Japanese Journal of Applied

Physics

RAPID COMMUNICATION You may also like

- Structural stabilities of organic–inorganic
Germanium ion doping of CsPbI3 to obtain perovskite crystals
Hiroki Tanaka, Takeo Oku and Naoki
inorganic perovskite solar cells with low Ueoka

- Review—Recent Advances of
temperature processing Electrochemical Techniques in Food,
Energy, Environment, and Forensic
Applications
To cite this article: Atsushi Kogo et al 2022 Jpn. J. Appl. Phys. 61 020904 Fernanda Da Silva Santos, Leandro Vitor
da Silva, Paulo Victor Soares Campos et
al.

- Improving the phase stability of inorganic

lead halide perovskites through K/Rb
View the article online for updates and enhancements. doping
Si-Qin Wang, Shiyu Shen, Xiong-Xiong
Xue et al.

This content was downloaded from IP address 49.36.180.217 on 22/04/2022 at 14:55

Japanese Journal of Applied Physics 61, 020904 (2022) RAPID COMMUNICATION
https://doi.org/10.35848/1347-4065/ac4927

Germanium ion doping of CsPbI3 to obtain inorganic perovskite solar cells with
low temperature processing
Atsushi Kogo* , Kohei Yamamoto , and Takurou N. Murakami
National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan
*
E-mail: [email protected]
Received December 22, 2021; accepted January 7, 2022; published online January 24, 2022

Although the all-inorganic perovskite CsPbI3 exhibits superior thermal- and photo-stability compared with organic–inorganic perovskites, formation
of the photoactive α-phase requires sintering at approximately 320 °C. Herein, we report the partial substitution of Ge2+ ions for Pb2+ as a means
of tuning the stability of the material and enabling α-phase formation at 90 °C. © 2022 The Japan Society of Applied Physics

Supplementary material for this article is available online

S
olution-processable organic–inorganic perovskites simply by evaporating a precursor solution at low tempera-
such as CH3NH3PbI3 and CH(NH2)2PbI3 have re- ture. As an example, in the case of the organic–inorganic
cently been investigated as novel light-harvesting mate- CH3NH3PbI3 perovskite having a t value of 0.91,17) the α-
rials for solar cells.1–3) The unique properties of these substances, phase can be obtained by drying at approximately 100 °C,
including strong optical absorption in the visible range (400–800 and is stable at room temperature. However, the CsPbI3 α-
nm, with ε values of approximately 105 cm−1)4) and long phase (rA = 167 pm, rB = 119 pm, rX = 220 pm)18) is un-
electron-hole diffusion distances (>175 μm),5) permit photoe- stable at room temperature (t = 0.81). Consequently,
nergy conversion with minimum charge carrier recombination. CsPbBr3 and CsPbI2Br, both of which exhibit relatively
At present, these materials exhibit a power conversion efficiency high α-phase stability, have been employed rather than
(PCE) as high as 25.5%, which is close to that for Si solar CsPbI3 despite their larger bandgap energies (2.3 eV19) and
cells.6,7) In addition, the solution-processability and low proces- 2.05 eV,20) respectively) that result in a narrower absorption
sing temperatures (100 °C–150 °C) of perovskites could enable wavelength range. It would therefore be beneficial to develop
solar cell fabrication on bendable plastic substrates,8–12) thus a means of doping CsPbX3 to obtain a highly stable
allowing cost-effective roll-to-roll mass printing. For these α-phase.15,16,21–26)
