- Email: [email protected]
fullerene blends employing different techniques
Accepted Manuscript Mismatch of non-geminate recombination in polymer TQ1/fullerene blends employing different techniques J. Važgėla, A. Galvelytė, G. Juška PII:
S1566-1199(17)30188-X
DOI:
10.1016/j.orgel.2017.04.030
Reference:
ORGELE 4068
To appear in:
Organic Electronics
Received Date: 17 March 2017 Accepted Date: 27 April 2017
Please cite this article as: J. Važgėla, A. Galvelytė, G. Juška, Mismatch of non-geminate recombination in polymer TQ1/fullerene blends employing different techniques, Organic Electronics (2017), doi: 10.1016/j.orgel.2017.04.030. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Mismatch of Non-Geminate Recombination in Polymer TQ1/Fullerene Blends Employing Different Techniques J. Važgėla1,a), A. Galvelytė1, and G. Juška1 Department of Solid State Electronics, Vilnius University, Saulėtekio 9 III k., 10222 Vilnius, Lithuania
RI PT
1
The non-geminate recombination was investigated in TQ1:PC61BM:PC71BM (poly[[2,3-bis(3-
SC
octyloxyphenyl)-5,8-quinoxalinediyl]-2,5-thiophenediyl]:[6,6]-Phenyl-C61-butyric-acid-methylester:[6,6]-Phenyl-C71-butyric-acid-methyl-ester) bulk heterojunctions employing the extraction of the photogenerated charge carriers by linearly increasing voltage (photo-CELIV) and integral
M AN U
Time-of-Flight (ToF) techniques. The shapes of photo-CELIV and ToF transients point out purely non-reduced Langevin recombination while the decay of the photogenerated charge carrier density measured with photo-CELIV technique indicates reduced third-order recombination. A clear discrepancy could be explained by spatially separated holes and electrons during the delay time while employing photo-CELIV method.
EP
Electronic mail: [email protected].
AC C
a)
TE D
_____________________________
1
ACCEPTED MANUSCRIPT 1. Introduction The power conversion efficiency of single junction organic solar cells (OSCs) and modules have reached 11.2 % and 8.7 % respectively [1]. In order to reach higher
RI PT
efficiencies, it is necessary to enhance the extraction of charge carriers and thus to reduce the recombination. Photo-induced charge carrier extraction by linearly increasing voltage (photo-CELIV) technique has been widely used for the studies of
SC
charge carrier transport properties [2–6]. Photo-CELIV allows to evaluate the relaxation of the photogenerated charge carrier mobility and the recombination of
M AN U
charge carriers [2,5,6]. The recombination rate can be evaluated both from the decay of charge carrier density [3–5] and the shape of current transient [6]. Photo-CELIV results give the information about the recombination rate and order. The concentration decay of electrons in solar cell can be described as:
TE D
dn 1 d jn =− + G − R, dt e dx
(1)
where n is the concentration of electrons, t – time, e – elementary charge, jn – the
EP
current of electrons, G – generation rate, R – recombination rate. In photo-CELIV experiments photogenerated charge carriers are extracted after certain delay time tdel
AC C
(G = 0). In order to avoid the extraction of charge carriers during the delay time tdel, an offset voltage Uoffset (approximately equal to the open circuit voltage Uoc) is applied to the sample. In this case the first member on the right side of Eq. (1) is almost equal to zero. Taking this into account Eq. (1) can be simplified to:
dn = − R = −k0nλ +1, dt
(2)
where k0 is the recombination coeffiecient independent of charge carrier density, λ+1 – the apparent recombination order.
ACCEPTED MANUSCRIPT The integration of Eq. (2) gives the charge carrier density dependence on time:
(
n(t ) = n(0)−λ + λk0t
)
−1 / λ
(3)
.
