- Email: [email protected]
Fluorescence spectroscopy-based study of balanced transport of charge carriers in hot-air-annealed perovskites
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 207 (2019) 68–72
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Fluorescence spectroscopy-based study of balanced transport of charge carriers in hot-air-annealed perovskites In-Wook Hwang a,⁎, Yanliang Liu b,c, Sung Heum Park b,c a b c
Advanced Photonics Research Institute, Gwangju Institute of Science and Technology, Gwangju 500-712, South Korea Department of Physics, Pukyong National University, Busan 48513, South Korea Hybrid Interface Materials Global Frontier Research Group, Pusan National University, Busan 46241, South Korea
a r t i c l e
i n f o
Article history: Received 19 June 2018 Received in revised form 27 August 2018 Accepted 2 September 2018 Available online 04 September 2018 Keywords: Fluorescence Perovskite Solar cell Charge carrier transport Time-resolved
a b s t r a c t Using fluorescence spectroscopy, we investigate the transport of charge carriers in (p-i-n) planar perovskite (CH3NH3PbI3−xClx) solar cells with the structure ITO/PEDOT:PSS/perovskites/PCBM/Ca/Al, in which the perovskite morphology is optimized by moisture pretreatment (MPT) and the hot-air-annealing process (HAAP). After annealing with hot air, the electron transport time of the perovskites is shortened by a factor of 3.6 (from 28.8 to 7.9 ns), eventually leading to balanced transport of electrons and holes (characterized by the fluorescence-decay time constants of 7.9 and 7.6 ns, respectively). These results are in good agreement with the observed increase in the photovoltaic conversion efficiency. © 2018 Published by Elsevier B.V.
1. Introduction Organic-inorganic halide perovskite solar cells (PESCs) have attracted considerable attention, because of their high device performance, low cost, and ease of fabrication. The high performance results from the excellent characteristics of perovskites, such as high absorbance [1,2], ambipolar carrier transport [3,4], small exciton binding energies (~20 meV) [5–7], long electron–hole diffusion lengths (100–1000 nm), and long exciton lifetimes (~100 ns) [8–10]. To fabricate high-performance PESCs, the morphology of perovskites needs to be controlled to realize high crystallinity, high coverage extent, and large grain sizes [11–14]. The presence of nonuniform crystalline networks in the perovskite layer results in a short electron–hole transport distance and high electron–hole charge recombination loss [13], while a poor surface coverage causes the incident photons to pass through the uncovered areas, leading to a low photocurrent [14]. Recently, we have improved the crystallinity and coverage of perovskites by annealing the pristine perovskite (CH3NH3PbI3−xClx) layer with exposure to moisture and hot air, and have realized enhancement in the photovoltaic conversion efficiency from 9.05% to 15.55% [15]. In this study, with the goal of understanding what currently limits the photovoltaic conversion efficiency in this system, we comparatively investigate the carrier transport dynamics of various PESCs. From analyses of the fluorescence-decay profiles of the conventional, moisture ⁎ Corresponding author. E-mail address: [email protected] (I.-W. Hwang).
https://doi.org/10.1016/j.saa.2018.09.005 1386-1425/© 2018 Published by Elsevier B.V.
pretreatment (MPT)- and hot-air-annealing process (HAAP)-perovskite layers coated on or under the carrier transporting layers of PEDOT:PSS and PCBM or sandwiched at their interface, we characterize the time constants for the transport of electrons and holes onto respective transporting layers. From analyses, we find that the efficiency of PESCs is significantly influenced by the carrier transport rates. Notably, the most-efficient cell fabricated by using both processes of MPT and HAAP exhibits short and balanced hole and electron transport times, consistent with enhancements in the short circuit current (Jsc), open circuit voltage (Voc), and fill factor of PESCs. 2. Experimental All reagents purchased from Sigma–Aldrich were used without purification. Methylammonium iodide (MAI) was synthesized according to the literature [16]. Methylamine (24 mL and 33 wt% in absolute ethanol) and HI (10 mL and 57 wt% in water) were reacted in a round bottom flask under N2 atmosphere at 0 °C for 2 h with stirring. Then, the resulting solution was dried by rotary evaporation at 50 °C to form a white powder of MAI. For the purification of the pristine MAI, the dried MAI powder was dissolved in ethanol followed by sedimentation in diethyl ether by stirring the solution. This process was repeated three times and an MAI powder with high purity was recovered and dried in a vacuum oven for 24 h at 60 °C. Thin-film perovskite samples were fabricated as follows. A quartz substrate was cleaned with detergent, ultrasonicated in acetone and isopropyl alcohol, and subsequently dried in an oven at 100 °C.
I.-W. Hwang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 207 (2019) 68–72
PEDOT:PSS (Baytron PH) was spin-casted on quartz at 4500 rpm from an aqueous solution to form a 40 nm thick film. The PEDOT:PSS-coated substrate was dried for 10 min at 140 °C in air and then transferred into a glove box to spin-cast the perovskite layers. A solution containing a mixture of MAI:PbCl2 with a molar ratio of 3:1 in DMF solvent (40 wt %) was spin-casted at 5000 rpm for 45 s on top of pure or PEDOT:PSScoated quart substrate. The precursor perovskite film was dried in the glove box for 30 min at room temperature. The thickness of the perovskite films was ~350 nm. In conventional process, the precursor perovskite film was directly heated on a hot plate in the glove box at 120 °C for 30 min. Meanwhile, in the moisture pretreatment process, the film was placed in a humid atmosphere (relative humidity 60%) for 3 min before the annealing process and subsequently annealed on a hot plate. For hot air annealing, the film was annealed in a hot-air environment in a N2 filled dry oven at 120 °C for 30 min after the moisture pretreatment. The PC61BM (20 mg/mL in chlorobenzene) was deposited by spin coating at 1000 rpm for 40 s to form a complete film. Further details of fabrication of the perovskite films and PESCs have been reported in a previous paper [15]. The steady-state fluorescence spectra and time-resolved fluorescence-decay profiles are recorded using a Hitachi F-4500 fluorescence spectrophotometer and a time-correlated single photon counting (TCSPC) system, respectively [17,18]. In the fluorescence measurements, we deliberately photoexcite the quartz sides of films of quartz/ perovskites, quartz/PEDOT:PSS/perovskites, quartz/perovskites/PCBM, and quartz/PEDOT:PSS/perovskites/PCBM for monitoring carrier transport dynamics under the same condition with sun light irradiation onto PESCs: Intensities of the steady-state fluorescence spectra and profiles of the time-resolved fluorescence-decays are also found to be influenced by the direction of photoexcitation of the films. The decay-time constants for the fluorescence-decay profiles are obtained by first deconvoluting the measured signal from the pump time profile (characterized by a full width at half maximum of ~100 ps) and then fitting to a sum of exponential terms [19]. The chi-square (χ2) values of the fitting are 1.0–1.5.
