Charge transport and recombination in bulk-heterojunction solar - PowerPoint PPT Presentation

Charge transport and recombination in bulk-heterojunction solar cells Ronald Österbacka Department of Physics and Center for Functional Materials Åbo Akademi University http://www.funmat.fi Finnish-Japanese Workshop on Functional Materials, 25-26.05.2009

FunMat Schematic Page 2 29.5.2009

Center for Functional Materials Our vision We believe that new functional materials will enable the printed intelligence revolution Our mission To create a strong multidisciplinary research environment for the development of new materials and demonstrating new functionalities by printing Key objectives • Excellence and innovativeness in research • Highly inter- and multidisciplinary approach to research • Strong national and international networking • To become the leader in paper based printed intelligence research Page 3 29.5.2009

Outline • Introduction to charge transport and recombination – Effect of Langevin recombination • Recombination studies with TOF – In pure polymers – In annealed bulk-heterojunction solar cells • Double Injection Transients (DoI) – Effect of trapping • Suggested model – Nanomorphology important – 2D delocalization of charge carriers • Effect of reduced recombination on magnetortransport • Summary

Bulk-Heterojunction Solar Cells PEDOT:PSS ITO Aluminum Glass OMe O O ) n Light ( OMe PCE ~ 2.5% I Mixture of electron accepting PCBM and electron donating polymer G. Yu, et al. Science 270 , 1789 (1995), J.J.M. Halls, et al., Nature 376 , 498 (1995)

Second generation BHSC j U FF sc oc PCE P light PCE ~ 5% FF= U oc j sc Xiaoniu Yang, et al., Nano Letters, 5 , 579-583 (2005)

Why transport and recombination? n = carrier density Efficiency proportional to the current = carrier mobility j en E e = electron charge E =electric field • Organic materials have low mobilities • To have same efficiency we need higher carrier densities • Higher density leads to lower carrier lifetime -> Lower current! 1 n ( 0 ) and the second order recombination parameter are important • parameters for testing suitable materials. • Main goal – to understand transport and recombination in polymeric solar cells.

Langevin Recombination Expected in low-mobility ( <1 cm 2 /Vs) materials Langevin recombination is determined by the probability for the charge carriers to meet in space, independent of the subsequent fate of the carriers dp dn e ( ) 2 n p np n ( F , T ) L f L L dt dt 0 Necessary condition: The carrier mean free path is much smaller than the Coulomb capture radius r c , i.e. a << r C. 2 e r 19 nm c 4 kT 0 15mA/cm 2 for d=300 To reach a photocurrent density of L / > 5 · 10 -3 cm 2 /Vs . nm and V oc =0.5 V: P. Langevin, Ann. Chim. Phys. 28 , 433 (1903).

Consequences of Langevin recombination • Langevin recombination leads to low photogeneration efficiency of Onsager type – Field-dependent generation! – Lower the Fill-Factor Q ext [Arb. Units] r r c kT Untreated Treated -1 0 1 2 3 4 U 0 [V] A. Pivrikas et al, unpublished

Consequences of Langevin recombination • Langevin recombination leads to low photogeneration efficiency of Onsager type – Field-dependent generation! – Lower the Fill-Factor • Efficiency will be limited – Only ~CU can be extracted from the device Bimolecular 1 t tr t 0 0 t tr t tr t t lifetime Photo-SCLC: ne ne n Langevin!

Recombination measured using TOF t [ms] 0.0 0.2 0.4 0.6 1.5 0.3 L = 300 J t e L = 100 J t cusp L = 30 J 0.2 L = 10 J t e j [mA] L = 3 J 1.0 L = 1 J j/j SCLC L = 0.3 J 0.1 L = 0.1 J L = 0.03 J -3 J L = 10 0.0 P3HT/PCBM Q = CU 0.5 0 20 Q <<CU 0 RRa-PHT 15 110 J 0.0 2 ] j [ A/cm 71 J 0 1 2 3 4 40 J 10 t/t tr 13 J 4.7 J 5 1.6 J 0.3 J 0.07 J 1. Small Charge Current (SCC) mode. 0 0 1 2 3 4 2. Space Charge Perturbed Current (SCPC). t [ms] 3. Space Charge Limited Current (SCLC). A. Pivrikas et al. PRB 71 , 125205, (2005) A. Pivrikas, et. al., PRL 94 , 176806 (2005).

Reduced Recombination in RRPHT/PCBM Solar cells / L =0,0001 / L =0,01 10 / L =1 Q e /CU 0 1 Q ext 30CU 0 a) 0.1 / L =0,0001 150 / L =0,01 / L =1 100 t 1/2 /t tr b) 50 0 0.1 1 10 100 1000 L =10 -4 / Q 0 /CU 0 A. Pivrikas, et. al., PRL. 94 , 176806 (2005).