In disordered materials the recombination is commonly considered as a second-
RI PT
order Langevin type recombination that is limited by the probability for the electrons and holes to meet in space [7]:
R = kL n2 ,
(4)
SC
where kL is Langevin recombination coefficient, kL = e(µn+µp)/εε0, µn (µp) – electron (hole) mobility, ε (ε0) – relative (absolute) dielectric permittivity. However, it
M AN U
was shown that in some polymer blends the recombination is reduced compared with the one predicted by Langevin [3,4,8–14]. In such case the coefficient of the secondorder recombination can be described as:
k = ξkL = ξ
e(µn + µp )
εε0
,
(5)
TE D
where ξ is the Langevin prefactor, ξ = 1 in a pure Langevin recombination case and ξ < 1 if the recombination is reduced.
EP
Nonetheless, experimental results [15–22] and theoretical simulations [23–25] indicate the recombination with the order higher than 2. One of the explanations is
AC C
related to the dependence of non-geminate recombination rate along with the mobility on charge carrier density. This dependence has been assigned to the presence of an energetic tail of localized trap sites [21,22]:
R = k (n)n2 = k0n2+δ ,
(6)
µ (n) = µ0nδ ,
(7)
ACCEPTED MANUSCRIPT where µ0 is charge carrier density-independent mobility, δ – constant, δ > 0. In disordered organic materials charge transport is explained by charge carrier hopping between localized states of the density of states. With increasing disorder the mobility becomes more dependent on charge carrier density (constant δ increases). It was
RI PT
shown that in P3HT:PC61BM blend the recombination order is 2 in case P3HT is completely crystalline (a single crystal) and with the increasing fraction of amorphous phase the concentration of traps and trapped charges increases along with the
SC
recombination order, connected to a trap-assisted recombination [18].
Another explanation of the recombination order being higher than 2 in bulk
M AN U
heterojunction is related to donor–acceptor phase separation [19]. Due to the phase separation trapped charge carrier cannot recombine with the different type of charge carrier. The recombination occurs only when trapped charge carrier becomes mobile and meets a mobile charge carrier which is oppositely charged. To sum up, in the
TE D
presence of charge carrier trapping, the recombination order can be higher than 2 due to the mobility dependence on the charge carrier density and the separation of donor– acceptor phase.
EP
In this paper we have employed photo-CELIV and integral time-of-flight (ToF)
AC C
techniques to investigate the recombination processes in TQ1:PC61BM:PC71BM solar cells. While the shape of photo-CELIV and ToF results indicate a non-reduced Langevin recombination, charge carrier decay evaluated with photo-CELIV technique shows strongly reduced recombination processes. We have shown that in case of concentrated internal electric field, the decay of carrier density measured using photoCELIV technique is governed by strongly reduced recombination with the order
ACCEPTED MANUSCRIPT higher than 2+δ and is extremely approximate. In such a case, photo-CELIV method is not suitable for measuring the charge carrier decay. 2. Materials and methods
RI PT
The samples used for the following experiments were prepared as follows: the glass substrates covered with ITO were cleaned and dried out, then a thin layer of PEDOT:PSS was spin-coated at 3000 RPM for 40 s and heated at 120 ℃ for 15 min. (TQ1),
SC
Poly[[2,3-bis(3-octyloxyphenyl)-5,8-quinoxalinediyl]-2,5-thiophenediyl]
[6,6]-Phenyl-C61-butyric-acid-methyl-ester (PC61BM) and [6,6]-Phenyl-C71-butyric-
M AN U
acid-methyl-ester (PC71BM) with a mass ratio of 5:8:2 were dissolved in o-xylene (20 and 30 mg/ml) and spin-coated (the thickness was 80 nm and 190 nm, respectively) on top of ITO/PEDOT:PSS layers. Afterwards, the layers of LiF (0.6 nm) and Al (90 nm) were evaporated under the high vacuum conditions (5·10-4 Pa). Finally, the
TE D
devices were encapsulated. The active area of the device was 0.048 cm2. Photo-CELIV measurements were performed in three different setups: ordinary (I), ordinary setup with an offset voltage (II) and setup with a voltage switcher (III,
EP