69
3. Results and Discussion The scanning electron microscope images and time-resolved fluorescence-decay profiles of the conventional, MPT-, and HAAPperovskite films and corresponding solar cell device characteristics (current–voltage curves) are shown in Fig. 1. After annealing the perovskite film with moisture and hot air, the number of pinholes systematically decreases, while the fluorescence-decay time and device power conversion efficiency (PCE) increase, as characterized by the decaytime constants of 68, 207, and 397 ns and the PCEs of 9.05%, 11.87%, and 15.55%, respectively (Fig. 1). We note that in Fig. 1b, the timeresolved fluorescence-decay profiles of the perovskite samples are consistently fitted by using single exponential-decay functions. These single exponential decays are comparable to those of other small grain-sized perovskites with diameters of a few micrometres [20]. Lie et al., previously demonstrated that these single exponential decays result from trap-assisted carrier recombination rather than free-carrier-assisted carrier recombination characterized by multi-exponential fluorescence-decay profiles of larger grain-sized perovskites [20]. Following these, the elongated fluorescence-decay times after annealing the perovskites are indicative of reduced number of carrier trapping sites, in good agreement with the reduced number of pin holes shown in Fig. 1a. Note that in Fig. 2, the reduced number of carrier trapping sites after annealing the perovskites results in the intensified fluorescence spectra of MPT- and HAAP-perovskites compared to the one of conventional perovsite. Charge carrier transports of PESCs are investigated by comparing the steady-state fluorescence spectra and time-resolved fluorescencedecay profiles obtained from films of pure perovskites, PEDOT:PSS/perovskites, perovskites/PCBM, and PEDOT:PSS/perovskites/PCBM, fabricated with the conventional, MPT- and HAAP-perovskite layers (Figs. 2 and 3). When coated on or under the carrier transporting layers of PEDOT: PSS and PCBM, the conventional, MPT- and HAAP-perovskites exhibit differing degrees of fluorescence quenching (Fig. 2). The fluorescence
a
c 1
0.1
0.01
5
2
Conventinal MPT HAAP
Current density (mA/cm )
PL intensity (normalized)
b
0
-10 -15 -20 -25
0
200
400
Time (ns)
600
800
Conventional MPT HAAP
-5
0.0
0.2
0.4
0.6
0.8
1.0
Voltage (V)
Fig. 1. (a) Scanning electron microscope images, (b) time-resolved fluorescence-decay profiles of the conventional, MPT-, and HAAP-perovskite films, and (c) current density-voltage (J-V) characteristics of corresponding PESCs, adapted from [15]. The excitation wavelength of 470 nm is used for measurement of time-resolved fluorescence-decay profiles. The optimized HAAP cell shows the best device characteristic parameters (PCE, FF, Jsc, and Voc of 15.55%, 80.30%, 20.39 mA/cm2, and 0.95 V, respectively), compared to the conventional cell (9.05%, 64.43%, 16.34 mA/cm2, and 0.67 V, respectively) and the MPT cell (11.87%, 74.07%, 17.80 mA/cm2, and 0.90 V, respectively).
70
I.-W. Hwang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 207 (2019) 68–72
PL intensity (a.u.)
Conventional
Perovskite Perovskite/PCBM PEDOT:PSS/perovskite PEDOT:PSS/perovskite/PCBM
2000
0
PL intensity (a.u.)
MPT 5000
0
PL intensity (a.u.)
HAAP 5000
0 700
750
800
850
Wavelength (nm)
Fig. 2. Steady-state fluorescence spectra of films of pure perovskites, perovskites/PCBM, PEDOT:PSS/perovskites, and PEDOT:PSS/perovskites/PCBM fabricated by using the conventional, MPT-, and HAAP-perovskite, respectively. The excitation wavelength of 470 nm is used.