reasons, perovskite-based solar cells have attracted significant The present study examined the effects of introducing Ge2+
attention as new and promising candidates for next-generation ions, which have a smaller ionic radius than Pb2+, into CsPbI3.
photovoltaics. The partial replacement of Ge2+ with Pb2+ in the perovskite
However, perovskite solar cells continue to suffer from poor was expected to increase its t value by decreasing rB, so as to
stability. In response to light exposure or high temperatures, allow the α-phase perovskite to be formed at relatively low
the CH3NH3+ and CH(NH2)2+ cations in the perovskite layers temperatures. This work demonstrates the formation of a
readily decompose as the compounds are converted to PbI2.13) CsPb0.8Ge0.2I3 perovskite after drying at 90 °C, with a PCE
To address this issue, the fully inorganic perovskites CsPbBr3 of 3.97%. The α-phase CsPb0.8Ge0.2I3 was found to be stable
and CsPbI3, in which stable Cs+ cations replace CH3NH3+ and did not exhibit a transition to the δ-phase at room
and CH(NH2)2+, have been developed.13,14) The all-inorganic temperature under N2. Compared to previous reports using
perovskite CsPbI3 has been reported to have a PCE of 19%, organic additives to precursor solutions to stabilize α-phase of
which is comparable to the values obtained from organic– CsPbI3,23–25) our process, substituting Ge2+ to Pb2+ ions have
inorganic hybrid perovskites.15) Although CsPbI3 shows an advantage that we could reduce usage of toxic Pb2+ in the
greater stability in response to light and elevated temperatures, perovskite mass production process in the future.
the phase transition of this material is an important issue. CsPb1−xGexI3 perovskite samples were formed on SnO2
Specifically, the α-phase of CsPbI3, which exhibits a high electron transport layers on indium tin oxide (ITO) substrates
degree of light absorption in the visible range (with a bandgap by spin-coating of precursor solutions at 3000 rpm for 30 s
energy of approximately 1.7 eV)14,16) and good photovoltaic followed by drying at 100 °C for 30 min under N2. These
performance, must be generated by sintering at temperatures precursor solutions were prepared by mixing 0.8 M CsPbI3
above 320 °C.16) Furthermore, α-phase CsPbI3 is unstable at and 0.8 M CsGeI3 solutions with anhydrous dimethyl sulf-
room temperature and exhibits a phase transition to a yellow oxide (DMSO) at volume ratios of 1 − x:x:0.2. The 0.8 M
orthorhombic δ-phase that is stable but does not exhibit visible CsPbI3 solution was prepared by dissolving CsI and PbI2 in a
light absorption (with a bandgap energy of approximately mixture of anhydrous N,N-dimethylformamide (DMF) and
2.8 eV)14,16) or photovoltaic performance.14) anhydrous DMSO (volume ratio = 4:1). The 0.8 M CsGeI3
The stability of the α-phase of an ionic ABX3-type solution was prepared by dissolving CsI and GeI2 in
perovskite material depends on its Goldschmidt tolerance anhydrous DMF followed by filtering through a 0.2 micro-
factor. This is defined as t = 2−0.5(rA + rX)/(rB + rX), where meter pore size membrane. Degradation of the materials by
rA, rB, and rX are the ionic radii of the A, B and X ions.14) humid ambient air during characterization was prevented by
When t is in the range of 0.9–1.0, the α-phase perovskite spin-coating solutions of 2,2′,7,7′-tetrakis-(N,N-di-p-methox-
structure is stable at room temperature and can be formed yphenylamine)−9,9′-spirobifluorene (spiro-OMeTAD) in