Fig. 1. The schematic view of photo-CELIV experimental setup with a voltage switcher for the investigation of charge carrier recombination.). The measurements of the all setups
AC C
were conducted with a pulsed second-harmonic 532 nm Nd:YAG laser (EKSPLA PL2143), oscilloscope (Tektronix DPO4054B) and generator (Tektronix AFG3022B). The additional generator (Tektronix AFG3022B) and a controllable voltage switcher were used for the III setup. The laser pulse duration was 30 ps and the beam diameter 6 mm. At 532 nm the absorption coefficients of the samples were α = 4.1·104 cm-1 and α = 1.03·105 cm-1, giving αd = 0.33 and αd = 0.82 for 20 mg/ml and 30 mg/ml
ACCEPTED MANUSCRIPT sample respectively. The sample was excited through Al contact side. The laser was synchronized with triangle (Generator II) and square (Generator I, only III setup) voltage pulse generators. In order to avoid the extraction by inserted voltage during the delay time tdel, an offset voltage Uoffset was used in II setup. During the delay time
RI PT
tdel in III setup the extraction of photogenerated charge carriers was avoided by employing a voltage switcher. The sample’s contacts were disconnected from the electrical circuit and after a certain delay time tdel Generator I applied square voltage
SC
pulse to the controllable voltage switcher which connected the contacts to the circuit (III setup). Photogenerated charge carriers were extracted after the same delay time
M AN U
tdel by the application of a linearly increasing voltage.
In case of bulk photogeneration the mobility of charge carriers can be found as [5]:
2d 2 , 2 [1 + 0.36∆j / j(0)] 3 Atmax
TE D
µ=
(8)
where d is the thickness of the active layer, A – the voltage ramp, A = dU/dt, tmax – the time when photo-CELIV transient reaches its maximum (Fig. 2), j(0) – the
AC C
EP
displacement current density, ∆j – the extraction current density.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 1. The schematic view of photo-CELIV experimental setup with a voltage switcher for the investigation of charge carrier recombination. In a time window of 0 ≤ t ≤ tp, where tp is CELIV pulse duration, photo-CELIV transient in a sample consists of a displacement j(0) and extraction ∆j current
TE D
densities (Fig. 2). Applying long enough CELIV pulse, ∆j = 0 at the end of the pulse. Thus, by deducting a dark transient from the illuminated one and integrating the result of it, the extracted charge Qe can be estimated. By changing delay time tdel between
EP
the laser and CELIV pulses, tmax increases (charge carrier mobility decreases, Eq. (8))
AC C
and the extracted charge Qe decreases due to the recombination. Hence, the mobility and carrier density dependences on time (µ(t) and n(t) respectively), where t is the delay time tdel, can be measured [2]. Using the measured mobility dependence on time µ(tdel), Eq. (5), Eq. (3) and the recombination order of 2 (λ = 1), the decay n(tdel) of charge carrier density can be calculated employing the different values of Langevin prefactor ξ. These theoretically calculated n(tdel) trends are compared with the
ACCEPTED MANUSCRIPT measured n(tdel) decay. Langevin prefactor ξ can be found from the theoretically calculated n(tdel) curve that matches experimental n(tdel) results the best. Using photo-CELIV technique Langevin prefactor ξ can be found in a different way. If the recombination is of Langevin type, the recombination is fast and the
RI PT
extraction current density ∆j does not exceed the displacement current density j(0) (Fig. 2) even at the saturation light intensity [26]. If the recombination is reduced (ξ <
intensity Langevin prefactor ξ can be evaluated [6]:
β j(0) = βL ∆j
(9)
M AN U
ξ=
SC
1 in Eq. (5)), ∆j exceeds j(0) and saturates with L. From j(0)/∆j ratio at a high light
Time-of-flight (ToF) experiments were made in integral mode ToF (τRC >> ttr, where τRC – is the time constant of the setup, ttr – charge carrier transit time, ttr = d2/µU, U – applied voltage, d – the thickness of the active layer). Using this
TE D
technique, the coefficient ξ could be estimated as described in Ref. 27:
ξ=
β CU ttr , = βL Qe te
(10)
where C is the capacitance of the sample, Qe – the extracted charge that is equal to
EP
integral of current transient, te is the extraction time, te = t1/2(LQg >> CU) - t1/2(LQg << CU).