of the HAAP-perovskite is almost completely quenched in the PEDOT: PSS/HAAP-perovskite, HAAP-perovskite/PCBM, and PEDOT:PSS/HAAPperovskite/PCBM films, while fluorescence of the conventional and MPT-perovskites are not completely quenched in the PEDOT:PSS/conventional perovskite, conventional perovskite/PCBM, and MPTperovskite/PCBM films. Complete quenching of the fluorescence is indicative of very efficient charge carrier transport onto PEDOT:PSS and PCBM layers after the photoexcitation of the perovskite layer, while the residual fluorescence intensities indicate significant trapping of charge carriers and recombination at defect sites after photoexcitation of the perovskites and before arrival at the carrier transporting layers. The time constants of the carrier transports are determined by measurement of the time-resolved fluorescence-decay profiles of the multilayered samples. The fluorescence time-decay profiles are measured at short (0–60 ns) and long (0–800 ns) time scales, as plotted in Fig. 3 and its insets, respectively, while the decay-time constants and relative amplitudes obtained from fitting the data to exponential terms are listed in Table 1. Films of the PEDOT:PSS/perovskites/PCBM fabricated using the conventional, MPT-, and HAAP-perovskites consistently show very fast fluorescence time decays (blue lines of Fig. 3), characterized by initial and major (N95%) time decays at b1 ns and long and minor (b5%) time decays at tens of nanoseconds (Table 1). These fast time decays are consistent with the significant quenching of the fluorescence spectra shown in Fig. 2, and demonstrated as resulting from the simultaneous contribution of the hole and electron transfers to the fluorescence quenching. The PEDOT:PSS/perovskite films fabricated using the conventional, MPT-, and HAAP-perovskites show very short initial time decays at ≤0.5 ns, followed by long time decays at tens of nanoseconds, while the perovskites/PCBM films show initial time decays of a few nanoseconds and longtime decays of tens of nanoseconds. Regarding the photoexcitation direction through the PEDOT:PSS and perovskite layers, very
short time decays, i.e., ≤0.5 ns of the PEDOT:PSS/perovskites may result from the fast hole transfer to an adjacent PEDOT:PSS layer after photoexcitation of the perovskite layer: This interpretation is also confirmed by the absence of these fast time-decay signals upon photoexcitation at the opposite direction of the films. Films of the PEDOT:PSS/perovskites and perovskites/PCBM show multi-exponential fluorescence time decays, probably because of involvements of short- and longrange carrier transports before arrival at the carrier transporting layers. We determined the carrier transport time constants by averaging the fluorescence-decay times of the PEDOT:PSS/perovskites and perovskites/PCBM, as listed in Table 1. The conventional perovskite shows the hole and electron transport times of 13.1 and 7.9 ns, as characterized by averaged fluorescence decay times of the PEDOT:PSS/conventional perovskite and conventional perovskite/PCBM, while the MPT-perovskite shows the hole and electron transport times of 2.4 and 28.8 ns. The HAAP-perovskite shows the hole and electron transport times of 7.6 and 7.9 ns. Note that the conventional and MPT-perovskites exhibit unbalanced characteristics for the hole and electron transports, while the HAAP-perovskite shows balanced characteristic. The fluorescence time-decay profiles shown in the insets of Fig. 3 (green and red lines) more clearly show balanced fluorescence time-decay profiles of PEDOT:PSS/HAAP-perovskite and HAAP-perovskite/PCBM films. The balanced hole and electron transport after photoexcitation of the HAAP-perovskite layer may suppress loss processes of carrier accumulation and recombination [9,21,22], leading to the improved device parameters (i.e., Jsc, Voc, and FF) shown in Fig. 1c. We find from Table 1 that the electron transport time is shortened by a factor of 3.6 (from 28.8 to 7.9 ns) after the perovskite film is annealed with hot air, and this eventually leads to the balanced hole and electron transport in the HAAP-perovskite. This result is in good agreement with our previous simulation result about the temperature gradient generated along the longitudinal direction of the film during the annealing processes, and demonstrated as a major factor for determining the PCEs [15]. The MPT preceded by placing the moisture-exposed perovskite film on a hot plate may result in a temperature gradient at the top and bottom sides of the film, while the HAAP preceded by placing the moisture-exposed perovskite film in a hot air oven may eliminate such a temperature gradient, leading to the homogeneous recrystallization of the perovskites along the longitudinal direction of the film. The enhanced electron transport rate after the HAAP is consistent with this recrystallization of the perovskite along the longitudinal direction of the film. 4. Conclusions Using fluorescence spectroscopy, we investigated carrier (electron and hole) transport dynamics of high-performance perovskite (CH3NH3PbI3−xClx) solar cells with the ITO/PEDOT:PSS/perovskites/ PCBM/Ca/Al structure. The fluorescence of the perovskites is significantly quenched by efficient electron and hole transport onto respective carrier transporting layers (PEDOT:PSS and PCBM) after photoexcitation of the perovskite layers. From analyses of the fluorescence-decay profiles for conventional, MPT-, and HAAP-perovskite layers coated on and under the carrier transporting layers of PEDOT:PSS and PCBM, we found that the most-efficient cell, fabricated by using both MPT and HAAP, exhibits the fast and balanced transport of electrons and holes, as characterized by the time constants of 7.9 and 7.6 ns, respectively, while the other cells fabricated by using the conventional and MPT-perovskites do not show such balanced characteristics. The balanced carrier transport results from the accelerated electron transport rate realized by annealing the perovskite layer with hot air, which leads to complete recrystallization of the perovskite along the longitudinal direction of the film. Acknowledgement This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the
71
Perovskite Perovskite/PCBM PEDOT:PSS/perovskite PEDOT:PSS/perovskite/PCBM
Conventional
1
1
0.1 PL intensity
PL intensity (normalized)
I.-W. Hwang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 207 (2019) 68–72
0.01
Long time 0.1
0.01
0
200
400
600
800
1
MPT
0.1
1
PL intensity
PL intensity (normalized)
Time (ns)
0.01
Long time 0.1
0.01
0
200
400
600
800
1
HAAP 1
0.1 PL intensity
PL intensity (normalized)
Time (ns)
0.01
Long time
0.1
0.01
0
200
400
600
800
Time (ns)
0
20
40
60
80
Time (ns) Fig. 3. Time-resolved fluorescence-decay profiles of films of pure perovskites, perovskites/PCBM, PEDOT:PSS/perovskites, and PEDOT:PSS/perovskites/PCBM fabricated by using the conventional, MPT-, and HAAP-perovskite, respectively. The insets show the fluorescence-decay profiles plotted at long time scales. The excitation wavelength of 470 nm is used. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
Ministry of Education, Science and Technology (NRF2017R1D1A1B03030669), the GIST Research Institute (GRI) grant funded by the GIST in 2018, and the Korea Technology and Information Promotion Agency for SMEs (C0565685).