020904-1 © 2022 The Japan Society of Applied Physics

Jpn. J. Appl. Phys. 61, 020904 (2022) A. Kogo et al.

(a)

(b)
(a)

(c)

(b)

Fig. 2. (Color online) (a) Absorption spectra of

ITO/SnO2/CsPb0.8Ge0.2I3/spiro-OMeTAD/PVP specimens annealed at dif-
ferent temperature for 30 min. (b) Absorption spectra of an
ITO/SnO2/CsPb0.8Ge0.2I3/spiro-OMeTAD/PVP specimen prepared at 100 °C
(d) before and after storage under N2 for 12 d.

Images of the CsPb1−xGexI3 perovskites are provided in

Fig. 1(a). CsGeI3 proportions of 20%–60% in CsPbI3
resulted in a brownish color, suggesting the formation of an
α-phase perovskite. In contrast, CsPb1−xGexI3 specimens for
which x was less than 0.2 were yellow, and so likely
comprised an orthorhombic δ-phase. In addition,
(e) CsPb1−xGexI3 specimens for which x was greater than 0.6
were transparent. UV–visible spectra of the
ITO/SnO2/CsPb1−xGexI3/spiro-OMeTAD/PVP devices are
presented in Fig. 1(b). The yellow CsPbI3 (x = 0) had a
bandgap energy of 447 nm (=2.77 eV), equal to that of
δ-CsPbI3.14,16) The CsPb1−xGexI3 perovskites having x va-
lues of 0.2−0.6 exhibited light absorption over a wide range
of visible wavelengths (up to 700 nm), suggesting the
Fig. 1. (Color online) (a) Photographic images, (b) UV–visible formation of α-CsPbI3. The bandgap energy for each such
absorption spectra and (c), (d) XRD patterns for ITO/SnO2/CsPb1-xGexI3/ material was 1.74 eV, corresponding to that for
spiro-OMeTAD/PVP specimens. (e) XRD pattern of ITO/SnO2/CsPb0.8 α-CsPbI3.14,16)
Ge0.2I3/spiro-OMeTAD/PVP in the range of 2θ = 28.6 –29.2°. Peak position of The crystal structure of the CsPb1−xGexI3 perovskites was
α-CsPbI3 calculated by CrystalDiffract software (CrystalMaker Software Ltd) and
CsPb0.8Ge0.2I3 are indicated by black and yellow dashed line.
characterized by performing X-ray diffraction (XRD) analyses
[Figs. 1(c) and 1(d)]. The CsPbI3 and CsPb0.9Ge0.1I3 layers
both generated diffraction peaks at 13.08° and 22.68° that were
chlorobenzene and polyvinylpyrrolidone (PVP) in 2-propanol attributed to diffraction from the (012) and (112) crystal planes
over top of the ITO/SnO2/CsPb1−xGexI3 specimens, followed of the orthorhombic δ-phase.21,26) However, CsPb1−xGexI3
by drying under N2. Further experimental details are pro- samples with x values ranging from 0.2 to 0.6 produced peaks
vided in the Supporting Information to this paper. at 14.34°, 21.32° and 28.90° that were associated with the
020904-2 © 2022 The Japan Society of Applied Physics
Jpn. J. Appl. Phys. 61, 020904 (2022) A. Kogo et al.

(a)

(b)

(c)

Fig. 3. (Color online) (a) A cross-sectional SEM image of an ITO/SnO2/CsPb0.8Ge0.2I3/spiro-OMeTAD/Au perovskite solar cell. (b) J–V curves obtained
from the best performing CsPb0.8Ge0.2I3 perovskite solar cell. Forward (−0.2 V → 1.3 V) and backward (1.3 V → −0.2 V) voltage scans are indicated by solid
and dashed lines, respectively. (c) EQE spectrum of the short-circuit photocurrent generated by the CsPb0.8Ge0.2I3 solar cell.

Table I. Performance parameters for the best performing ITO/SnO2/CsPb0.8Ge0.2I3/spiro-OMeTAD/Au solar cell.

Scan direction JSC/mA cm−2 VOC/V FF PCE/%

Forward scan (−0.2 V → 1.3 V) 9.80 0.678 0.566 3.76

Backward scan (1.3 V → −0.2 V) 9.58 0.701 0.590 3.97

(100), (110) and (200) crystal planes of the α-phase planes of CsI,27) demonstrating that perovskite formation did
perovskite.21) The CsPb0.8Ge0.2I3 perovskite provided the not take place. These results clearly indicate that replacing 20%
sharpest α-phase peaks and thus had the highest crystallinity of the Pb2+ ions with Ge2+ promoted the formation of α-phase
among the CsPb1−xGexI3 materials. When x was larger than CsPbI3 having a high degree of crystallinity at a low
0.8, a peak appeared at 26.59° due to diffraction from the (110) temperature of 100 °C. To examine incorporation of Ge2+
020904-3 © 2022 The Japan Society of Applied Physics
Jpn. J. Appl. Phys. 61, 020904 (2022) A. Kogo et al.