AC C
t1/2 is the time at which current transient reduces to the half of its maximum value, t1/2(LQg >> CU) is the time t1/2 at a high light intensity L where photogenerated charge Qg is much higher than the charge stored on the contacts CU and t1/2(LQg << CU) is the t1/2 at low L where photogenerated charge Qg is much lower than the charge stored on the contacts CU. The condition of t1/2 (LQg
>> CU)
is obtained when the input light
intensity L is high enough to saturate the ToF current transient, creating a reservoir of
ACCEPTED MANUSCRIPT photogenerated charges. If the dominant recombination is of pure Langevin type, the photogenerated charge Qg quickly recombines and the extracted charge Qe remains equal to the charge stored on the contacts CU, since t1/2(LQg >> CU) ≈ t1/2(LQg << CU). If the recombination is strongly reduced, both t1/2(LQg >> CU) and the extracted charge Qe
RI PT
increases with light intensity L until it ultimately saturates. This results in the condition t1/2(LQg >> CU) > t1/2(LQg << CU); therefore, the extracted charge Qe becomes
SC
higher than CU. 3. Results and discussion
M AN U
In order to evaluate Langevin prefactor ξ, TQ1:PC61BM:PC71BM sample was measured with an ordinary photo-CELIV technique using 2 V amplitude and 10 µs duration triangle voltage pulse. In Fig. 2a depicted photo-CELIV current transients at different laser intensity L. With increasing L due to the fast recombination, the
TE D
extraction current ∆j saturates to the displacement current j(0) (Fig. 2b). It is an
AC C
EP
explicit evidence of non-reduced Lagevin recombination (Eq. 9).
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 2. (a) TQ1:PC61BM:PC71BM 5:8:2 20 mg/ml sample’s photo-CELIV transients for the A = 2×105 V/s, tdel = 3.5 µs at a different light intensity L (arbitrary units) measured with an ordinary setup. (b) The comparison between experimentally measured (squares) and theoretically calculated (line) [26] dependence of ∆j/j(0) ratio on light intensity L, where ∆j and j(0) are the densities of extraction and displacement current respectively. Integral ToF measurements were performed additionally to assure that the recombination is of non-reduced Langevin type. The sample was excited at different
ACCEPTED MANUSCRIPT light intensities L; the time t1/2 and the extracted charge Qe were evaluated from the current transient. From Fig. 3. Extracted charge Qe (black squares) and extraction half-time t1/2 (blue circles) dependence on light intensity measured with integral mode ToF technique in TQ1:PC61BM:PC71BM 5:8:2 20 mg/ml sample for the E = 2.5·105
RI PT
V/cm, R = 1 MΩ. it is obvious that t1/2, the time when ToF transient decreases twice, slightly depends on light intensity L. With increasing L, t1/2 increases only slightly and the extracted charge Qe saturates to CU. It shows that due to fast non-reduced
EP
TE D
M AN U
SC
Langevin recombination, no charge carrier reservoir is created [14,28,29].
AC C
Fig. 3. Extracted charge Qe (black squares) and extraction half-time t1/2 (blue circles) dependence on light intensity measured with integral mode ToF technique in TQ1:PC61BM:PC71BM 5:8:2 20 mg/ml sample for the E = 2.5·105 V/cm, R = 1 MΩ. We have also employed photo-CELIV technique to analyse the charge carrier
density decay which gives valuable insights about the order and the rate of recombination. During photo-CELIV delay time tdel it is common to use offset voltage Uoffset for the investigation of recombination processes. We can see from Fig.