Table 1 Fluorescence-decay parameters of conventional, MPT-, and HAAP-perovskites.a Sample structures
Fitted decay timesb (ns) τ1
τ2
τavgc
Conventional Perovskite PEDOT:PSS/perovskite Perovskite/PCBM PEDOT:PSS/perovskite/PCBM
67.4 (100%) 0.4 (81%) 4.0 (61%) 0.8 (95%)
67.0 (19%) 14.1 (39%) 23.4 (5%)
67.4 13.1 7.9 1.9
MPT Perovskite PEDOT:PSS/perovskite Perovskite/PCBM PEDOT:PSS/perovskite/PCBM
207.1 (100%) 0.5 (92%) 8.1 (53%) 1.0 (98%)
23.7 (8%) 52.2 (47%) 22.9 (2%)
207.1 2.4 28.8 1.4
HAAP Perovskite PEDOT:PSS/perovskite Perovskite/PCBM PEDOT:PSS/perovskite/PCBM
397.2 (100%) 0.4 (75%) 3.1 (73%) 0.8 (96%)
29.0 (25%) 20.9 (27%) 28.0 (4%)
397.2 7.6 7.9 1.9
a
The excitation and emission wavelength of 470 and 770 nm, respectively, is used. The following fitting function is used: I(t) = A1 exp(−t/τ1) + A2 exp(−t/τ2), where I (t) is the time-dependent fluorescence intensity, A is the amplitude (noted in parentheses as the normalized percentage, i.e., [Ai / (A1 + A2)] × 100), and τ is the fitted decay time. c The averaged decay times (τavg) are obtained using (A1τ1 + A2τ2) / (A1 + A2). The χ2 values for the deconvolution fitting are 1.0–1.5. b
References [1] M.M. Lee, J. Teuscher, T. Miyasaka, T.N. Murakami, H.J. Snaith, Science 338 (2012) 643. [2] C.-W. Chen, H.-W. Kang, S.-Y. Hsiao, P.-F. Yang, K.-M. Chiang, H.-W. Lin, Adv. Mater. 26 (2014) 6647. [3] C.R. Kagan, D.B. Mitzi, C.D. Dimitrakopoulos, Science 286 (1999) 945. [4] J.H. Heo, S.H. Im, J.H. Noh, T.N. Mandal, C.-S. Lim, J.A. Chang, Y.H. Lee, H. Kim, A. Sarkar, M.K. Nazeeruddin, M. Grätzel, S.I. Seok, Nat. Photonics 7 (2013) 486. [5] V. D'Innocenzo, G. Grancini, M.J.P. Alcocer, A.R.S. Kandada, S.D. Stranks, M.M. Lee, G. Lanzani, H.J. Snaith, A. Petrozza, Nat. Commun. 5 (2014) 3586. [6] H.-S. Kim, S.H. Im, N.-G. Park, J. Phys. Chem. C 118 (2014) 5615. [7] M. Saba, M. Cadelano, D. Marongiu, F. Chen, V. Sarritzu, N. Sestu, C. Figus, M. Aresti, R. Piras, A. Geddo Lehmann, C. Cannas, A. Musinu, F. Quochi, A. Mura, G. Bongiovanni, Nat. Commun. 5 (2014) 5049. [8] S.D. Stranks, G.E. Eperon, G. Grancini, C. Menelaou, M.J.P. Alcocer, T. Leijtens, L.M. Herz, A. Petrozza, H.J. Snaith, Science 342 (2013) 341. [9] G. Xing, N. Mathews, S. Sun, S.S. Lim, Y.M. Lam, M. Grätzel, S. Mhaisalkar, T.C. Sum, Science 342 (2013) 344. [10] C. Wehrenfennig, G.E. Eperon, M.B. Johnston, H.J. Snaith, L.M. Herz, Adv. Mater. 26 (2014) 1584. [11] G.E. Eperon, V.M. Burlakov, P. Docampo, A. Goriely, H.J. Snaith, Adv. Funct. Mater. 24 (2014) 151. [12] T.-B. Song, Q. Chen, H. Zhou, S. Luo, Y. (Michael) Yang, J. You, Y. Yang, Nano Energy 12 (2015) 494.
72
I.-W. Hwang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 207 (2019) 68–72
[13] L. Huang, J. Xu, X. Sun, Y. Du, H. Cai, J. Ni, J. Li, Z. Hu, J. Zhang, ACS Appl. Mater. Interfaces 8 (2016) 9811. [14] L. Huang, Z. Hu, G. Yue, J. Liu, X. Cui, J. Zhang, Y. Zhu, Phys. Chem. Chem. Phys. 17 (2015) 22015. [15] Y. Liu, I. Shin, I.-W. Hwang, J. Lee, S. Kim, D.Y. Lee, S.-H. Lee, J.-W. Jang, Y.K. Jung, J.H. Jeong, S.H. Park, K.H. Kim, Sol. Energy 146 (2017) 359. [16] N.J. Jeon, J.H. Noh, Y.C. Kim, W.S. Yang, S. Ryu, S.I. Seok, Nat. Mater. 13 (2014) 897. [17] I.-W. Hwang, Y.-R. Kim, Chem. Phys. Lett. 571 (2013) 34. [18] I.-W. Hwang, J. Kong, H.K. Yoo, K. Lee, J. Phys. Chem. C 119 (2015) 12896.
[19] D.V. O'Connor, D. Philips, Time-correlated Single Photon Counting, Academic Press, New York, 1984. [20] W. Nie, H. Tsai, R. Asadpour, J.-C. Blancon, A.J. Neukirch, G. Gupta, J.J. Crochet, M. Chhowalla, S. Tretiak, M.A. Alam, H.-L. Wang, A.D. Mohite, Science 347 (2015) 522. [21] Q. Dong, Y. Fang, Y. Shao, P. Mulligan, J. Qiu, L. Cao, J. Huang, Science 347 (2015) 967. [22] K. Chen, Q. Hu, T. Liu, L. Zhao, D. Luo, J. Wu, Y. Zhang, W. Zhang, F. Liu, T.P. Russell, R. Zhu, Q. Gong, Adv. Mater. 28 (2016) 10718.