Double Injection (DoI) Currents t/t a • Apply a voltage pulse 0 2 4 6 8 • Record the transient (a) 6 j (t) current t 1/2 j/j ; dj / dt j=j S - j m 4 • First the RC-current is observed 2 • Then the build-up of a d j /d t t m 0 carrier density is observed (b) 120 10 V • The saturation curent is 2 ] j [ m A/cm 80 t m 7V limited by recombination 5V • At switch-off, a reservoir 40 3V extraction is observed 0 0.0 0.2 0.4 t [ms] R.H. Dean. J. Appl. Phys. 40 , 585 (1969) Mark & Lampert, Current Injection in Solids , Academic Press NY (1970)

Simulated current transients Q s / L =0.001 -3 L =10 / 10 Q 10 0.01 Q ex /CU j/j SCLC 0.1 d Q /d t 1 1 1 t slow t cusp 0.1 1 10 100 0.1 1 10 100 t p /t tr t/t tr As a function of pulse width As a function of G. Juska et al., JAP 101, 114305 (2007)

Effect of trapping n / p =100 10 without trapping j / j SCLC 1 slow-carrier trapping faster carrier trapping 0.1 0.1 1 10 100 1000 t / t tr

Trapping in real solar cells a-Si:H RR-P3HT/PCBM a-Si:H thickness 6 m 2 ) current density ( A/cm -3 10 delay between pulses 0.1 s single shot -4 10 transit time -7 -6 -5 -4 -3 -2 10 10 10 10 10 10 time (s) Note the absence of trapping in annealed RRPHT/PCBM BHSC! Juska et al., in press

Why is there reduced recombination? After annealing Before annealing Xiaoniu Yang, et al., Nano Letters, 5 , 579-583 (2005)

A possible model for reduced recombination in RR-PHT/PCBM • Increased carrier generation / reduced recombination due to an effective energy barrier for geminate pair recombination at the interface ( ~0.5 eV Osikowicz et al.) • To reach the chain nearest the interface the hole on the polymer has to overcome the same barrier • Probability for carriers to meet in space is reduced! • Requires lamellar structures at the interface V. Arkhipov et al., APL 82 , 4605 (2003); Osikowicz et al., Advanced Materials 19 , 4213 (2007)

Lamellar structures important Head-to-tail ~ 80% Head-to-tail = 100% Sirringhaus et al., Nature 401 , 685 (1999); Österbacka et al., Science 287 , 839 (2000)

Langevin recombination in MDMO- PPV/PCBM systems 4 Increasing Intensity 10 3 2 ] j [mA/cm 1 e /CU 0 1 2 Q 0.1 2 L 1 0.01 -8 -6 -4 -2 0 10 10 10 10 10 I/I 0 0 0.0 -5 -5 4.0x10 8.0x10 t [s] Pivrikas et al., Nonlinear Optics, Quantum Optics: Concepts in Modern Optics, 37 , 169-177 (2007)

Role of e-h recombination on OMAR 1 10 0 10 -1 10 % MR -2 10 P3HT Diode L ~1 MDMO-PPV/PCBM L ~0.5 -3 10 -1 Un-treated L ~10 -3 Treated L ~10 -4 10 -300 -200 -100 0 100 200 300 B (mT) S. Majumdar et al., Physical Review B 79 , 201202R (2009).

Correlation with magnetic behavior -8 2x10 P3HT Magnetization (Am2) -8 1x10 1 10 0 L ~ 1 -8 -1x10 L ~ 0.5 0 10 | % MR | -1 -8 L ~ 10 -2x10 -3 L ~ 10 -200 -100 0 100 200 -1 10 B (mT) -7 2x10 -2 5 K 10 100 K -7 1x10 -300 -200 -100 0 100 200 300 Magnetization B (mT) 0 P3HT/PCBM -7 -1x10 -7 -2x10 -200 -100 0 100 200 B (mT) S. Majumdar et al., submitted. V.I. Krinichnyi, Solar cells and Solar energy Materials 92 , 942 (2008).

Summary • Bimolecular recombination is important for characterization of solar cell materials • Langevin recombination: untreated treated 0.5 – Probability for charge carriers to meet in space 0.4 0.3 EQE – Field dependent generation of Onsager type 0.2 – Lower extraction efficiency 0.1 400 500 600 700 800 900 • Treated RR-PHT/PCBM blends shows greatly reduced wavelength [nm] recombination compared to Langevin L ~10 -3 ~ • Untreated RR-PHT/PCBM blends show ~ L ~ 0.1 • MDMO-PPV/PCBM blends show ~ L ~ 0.5 • Nanomorphology important ->delocalization in lamellas important! • DoI is an extremely useful measurement technique for materials with reduced recombination • Take-home message: The probability to form electron – hole pairs is crucial for the magnetotransport response.

The people! • S. Majumdar, H. Majumdar, T. Mäkelä, L. Jian, R. Mohan, ÅA • H. Aarnio, G. Sliauzys, J.K. Baral, N. Kaihovirta, D. Tobjörk, F. Jansson, N. Björklund, M. Nyman, S. Sanden, F. Petterson, ÅA • H. Stubb (emeritus), K.-M. Källman (lab engineer), ÅA • Left the group: A. Pivrikas (Linz), T. Bäcklund (UPM), H. Sandberg (VTT), M. Westerling (Perkin-Elmer) • G. Ju š ka, K. Arlauskas, K. Genevicius/Vilnius Univ • R. Laiho/University of Turku • N.S. Sariciftci, LIOS • G. Dennler and M. Scharber, Konarka Austria • D. Vanderzande, Universiteit Hasselt, Belgium • Financial Support by the Academy of Finland, Technology Development Center TEKES, and Magnus Ehrnrooth foundation

Organic Electronics group 2008 Page 25 29.5.2009

Thank you!