ACCEPTED MANUSCRIPT 4 that the employment of an offset voltage Uoffset increases the lifetime of charge carriers. However, during the delay time even applying a large Uoffset (-0.9 V) we observed the extraction of photogenerated charge carriers caused by a not fully compensated internal electric field. It was impossible to use a larger Uoffset due to the
RI PT
commenced injection of charge carriers. Thus, when Uoffset = -0.9 V, the charge carrier density n decreased due to the recombination and extraction. In order to avoid
TE D
M AN U
SC
this extraction, the voltage switcher was used (Fig. 4, black squares).
AC C
EP
Fig. 4. Extracted charge carrier density dependences on delay time between the laser pulse and voltage ramp measured in TQ1:PC61BM:PC71BM 5:8:2 30 mg/ml sample using different photo-CELIV setups: ordinary setup (green triangles), ordinary setup with an offset voltage (red stars), setup with a voltage switcher (black squares). CELIV pulse with a voltage ramp of 200 V/s was used. Employing ordinary setup photo-CELIV and Eq. (8), the mobility dependence on
delay time µ(tdel) was evaluated (Fig. 5, black circles) and it has a slope of 0.14. It is clearly seen that the mobility is almost time-independent and thus trapping of charge carriers is obscure. Using Eq. (5), Eq. (3) and time-independent mobility, we calculated the decay of carrier density n(t) for the non-reduced (ξ = 1) second-order
ACCEPTED MANUSCRIPT Langevin recombination (Fig. 5, black line). Taking µ(tdel) dependence into account, it gives the recombination order of 2.14 (Fig. 5, blue line). Thus, µ(tdel) slightly increases the recombination order compared to the time (or carrier density) independent recombination. However, we can see that the recombination is much
RI PT
slower than Langevin using III photo-CELIV setup (Fig. 5, grey squares). We fitted the carrier density decay with k/kL = 0.0025 second-order recombination employing Eq. (3) and Eq. (5) (Fig. 5, red line). Strongly reduced Langevin recombination
SC
contradicts the previous photo-CELIV (Fig. 2) and ToF (Fig. 3) results. The reduced Langevin type recombination was explained by the separation of the electron and hole
M AN U
pathways (i) [30,31], the limitation of the minimum mobility (ii) [32], the extent of charge delocalization of opposite charges in a charge transfer (CT) state (iii) [3] or the discrepancy between the spatially dependent non-geminate recombination rate, proportional to the local product of electron and hole densities and the recombination
TE D
rate based on the average concentrations of charge carrier (iv) [8]. The (i), (ii) and (iii) explanations are related to the morphology and are not able to define the nonreduced Langevin recombination observed in the results of photo-CELIV (Fig. 2) and
EP
ToF (Fig. 3). Thus, the reduction of the recombination rate compared with Langevin
AC C
could be explained by spatially separated holes and electrons during the photoCELIV delay time tdel. Consequently, carrier density decay n(t) is strongly reduced.
SC
RI PT
ACCEPTED MANUSCRIPT
TE D
M AN U
Fig. 5. Experimentally measured mobility (black circles) of TQ1:PC61BM:PC71BM 5:8:2 20 mg/ml sample using photo-CELIV setup with Uoffset = 0 V, Α = 2×105 V/s; extracted charge density dependence on delay time using photo-CELIV technique with a voltage switcher (grey squares) with Α = 400 V/s and calculated charge density decay using measured time-dependent mobility in case of non-reduced (blue line) and reduced with k/kL = 0.0025 (red line) Langevin recombination. Additionally, the charge carrier decay dependence on time was calculated using time-independent mobility for non-reduced Langevin (black line) and the third-order recombination with the coefficient of k = 2×10-30 cm6/s (green line). While in a time range of µs-ms reduced Langevin recombination can explain an experimentally observed decay of charge carrier, it fails to define the third-order
EP
decay in ms timescale. It was shown that in systems with a partial phase separation,
AC C
trapped charge carriers can be protected from the recombination. Therefore, it increases the order of recombination [19]. As it was mentioned before, indicated pure Langevin recombination denies the morphological phase separation. However, due to the spatial separation, trapped electrons in an electron-rich region (near the cathode) and trapped holes in a hole-rich region (near the anode) are protected from the recombination and it could lead to the increased third-order recombination.