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Fluorescence spectroscopy-based study of balanced transport of charge carriers in hot-air-annealed perovskites In-Wook Hwang a,⁎, Yanliang Liu b,c, Sung Heum Park b,c a b c
Advanced Photonics Research Institute, Gwangju Institute of Science and Technology, Gwangju 500-712, South Korea Department of Physics, Pukyong National University, Busan 48513, South Korea Hybrid Interface Materials Global Frontier Research Group, Pusan National University, Busan 46241, South Korea
a r t i c l e
i n f o
Article history: Received 19 June 2018 Received in revised form 27 August 2018 Accepted 2 September 2018 Available online 04 September 2018 Keywords: Fluorescence Perovskite Solar cell Charge carrier transport Time-resolved
a b s t r a c t Using fluorescence spectroscopy, we investigate the transport of charge carriers in (p-i-n) planar perovskite (CH3NH3PbI3−xClx) solar cells with the structure ITO/PEDOT:PSS/perovskites/PCBM/Ca/Al, in which the perovskite morphology is optimized by moisture pretreatment (MPT) and the hot-air-annealing process (HAAP). After annealing with hot air, the electron transport time of the perovskites is shortened by a factor of 3.6 (from 28.8 to 7.9 ns), eventually leading to balanced transport of electrons and holes (characterized by the fluorescence-decay time constants of 7.9 and 7.6 ns, respectively). These results are in good agreement with the observed increase in the photovoltaic conversion efficiency. © 2018 Published by Elsevier B.V.
1. Introduction Organic-inorganic halide perovskite solar cells (PESCs) have attracted considerable attention, because of their high device performance, low cost, and ease of fabrication. The high performance results from the excellent characteristics of perovskites, such as high absorbance [1,2], ambipolar carrier transport [3,4], small exciton binding energies (~20 meV) [5–7], long electron–hole diffusion lengths (100–1000 nm), and long exciton lifetimes (~100 ns) [8–10]. To fabricate high-performance PESCs, the morphology of perovskites needs to be controlled to realize high crystallinity, high coverage extent, and large grain sizes [11–14]. The presence of nonuniform crystalline networks in the perovskite layer results in a short electron–hole transport distance and high electron–hole charge recombination loss [13], while a poor surface coverage causes the incident photons to pass through the uncovered areas, leading to a low photocurrent [14]. Recently, we have improved the crystallinity and coverage of perovskites by annealing the pristine perovskite (CH3NH3PbI3−xClx) layer with exposure to moisture and hot air, and have realized enhancement in the photovoltaic conversion efficiency from 9.05% to 15.55% [15]. In this study, with the goal of understanding what currently limits the photovoltaic conversion efficiency in this system, we comparatively investigate the carrier transport dynamics of various PESCs. From analyses of the fluorescence-decay profiles of the conventional, moisture ⁎ Corresponding author. E-mail address: [email protected] (I.-W. Hwang).
https://doi.org/10.1016/j.saa.2018.09.005 1386-1425/© 2018 Published by Elsevier B.V.
pretreatment (MPT)- and hot-air-annealing process (HAAP)-perovskite layers coated on or under the carrier transporting layers of PEDOT:PSS and PCBM or sandwiched at their interface, we characterize the time constants for the transport of electrons and holes onto respective transporting layers. From analyses, we find that the efficiency of PESCs is significantly influenced by the carrier transport rates. Notably, the most-efficient cell fabricated by using both processes of MPT and HAAP exhibits short and balanced hole and electron transport times, consistent with enhancements in the short circuit current (Jsc), open circuit voltage (Voc), and fill factor of PESCs. 2. Experimental All reagents purchased from Sigma–Aldrich were used without purification. Methylammonium iodide (MAI) was synthesized according to the literature [16]. Methylamine (24 mL and 33 wt% in absolute ethanol) and HI (10 mL and 57 wt% in water) were reacted in a round bottom flask under N2 atmosphere at 0 °C for 2 h with stirring. Then, the resulting solution was dried by rotary evaporation at 50 °C to form a white powder of MAI. For the purification of the pristine MAI, the dried MAI powder was dissolved in ethanol followed by sedimentation in diethyl ether by stirring the solution. This process was repeated three times and an MAI powder with high purity was recovered and dried in a vacuum oven for 24 h at 60 °C. Thin-film perovskite samples were fabricated as follows. A quartz substrate was cleaned with detergent, ultrasonicated in acetone and isopropyl alcohol, and subsequently dried in an oven at 100 °C.
I.-W. Hwang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 207 (2019) 68–72
PEDOT:PSS (Baytron PH) was spin-casted on quartz at 4500 rpm from an aqueous solution to form a 40 nm thick film. The PEDOT:PSS-coated substrate was dried for 10 min at 140 °C in air and then transferred into a glove box to spin-cast the perovskite layers. A solution containing a mixture of MAI:PbCl2 with a molar ratio of 3:1 in DMF solvent (40 wt %) was spin-casted at 5000 rpm for 45 s on top of pure or PEDOT:PSScoated quart substrate. The precursor perovskite film was dried in the glove box for 30 min at room temperature. The thickness of the perovskite films was ~350 nm. In conventional process, the precursor perovskite film was directly heated on a hot plate in the glove box at 120 °C for 30 min. Meanwhile, in the moisture pretreatment process, the film was placed in a humid atmosphere (relative humidity 60%) for 3 min before the annealing process and subsequently annealed on a hot plate. For hot air annealing, the film was annealed in a hot-air environment in a N2 filled dry oven at 120 °C for 30 min after the moisture pretreatment. The PC61BM (20 mg/mL in chlorobenzene) was deposited by spin coating at 1000 rpm for 40 s to form a complete film. Further details of fabrication of the perovskite films and PESCs have been reported in a previous paper [15]. The steady-state fluorescence spectra and time-resolved fluorescence-decay profiles are recorded using a Hitachi F-4500 fluorescence spectrophotometer and a time-correlated single photon counting (TCSPC) system, respectively [17,18]. In the fluorescence measurements, we deliberately photoexcite the quartz sides of films of quartz/ perovskites, quartz/PEDOT:PSS/perovskites, quartz/perovskites/PCBM, and quartz/PEDOT:PSS/perovskites/PCBM for monitoring carrier transport dynamics under the same condition with sun light irradiation onto PESCs: Intensities of the steady-state fluorescence spectra and profiles of the time-resolved fluorescence-decays are also found to be influenced by the direction of photoexcitation of the films. The decay-time constants for the fluorescence-decay profiles are obtained by first deconvoluting the measured signal from the pump time profile (characterized by a full width at half maximum of ~100 ps) and then fitting to a sum of exponential terms [19]. The chi-square (χ2) values of the fitting are 1.0–1.5.