ions in CsPbI3 crystal structure, we compared XRD peak and low-resistive CsPb0.8Ge0.2I3 layer can be increased to
position of CsPb0.8Ge0.2I3 and CsPbI3 [Fig. 1(e)]. The (200) enhance optical absorption and photocurrent of solar cells to
diffraction peak position of CsPb0.8Ge0.2I3 at 2θ = 28.90° was improve PCE.
larger than that of α-CsPbI3 at 28.88° indicating that Compared to CsPb0.8Ge0.2I3, CsPb0.6Ge0.4I3 as well ex-
CsPb0.8Ge0.2I3 has a smaller lattice parameter. This is due to hibit optical absorption in the visible range [Fig. 1(b)].
the replacement of Pb2+ ions to Ge2+, which has a smaller ion However, the solar cells with CsPb0.6Ge0.4I3 gave poor
radius. photovoltaic performance (PCE ∼ 0.08%) due to low
The lowest temperature at which α-phase formation could be crystallinity of α-phase perovskite [Fig. 1(c)].
achieved was examined by spin-coating ITO/SnO2 substrates In conclusion, we fabricated CsPb1−xGexI3 specimens
with the precursor solutions at different temperatures and then containing Ge2+ ions having a smaller radius than Pb2+
acquiring UV–visible spectra [Fig. 2(a)]. The specimen an- ions, as a means of controlling the α-phase stability of the
nealed at 60 °C did not exhibit absorption in the visible range, perovskite. The α-phase of CsPb0.8Ge0.2I3 could be formed
while the sample processed at 70 °C showed increased by annealing at 90 °C, which is much lower than the
absorption below 740 nm. Annealing at 70 °C–90 °C further temperature required to produce the α-phase CsPbI3 perovs-
increased the absorption of visible light. These results indicate kite (320 °C).16) The CsPb0.8Ge0.2I3 perovskite exhibited a
that the extent of α-phase CsPb0.8Ge0.2I3 formation was PCE of 3.97% and good EQE values of 40%–70%,
increased by the rapid drying associated with higher annealing suggesting photovoltaic properties. The α-phase of this
temperatures. The absorption spectra of the specimens annealed compound also exhibited good stability at room temperature.
at 90 °C and 100 °C were similar, suggesting saturation of the
Acknowledgments This research was financially supported by a Grant-in-
α-CsPb0.8Ge0.2I3 formation rate. The stability of the α-phase at Aid for Early-Career Scientists (no. 21K14733) from the Japan Society for the
room temperature was assessed by acquiring UV–visible spectra Promotion of Science (JSPS).
from the CsPb0.8Ge0.2I3 perovskite (formed by 100 °C an- ORCID iDs Atsushi Kogo https://orcid.org/0000-0002-9837-
nealing) before and after storage in N2 [Fig. 2(b)]. These trials 5585 Kohei Yamamoto https://orcid.org/0000-0002-0503-4432 Takurou
N. Murakami https://orcid.org/0000-0002-2066-3982
demonstrated that the spectrum of this material was not changed
after storage for 12 d, confirming that the CsPb0.8Ge0.2I3 α-
phase was stable at room temperature.
The photovoltaic performance of the CsPb0.8Ge0.2I3 per- 1) A. Kojima, K. Teshima, Y. Shirai, and T. Miyasaka, J. Am. Chem. Soc. 131,
6050 (2009).
ovskite was subsequently evaluated by fabricating solar cells. 2) M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami, and H. J. Snaith,
In the preparation process, spiro-OMeTAD hole-transport Science 338, 643 (2012).
layers were formed on ITO/SnO2/CsPb0.8Ge0.2I3 substrates 3) H.-S. Kim et al., Sci. Rep. 2012, 591 (2012).
by spin-coating with chlorobenzene solutions containing 4) S. Sun, T. Salim, N. Mathews, M. Duchamp, C. Boothroyd, G. Xing, T.
C. Sum, and Y. M. Lam, Energy Environ. Sci. 7, 399 (2014).
spiro-OMeTAD, LiN(CF3SO2)2, 4-tert-butylpyridine and 5) Q. Dong, Y. Fang, Y. Shao, P. Mulligan, J. Qiu, L. Cao, and J. Huang,
tris(2-(1H-pyrazol-1-yl)-4-tertbutylpyridine)-cobalt(III) tris Science 347, 967 (2015).
(bis(trifluoromethylsulfonyl)imide, followed by drying over- 6) Efficiency chart of NREL, (https://www.nrel.gov/pv/assets/pdfs/best-re-
search-cell-efficiencies-rev211214.pdf).
night at room temperature in dry air (dew point of approxi-
7) M. A. Green, E. D. Dunlop, J. Hohl‐Ebinger, M. Yoshita, N. Kopidakis, and