ACCEPTED MANUSCRIPT 4. Conclusions The shape of standard setup photo-CELIV and the results of ToF experiments confirm the non-reduced Langevin recombination of the second-order in
RI PT
TQ1:PC61BM:PC71BM blends. The recombination processes were also investigated by photo-CELIV technique from charge carrier density decay. However, in case of the concentrated electric field at the contacts, the density of photogenerated charge
SC
carrier decreases due to both the recombination and the extraction by the internal electric field in standard photo-CELIV setup even with applied offset voltage. A
M AN U
voltage switcher that disconnects sample’s contacts from the electrical circuit during delay time can also be used in order to avoid the charge carrier extraction. Employing photo-CELIV setup with a voltage switcher, the decay of the charge carrier density was measured, and it shows the non-geminate strongly reduced recombination of the
TE D
third-order. Due to the concentrated electric field, the electrons are concentrated at the cathode and the holes – at the anode. This leads to the spatially dependent recombination rate that is proportional to the local product of electron and hole
EP
concentrations and this recombination is strongly reduced compared to Langevin recombination. To sum up, in case of the concentrated electric field, photo-CELIV
AC C
fails to evaluate the recombination rate and order from the density decay of charge carriers.
Acknowledgements
This research was funded by a grant MIP-091/2015 from the Research Council of Lithuania.
ACCEPTED MANUSCRIPT References [1]
[2]
RI PT
[3]
M. A. Green, K. Emery, Y. Hishikawa, W. Warta, E. D. Dunlop, D. H. Levi, and A. W. Y. HoBaillie, Solar cell efficiency tables (version 49), Prog. Photovoltaics Res. Appl., 25 (2016), 3–13, http://dx.doi.org/10.1002/pip.2855. R. Österbacka, A. Pivrikas, G. Juška, K. Genevičius, K.Arlauskas, and H. Stubb, Measuring the time relaxation of photogenerated charge carriers in organic materials, Current Appl. Phys. 415 (2004) 534–538, http://dx.doi.org/10.1016/j.cap.2004.01.013. D. H. K. Murthy, A. Melianas, Z. Tang, G. Juška, K. Arlauskas, F. Zhang, L. D. A. Siebbeles, O. Inganäs, T. J. Savenije, Origin of Reduced Bimolecular Recombination in Blends of Conjugated Polymers and Fullerenes, Adv. Funct. Mater. 23 (2013) 4262–4268, http://dx.doi.org/10.1002/adfm.201203852. C. Deibel, A. Baumann, and V. Dyakonov, Polaron recombination in pristine and annealed bulk heterojunction solar cells, Appl. Phys. Lett. 93 (2008) 163303, http://dx.doi.org/10.1063/1.3005593. A. J. Mozer, G. Dennler, N. S. Sariciftci, M. Westerling, A. Pivrikas, R. Österbacka, G. Juška, Time-dependent mobility and recombination of the photoinduced charge carriers in conjugated