar cells,” Sol. Energy Mater. Sol. Cells, vol. 93,
electroluminescence,” Prog. Photovoltaics, vol. 20, pp. 936–944, 2012. pp. 350–354, 2009.
[8] P. Chaturvedi, H. Bram and T. M. Walsh, “Broken metal fingers in silicon [29] A. Helbig, T. Kirchartz, R. Schaeffler, J. H. Werner and U. Rau,
wafer solar cells and PV modules,” Sol. Energy Mater. Sol. Cells, vol. 108, “Quantitative electroluminescence analysis of resistive losses in Cu(In,
pp. 78-81, 2013. Ga)Se2 thin-film modules,” Sol. Energy Mater. Sol. Cells, vol. 94, pp.
[9] A. Kitiyanan, A. Ogane, A. Tani, T. Hatayama, H. Yano, Y. Uraoka and T. 979–984, 2010.
Fuyuki, “Comprehensive study of electroluminescence in multicrystalline [30] C. Jordan, J. F. Donegan, J. Hegarty, B. J. Roycroft, S. Taniguchi, T. Hino,
silicon solar cells,” J. Appl. Phys., vol. 106, 043717, 2009. E. Kato, N. Noguchi, and A. Ishibashi, “Carrier-density dependence of
[10] M. Schneemann, T. Kirchartz, R. Carius and U. Rau, “Measurement and the photoluminescence lifetimes in ZnCdSe/ZnSSe quantum wells at
modeling of reverse biased electroluminescence in multi-crystalline room temperature,” Appl. Phys. Lett., vol.74, pp. 3359-3361, 1999.
silicon solar cells,” J. Appl. Phys., vol. 114, 134509, 2013.
[11] H. Nesswetter, W. Dyck, P. Lugli, A. W. Bett and C. G. Zimmermann,
“Luminescence based series resistance mapping of III-V multijunction
solar cells,” J. Appl. Phys., vol. 114, 194510, 2013.
[12] M. Seeland, C. Kästner, H. Hoppe, “Quantitative evaluation of
inhomogeneous device operation in thin film solar cells by luminescence
imaging,” Appl. Phys. Lett., vol. 107, 073302, 2015.
[13] K. Ramspeck, K. Bothe, D. Hinken, B. Fischer, J. Schmidt and R. Brendel,
“Recombination current and series resistance imaging of solar cells by
combined luminescence and lock-in thermography,” Appl. Phys. Lett.,
vol. 90, 153502, 2007.
[14] P. Wurfel, T. Trupke, T. Puzzer, E. Schaffer, W. Warta, and S. Glunz,
“Diffusion lengths of silicon solar cells from luminescence images,” J.
Appl. Phys. Vol. 101, 123110, 2007.
15. U. Rau, “Reciprocity relation between photovoltaic quantum efficiency and
electroluminescence emission of solar cells,” Phys. Rev. B, vol. 76,
085303, 2007.
[16] T. Kirchartz, A. Helbig, W. Reetz, M. Reuter, J. H. Werner, and U. Rau,
“Reciprocity between electroluminescence and quantum efficiency used
for the characterization of silicon solar cells,” Prog. Photovoltaics, vol. 17,
pp. 394-402, 2009.
[17] D. Hinken, K. Ramspeck, K. Bothe, B. Fischer, and R. Brendel, “Series
resistance imaging of solar cells by voltage dependent
electroluminescence,” Appl. Phys. Lett., vol. 91, 182104, 2007.
[18] O. Breitenstein, J. Bauer, T. Trupke and R. Bardos, “On the detection of
shunts in silicon solar cells by photo- and electroluminescence imaging,”
Prog. Photovoltaics, vol. 16, pp. 325–330, 2008.
[19] T. Potthoff, K. Bothe, U. Eitner, D. Hinken, and M. Kontges, “Detection of
the voltage distribution in photovoltaic modules by electroluminescence
image,” Prog. Photovoltaics, vol.18, pp. 100–106, 2010.
[20] O. Breitenstein, A. Khanna, Y. Augarten, J. Bauer, J. M.Wagner and K.
Iwig, “Quantitative evaluation of electroluminescence images of solar
cells,” physica status solidi (RRL), vol. 4, pp. 7-9, 2010.
[21] J. Haunschild, M. Glatthaar, M. Kasemann, S. Rein and E. R. Weber,
“Fast series resistance imaging for silicon solar cells using
electroluminescence,” physica status solidi (RRL), vol. 3, pp. 227-229,

1943-0655 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.