69
3. Results and Discussion The scanning electron microscope images and time-resolved fluorescence-decay profiles of the conventional, MPT-, and HAAPperovskite films and corresponding solar cell device characteristics (current–voltage curves) are shown in Fig. 1. After annealing the perovskite film with moisture and hot air, the number of pinholes systematically decreases, while the fluorescence-decay time and device power conversion efficiency (PCE) increase, as characterized by the decaytime constants of 68, 207, and 397 ns and the PCEs of 9.05%, 11.87%, and 15.55%, respectively (Fig. 1). We note that in Fig. 1b, the timeresolved fluorescence-decay profiles of the perovskite samples are consistently fitted by using single exponential-decay functions. These single exponential decays are comparable to those of other small grain-sized perovskites with diameters of a few micrometres [20]. Lie et al., previously demonstrated that these single exponential decays result from trap-assisted carrier recombination rather than free-carrier-assisted carrier recombination characterized by multi-exponential fluorescence-decay profiles of larger grain-sized perovskites [20]. Following these, the elongated fluorescence-decay times after annealing the perovskites are indicative of reduced number of carrier trapping sites, in good agreement with the reduced number of pin holes shown in Fig. 1a. Note that in Fig. 2, the reduced number of carrier trapping sites after annealing the perovskites results in the intensified fluorescence spectra of MPT- and HAAP-perovskites compared to the one of conventional perovsite. Charge carrier transports of PESCs are investigated by comparing the steady-state fluorescence spectra and time-resolved fluorescencedecay profiles obtained from films of pure perovskites, PEDOT:PSS/perovskites, perovskites/PCBM, and PEDOT:PSS/perovskites/PCBM, fabricated with the conventional, MPT- and HAAP-perovskite layers (Figs. 2 and 3). When coated on or under the carrier transporting layers of PEDOT: PSS and PCBM, the conventional, MPT- and HAAP-perovskites exhibit differing degrees of fluorescence quenching (Fig. 2). The fluorescence
a
c 1
0.1
0.01
5
2
Conventinal MPT HAAP
Current density (mA/cm )
PL intensity (normalized)
b
0
-10 -15 -20 -25
0
200
400
Time (ns)
600
800
Conventional MPT HAAP
-5
0.0
0.2
0.4
0.6
0.8
1.0
Voltage (V)
Fig. 1. (a) Scanning electron microscope images, (b) time-resolved fluorescence-decay profiles of the conventional, MPT-, and HAAP-perovskite films, and (c) current density-voltage (J-V) characteristics of corresponding PESCs, adapted from [15]. The excitation wavelength of 470 nm is used for measurement of time-resolved fluorescence-decay profiles. The optimized HAAP cell shows the best device characteristic parameters (PCE, FF, Jsc, and Voc of 15.55%, 80.30%, 20.39 mA/cm2, and 0.95 V, respectively), compared to the conventional cell (9.05%, 64.43%, 16.34 mA/cm2, and 0.67 V, respectively) and the MPT cell (11.87%, 74.07%, 17.80 mA/cm2, and 0.90 V, respectively).
70
I.-W. Hwang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 207 (2019) 68–72
PL intensity (a.u.)
Conventional
Perovskite Perovskite/PCBM PEDOT:PSS/perovskite PEDOT:PSS/perovskite/PCBM
2000
0
PL intensity (a.u.)
MPT 5000
0
PL intensity (a.u.)
HAAP 5000
0 700
750
800
850
Wavelength (nm)
Fig. 2. Steady-state fluorescence spectra of films of pure perovskites, perovskites/PCBM, PEDOT:PSS/perovskites, and PEDOT:PSS/perovskites/PCBM fabricated by using the conventional, MPT-, and HAAP-perovskite, respectively. The excitation wavelength of 470 nm is used.