polymer/fullerene bulk heterojunction solar cells, Phys. Rev. B 72 (2005) 035217, http://dx.doi.org/10.1103/PhysRevB.72.035217. N. Nekrašas, K. Genevičius, M. Viliūnas, G. Juška, Features of current transients of photogenerated charge carriers, extracted by linearly increased voltage, Chem. Phys. 404 (2012) 56–59, http://dx.doi.org/10.1016/j.chemphys.2012.01.008. P. Langevin, Ann. Chim. Phys. 28 (1903) 433. C. Deibel, A. Wagenpfahl, and V. Dyakonov, Origin of reduced polaron recombination in organic semiconductor devices, Phys. Rev. B 80 (2009) 07520, http://dx.doi.org/10.1103/PhysRevB.80.075203. J.-M. Nunzi, G. Juška, F. Jean and K.Arlauskas, Slowup of Bimolecular Recombination in Organic Polymer Solar Cells. Acta Phys. Pol. 107 (2) (2005) 377–380, http://dx.doi.org/10.12693/APhysPolA.107.377. B. Philippa, M. Stolterfoht, R. D. White, M. Velusamy, P. L. Burn, P. Meredith, A. Pivrikas, Molecular weight dependent bimolecular recombination in organic solar cells, J. Chem. Phys. 141 (2014) 054903, http://dx.doi.org/10.1063/1.4891369. G. Juška, K. Arlauskas, G. Šliaužys, A. Pivrikas, A. J. Mozer, N. S. Sariciftci, M. Scharber, R. Österbacka, Double injection as a technique to study charge carrier transport and recombination in bulkheterojunction solar-cells, Appl. Phys. Lett. 87 (2005) 222110, http://dx.doi.org/10.1063/1.2137454. S. Roland, M. Schubert, B. A. Collins, J. Kurpiers, Z. Chen, A. Facchetti, H. Ade, and Dieter Neher, Fullerene-Free Polymer Solar Cells with Highly Reduced Bimolecular Recombination and Field-Independent Charge Carrier Generation, J. Phys. Chem. Lett. 5 (2014) 2815–2822, http://doi.org/10.1021/jz501506z. P. Zalar, M. Kuik, N. A. Ran, J. A. Love, and T.-Q. Nguyen, Effects of Processing Conditions on the Recombination Reduction in Small Molecule Bulk Heterojunction Solar Cells, Adv. Energy Mater. 4 (2014) 1400438, http://doi.org/10.1002/aenm.201400438. N. Nekrašas, G. Šliaužys, G. Juška, K. Arlauskas, J. Stuchlik, J. Kočka, Ultrafast Bimolecular Recombination in Nanocrystalline Hydrogenated Silicon, Acta Phys. Pol. A 107 (2005) 373–376, http://dx.doi.org/10.12693/APhysPolA.107.373. G. Juška, K. Genevičius, N. Nekrašas, G. Sliaužys, and G. Dennler, Trimolecular recombination in polythiophene: fullerene bulk heterojunction solar cells, Appl. Phys. Lett. 93 (2008) 143303, http://dx.doi.org/10.1063/1.2996588. C. G. Shuttle, B. O’Regan, A. M. Ballantyne, J. Nelson, D. D. C. Bradley, J. de Mello, and J. R. Durrant, Experimental determination of the rate law for charge carrier decay in a polythiophene:Fullerene solar cell, Appl. Phys. Lett. 92 (2008) 093311, http://doi.org/10.1063/1.2891871. C. Deibel, A. Wagenpfahl, V. Dyakonov, phys. stat. sol. (RRL) 2 (4) (2008) 175–177, http://dx.doi.org/10.1002/pssr.200802110.