mately −30 °C). Au contacts (3 mm × 3 mm) were then X. Hao, Prog. Photovolt. Res. Appl. 30, 3 (2022).
formed by vapor deposition. 8) M. Kaltenbrunner et al., Nat. Mater. 14, 1032 (2015).
A cross-sectional scanning electron microscopy (SEM) 9) A. Kogo, M. Ikegami, and T. Miyasaka, Chem. Commun. 52, 8119 (2016).
10) A. Kogo, Y. Sanehira, Y. Numata, M. Ikegami, and T. Miyasaka, ACS
image of the ITO/SnO2/CsPb0.8Ge0.2I3/spiro-OMeTAD/Au
Appl. Mater. Interfaces 10, 2224 (2018).
solar cell is shown in Fig. 3(a). Although some voids can be 11) A. Kogo, S. Iwasaki, M. Ikegami, and T. Miyasaka, Chem. Lett. 46, 530
observed in the perovskite layer, this coating appears to have (2017).
fully covered the SnO2 electron transport layer. The thick- 12) P. Docampo, J. M. Ball, M. Darwich, G. E. Eperon, and H. J. Snaith, Nat.
Commun. 4, 2761 (2013).
nesses of the CsPb0.8Ge0.2I3 and spiro-OMeTAD layers were 13) Q. Tai, K. Tang, and F. Yan, Energy Environ. Sci. 12, 2375 (2019).
estimated to be approximately 300 and 250 nm, respectively, 14) J. Tian, Q. Xue, Q. Yao, N. Li, C. J. Brabec, and H. Yip, Adv. Energy Mater.
based on SEM observations. Photocurrent density–voltage 10, 2000183 (2020).
(J–V ) curves were acquired from these solar cells (active 15) X. Chang et al., Adv. Mater. 32, 2001243 (2020).
16) B. Zhao et al., J. Am. Chem. Soc. 140, 11716 (2018).
area = 3 mm × 3 mm) under 1 sun illumination [Fig. 3(b)] 17) M. Saliba et al., Science 354, 206 (2016).
and photovoltaic parameters were estimated (Table I). A PCE 18) R. D. Shannon, Acta Crystallogr. A 32, 751 (1976).
of 3.97% was obtained from the best-performing cell. The 19) M. B. Faheem, B. Khan, C. Feng, M. U. Farooq, F. Raziq, Y. Xiao, and
external quantum efficiency (EQE) action spectrum asso- Y. Li, ACS Energy Lett. 5, 290 (2020).
20) J. Duan, H. Xu, W. E. I. Sha, Y. Zhao, Y. Wang, X. Yang, and Q. Tang, J.
ciated with short-circuit photocurrent generation is provided Mater. Chem. A 7, 21036 (2019).
in Fig. 3(c). EQE values of 40%−70% were obtained in the 21) A. K. Jena, A. Kulkarni, Y. Sanehira, M. Ikegami, and T. Miyasaka, Chem.
visible range (400–700 nm), in agreement with the wave- Mater. 30, 6668 (2018).
length range over which the CsPb0.8Ge0.2I3 perovskite was 22) F. Yang, D. Hirotani, G. Kapil, M. A. Kamarudin, C. H. Ng, Y. Zhang,
Q. Shen, and S. Hayase, Angew. Chem. Int. Ed. 57, 12745 (2018).
found to absorb [Fig. 1(a)]. These results indicate that the 23) T. Ye, B. Zhou, F. Zhan, F. Yuan, S. Ramakrishna, D. Golberg, and
CsPb0.8Ge0.2I3 provided good photovoltaic performance. It is X. Wang, Sol. RRL 4, 2000014 (2020).
also likely that improving the CsPb0.8Ge0.2I3 film mor- 24) J. Zhang, C. Wang, H. Fu, L. Gong, H. He, Z. Fang, C. Zhou, J. Chen,
Z. Chao, and J. Fan, J. Alloys Compd. 862, 158454 (2021).
phology (eliminating voids) and crystallinity by process
25) Y. Zhang, Y. Yang, X. Zhang, T. Wang, L. Nian, Q. Rong, G. Zhou, and
engineering such as using additives,23–26) anti-solvents,28,29) N. Li, Org. Electron. 287, 105940 (2020).
and surface treatments30–33) would decrease the resistance of 26) P. Luo, W. Xia, S. Zhou, L. Sun, J. Cheng, C. Xu, and Y. Lu, J. Phys. Chem.
solar cells. Furthermore, the thickness of the high crystalline Lett. 7, 3603 (2016).

020904-4 © 2022 The Japan Society of Applied Physics

Jpn. J. Appl. Phys. 61, 020904 (2022) A. Kogo et al.

27) W. Song, X. Wu, Q. Di, T. Xue, J. Zhu, and Z. Quan, RSC Adv. 8, 18519 30) K. T. Cho, S. Paek, G. Grancini, C. Roldán-Carmona, P. Gao, Y. Lee, and
(2018). M. K. Nazeeruddin, Energy Environ. Sci. 10, 621
28) N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. Ryu, and S. Il Seok, Nat. (2017).
Mater. 13, 897 (2014). 31) A. Kogo, T. Miyadera, and M. Chikamatsu, ACS Appl. Mater. Interfaces 11,
29) M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A. Gray- 38683 (2019).
Weale, U. Bach, Y.-B. Cheng, and L. Spiccia, Angew. Chem. Int. Ed. 53, 32) A. Kogo and M. Chikamatsu, Chem. Commun. 56, 1235 (2020).
9898 (2014). 33) A. Kogo and M. Chikamatsu, Nanoscale 12, 21605 (2020).

020904-5 © 2022 The Japan Society of Applied Physics