of the HAAP-perovskite is almost completely quenched in the PEDOT: PSS/HAAP-perovskite, HAAP-perovskite/PCBM, and PEDOT:PSS/HAAPperovskite/PCBM films, while fluorescence of the conventional and MPT-perovskites are not completely quenched in the PEDOT:PSS/conventional perovskite, conventional perovskite/PCBM, and MPTperovskite/PCBM films. Complete quenching of the fluorescence is indicative of very efficient charge carrier transport onto PEDOT:PSS and PCBM layers after the photoexcitation of the perovskite layer, while the residual fluorescence intensities indicate significant trapping of charge carriers and recombination at defect sites after photoexcitation of the perovskites and before arrival at the carrier transporting layers. The time constants of the carrier transports are determined by measurement of the time-resolved fluorescence-decay profiles of the multilayered samples. The fluorescence time-decay profiles are measured at short (0–60 ns) and long (0–800 ns) time scales, as plotted in Fig. 3 and its insets, respectively, while the decay-time constants and relative amplitudes obtained from fitting the data to exponential terms are listed in Table 1. Films of the PEDOT:PSS/perovskites/PCBM fabricated using the conventional, MPT-, and HAAP-perovskites consistently show very fast fluorescence time decays (blue lines of Fig. 3), characterized by initial and major (N95%) time decays at b1 ns and long and minor (b5%) time decays at tens of nanoseconds (Table 1). These fast time decays are consistent with the significant quenching of the fluorescence spectra shown in Fig. 2, and demonstrated as resulting from the simultaneous contribution of the hole and electron transfers to the fluorescence quenching. The PEDOT:PSS/perovskite films fabricated using the conventional, MPT-, and HAAP-perovskites show very short initial time decays at ≤0.5 ns, followed by long time decays at tens of nanoseconds, while the perovskites/PCBM films show initial time decays of a few nanoseconds and longtime decays of tens of nanoseconds. Regarding the photoexcitation direction through the PEDOT:PSS and perovskite layers, very
short time decays, i.e., ≤0.5 ns of the PEDOT:PSS/perovskites may result from the fast hole transfer to an adjacent PEDOT:PSS layer after photoexcitation of the perovskite layer: This interpretation is also confirmed by the absence of these fast time-decay signals upon photoexcitation at the opposite direction of the films. Films of the PEDOT:PSS/perovskites and perovskites/PCBM show multi-exponential fluorescence time decays, probably because of involvements of short- and longrange carrier transports before arrival at the carrier transporting layers. We determined the carrier transport time constants by averaging the fluorescence-decay times of the PEDOT:PSS/perovskites and perovskites/PCBM, as listed in Table 1. The conventional perovskite shows the hole and electron transport times of 13.1 and 7.9 ns, as characterized by averaged fluorescence decay times of the PEDOT:PSS/conventional perovskite and conventional perovskite/PCBM, while the MPT-perovskite shows the hole and electron transport times of 2.4 and 28.8 ns. The HAAP-perovskite shows the hole and electron transport times of 7.6 and 7.9 ns. Note that the conventional and MPT-perovskites exhibit unbalanced characteristics for the hole and electron transports, while the HAAP-perovskite shows balanced characteristic. The fluorescence time-decay profiles shown in the insets of Fig. 3 (green and red lines) more clearly show balanced fluorescence time-decay profiles of PEDOT:PSS/HAAP-perovskite and HAAP-perovskite/PCBM films. The balanced hole and electron transport after photoexcitation of the HAAP-perovskite layer may suppress loss processes of carrier accumulation and recombination [9,21,22], leading to the improved device parameters (i.e., Jsc, Voc, and FF) shown in Fig. 1c. We find from Table 1 that the electron transport time is shortened by a factor of 3.6 (from 28.8 to 7.9 ns) after the perovskite film is annealed with hot air, and this eventually leads to the balanced hole and electron transport in the HAAP-perovskite. This result is in good agreement with our previous simulation result about the temperature gradient generated along the longitudinal direction of the film during the annealing processes, and demonstrated as a major factor for determining the PCEs [15]. The MPT preceded by placing the moisture-exposed perovskite film on a hot plate may result in a temperature gradient at the top and bottom sides of the film, while the HAAP preceded by placing the moisture-exposed perovskite film in a hot air oven may eliminate such a temperature gradient, leading to the homogeneous recrystallization of the perovskites along the longitudinal direction of the film. The enhanced electron transport rate after the HAAP is consistent with this recrystallization of the perovskite along the longitudinal direction of the film. 4. Conclusions Using fluorescence spectroscopy, we investigated carrier (electron and hole) transport dynamics of high-performance perovskite (CH3NH3PbI3−xClx) solar cells with the ITO/PEDOT:PSS/perovskites/ PCBM/Ca/Al structure. The fluorescence of the perovskites is significantly quenched by efficient electron and hole transport onto respective carrier transporting layers (PEDOT:PSS and PCBM) after photoexcitation of the perovskite layers. From analyses of the fluorescence-decay profiles for conventional, MPT-, and HAAP-perovskite layers coated on and under the carrier transporting layers of PEDOT:PSS and PCBM, we found that the most-efficient cell, fabricated by using both MPT and HAAP, exhibits the fast and balanced transport of electrons and holes, as characterized by the time constants of 7.9 and 7.6 ns, respectively, while the other cells fabricated by using the conventional and MPT-perovskites do not show such balanced characteristics. The balanced carrier transport results from the accelerated electron transport rate realized by annealing the perovskite layer with hot air, which leads to complete recrystallization of the perovskite along the longitudinal direction of the film. Acknowledgement This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the
71
Perovskite Perovskite/PCBM PEDOT:PSS/perovskite PEDOT:PSS/perovskite/PCBM
Conventional
1
1
0.1 PL intensity
PL intensity (normalized)
I.-W. Hwang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 207 (2019) 68–72
0.01
Long time 0.1
0.01
0
200
400
600
800
1
MPT
0.1
1
PL intensity
PL intensity (normalized)
Time (ns)
0.01
Long time 0.1
0.01
0
200
400
600
800
1
HAAP 1
0.1 PL intensity
PL intensity (normalized)
Time (ns)
0.01
Long time
0.1
0.01
0
200
400
600
800
Time (ns)
0
20
40
60
80
Time (ns) Fig. 3. Time-resolved fluorescence-decay profiles of films of pure perovskites, perovskites/PCBM, PEDOT:PSS/perovskites, and PEDOT:PSS/perovskites/PCBM fabricated by using the conventional, MPT-, and HAAP-perovskite, respectively. The insets show the fluorescence-decay profiles plotted at long time scales. The excitation wavelength of 470 nm is used. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
Ministry of Education, Science and Technology (NRF2017R1D1A1B03030669), the GIST Research Institute (GRI) grant funded by the GIST in 2018, and the Korea Technology and Information Promotion Agency for SMEs (C0565685).
Table 1 Fluorescence-decay parameters of conventional, MPT-, and HAAP-perovskites.a Sample structures
Fitted decay timesb (ns) τ1
τ2
τavgc
Conventional Perovskite PEDOT:PSS/perovskite Perovskite/PCBM PEDOT:PSS/perovskite/PCBM
67.4 (100%) 0.4 (81%) 4.0 (61%) 0.8 (95%)
67.0 (19%) 14.1 (39%) 23.4 (5%)
67.4 13.1 7.9 1.9
MPT Perovskite PEDOT:PSS/perovskite Perovskite/PCBM PEDOT:PSS/perovskite/PCBM
207.1 (100%) 0.5 (92%) 8.1 (53%) 1.0 (98%)
23.7 (8%) 52.2 (47%) 22.9 (2%)
207.1 2.4 28.8 1.4
HAAP Perovskite PEDOT:PSS/perovskite Perovskite/PCBM PEDOT:PSS/perovskite/PCBM
397.2 (100%) 0.4 (75%) 3.1 (73%) 0.8 (96%)
29.0 (25%) 20.9 (27%) 28.0 (4%)
397.2 7.6 7.9 1.9
a
The excitation and emission wavelength of 470 and 770 nm, respectively, is used. The following fitting function is used: I(t) = A1 exp(−t/τ1) + A2 exp(−t/τ2), where I (t) is the time-dependent fluorescence intensity, A is the amplitude (noted in parentheses as the normalized percentage, i.e., [Ai / (A1 + A2)] × 100), and τ is the fitted decay time. c The averaged decay times (τavg) are obtained using (A1τ1 + A2τ2) / (A1 + A2). The χ2 values for the deconvolution fitting are 1.0–1.5. b
References [1] M.M. Lee, J. Teuscher, T. Miyasaka, T.N. Murakami, H.J. Snaith, Science 338 (2012) 643. [2] C.-W. Chen, H.-W. Kang, S.-Y. Hsiao, P.-F. Yang, K.-M. Chiang, H.-W. Lin, Adv. Mater. 26 (2014) 6647. [3] C.R. Kagan, D.B. Mitzi, C.D. Dimitrakopoulos, Science 286 (1999) 945. [4] J.H. Heo, S.H. Im, J.H. Noh, T.N. Mandal, C.-S. Lim, J.A. Chang, Y.H. Lee, H. Kim, A. Sarkar, M.K. Nazeeruddin, M. Grätzel, S.I. Seok, Nat. Photonics 7 (2013) 486. [5] V. D'Innocenzo, G. Grancini, M.J.P. Alcocer, A.R.S. Kandada, S.D. Stranks, M.M. Lee, G. Lanzani, H.J. Snaith, A. Petrozza, Nat. Commun. 5 (2014) 3586. [6] H.-S. Kim, S.H. Im, N.-G. Park, J. Phys. Chem. C 118 (2014) 5615. [7] M. Saba, M. Cadelano, D. Marongiu, F. Chen, V. Sarritzu, N. Sestu, C. Figus, M. Aresti, R. Piras, A. Geddo Lehmann, C. Cannas, A. Musinu, F. Quochi, A. Mura, G. Bongiovanni, Nat. Commun. 5 (2014) 5049. [8] S.D. Stranks, G.E. Eperon, G. Grancini, C. Menelaou, M.J.P. Alcocer, T. Leijtens, L.M. Herz, A. Petrozza, H.J. Snaith, Science 342 (2013) 341. [9] G. Xing, N. Mathews, S. Sun, S.S. Lim, Y.M. Lam, M. Grätzel, S. Mhaisalkar, T.C. Sum, Science 342 (2013) 344. [10] C. Wehrenfennig, G.E. Eperon, M.B. Johnston, H.J. Snaith, L.M. Herz, Adv. Mater. 26 (2014) 1584. [11] G.E. Eperon, V.M. Burlakov, P. Docampo, A. Goriely, H.J. Snaith, Adv. Funct. Mater. 24 (2014) 151. [12] T.-B. Song, Q. Chen, H. Zhou, S. Luo, Y. (Michael) Yang, J. You, Y. Yang, Nano Energy 12 (2015) 494.
72
I.-W. Hwang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 207 (2019) 68–72
[13] L. Huang, J. Xu, X. Sun, Y. Du, H. Cai, J. Ni, J. Li, Z. Hu, J. Zhang, ACS Appl. Mater. Interfaces 8 (2016) 9811. [14] L. Huang, Z. Hu, G. Yue, J. Liu, X. Cui, J. Zhang, Y. Zhu, Phys. Chem. Chem. Phys. 17 (2015) 22015. [15] Y. Liu, I. Shin, I.-W. Hwang, J. Lee, S. Kim, D.Y. Lee, S.-H. Lee, J.-W. Jang, Y.K. Jung, J.H. Jeong, S.H. Park, K.H. Kim, Sol. Energy 146 (2017) 359. [16] N.J. Jeon, J.H. Noh, Y.C. Kim, W.S. Yang, S. Ryu, S.I. Seok, Nat. Mater. 13 (2014) 897. [17] I.-W. Hwang, Y.-R. Kim, Chem. Phys. Lett. 571 (2013) 34. [18] I.-W. Hwang, J. Kong, H.K. Yoo, K. Lee, J. Phys. Chem. C 119 (2015) 12896.
[19] D.V. O'Connor, D. Philips, Time-correlated Single Photon Counting, Academic Press, New York, 1984. [20] W. Nie, H. Tsai, R. Asadpour, J.-C. Blancon, A.J. Neukirch, G. Gupta, J.J. Crochet, M. Chhowalla, S. Tretiak, M.A. Alam, H.-L. Wang, A.D. Mohite, Science 347 (2015) 522. [21] Q. Dong, Y. Fang, Y. Shao, P. Mulligan, J. Qiu, L. Cao, J. Huang, Science 347 (2015) 967. [22] K. Chen, Q. Hu, T. Liu, L. Zhao, D. Luo, J. Wu, Y. Zhang, W. Zhang, F. Liu, T.P. Russell, R. Zhu, Q. Gong, Adv. Mater. 28 (2016) 10718.