[4]
[9]
[10]
[11]
[12]
AC C
[13]
TE D
[7] [8]
EP
[6]
M AN U
SC
[5]
[14]
[15]
[16]
[17]
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[18] D. Spoltore, W. D. Oosterbaan, S. Khelifi, J. N. Clifford, A. Viterisi, E. Palomares, M. Burgelman, L. Lutsen, D. Vanderzande, J. Manca, Effect of Polymer Crystallinity in P3HT:PCBM Solar Cells on Band Gap Trap States and Apparent Recombination Order, Adv. Energy Mater. 3 (4) (2012) 466–471, http://dx.doi.org/10.1002/aenm.201200674. [19] D. Rauh, C. Deibel, V. Dyakonov, Charge Density Dependent Nongeminate Recombination in Organic Bulk Heterojunction Solar Cells, Adv. Funct. Mater. 22 (2012) 3371–3377, http://dx.doi.org/10.1002/adfm.201103118. [20] J. Gorenflot, M. C. Heiber, A. Baumann, J. Lorrmann, M. Gunz, A. Kämpgen, V. Dyakonov, C. Deibel, Nongeminate recombination in neat P3HT and P3HT:PCBM blend films, J. Appl. Phys. 115 (2014) 144502, http://dx.doi.org/10.1063/1.4870805. [21] C. G. Shuttle, B. O’Regan, A. M. Ballantyne, J. Nelson, D. D. C. Bradley, and J. R. Durrant, Bimolecular recombination losses in polythiophene: Fullerene solar cells. Phys. Rev. B 78 (2008) 113201, http://dx.doi.org/10.1103/PhysRevB.78.113201. [22] C. G. Shuttle, R. Hamilton, J. Nelson, B. C. O’Regan, and J. R. Durrant, Measurement of ChargeDensity Dependence of Carrier Mobility in an Organic Semiconductor Blend. Adv. Funct. Mater. 20 (2010) 698–702, http://dx.doi.org/10.1002/adfm.200901734. [23] R. C. I. MacKenzie, T. Kirchartz, G. F. A. Dibb, J. Nelson, J. Phys. Chem. C 115 (19) (2011) 9806–9813, http://dx.doi.org/10.1021/jp200234m. [24] A. V. Nenashev, M. Wiemer, A. V. Dvurechenskii, F. Gebhard, M. Koch, S. D. Baranovskii, Why the apparent order of bimolecular recombination in blend organic solar cells can be larger than two: A topological consideration, Appl. Phys. Lett. 109 (2016) 033301, http://dx.doi.org/10.1063/1.4959076. [25] J. Szmytkowski, Electrostatic interface recombination in the system of disordered materials characterized by different permittivities, Synth. Met. 206 (2015) 120–123, http://dx.doi.org/10.1016/j.synthmet.2015.05.001. [26] G. Juška, N. Nekrašas, V. Valentinavičius, P. Meredith, and A. Pivrikas, Extraction of photogenerated charge carriers by linearly increasing voltage in the case of Langevin recombination, Phys. Rev. B 84 (2011) 155202, http://dx.doi.org/10.1103/PhysRevB.84.155202. [27] A. Pivrikas, G. Juška, R. Österbacka, M. Westerling, M. Viliūnas, K. Arlauskas, and H. Stubb, Langevin recombination and space-charge-perturbed current transients in regiorandom poly(3hexylthiophene), Phys. Rev. B 71 (2005) 125205, http://dx.doi.org/10.1103/PhysRevB.71.125205. [28] G. Juška, M. Viliūnas, and K. Arlauskas, Space-charge-limited photocurrent transients: The influence of bimolecular recombination, Phys. Rev. B 51 (1995) 16668, http://dx.doi.org/10.1103/PhysRevB.51.16668. [29] G. Juška, J. Kočka, M. Viliūnas, and K. Arlauskas, Subnanosecond bimolecular non-radiative recombination in a-Si:H, J. Non-Cryst. Solids 164–166 (1993) 579–582, http://dx.doi.org/10.1016/0022-3093(93)90618-8. [30] I. Balberg, R. Naidis, M. K. Lee, J. Shinar, L. F. Fonseca, Bipolar phototransport in p-conjugated polymer/C60 composites, Appl. Phys. Lett. 79 (2001) 197–199, http://dx.doi.org/10.1063/1.1383801. [31] G. J. Adriaenssens, V. I. Arkhipov, Non-Langevin recombination in disordered materials with random potential distributions, Solid State Commun. 103 (1997) 541–543, http://dx.doi.org/10.1016/S0038-1098(97)00233-0. [32] L. J. A. Koster, V. D. Mihailetchi, and P. W. M. Blom, Bimolecular recombination in polymer/fullerene bulk heterojunction solar cells, Appl. Phys. Lett. 88 (2006) 052104, http://dx.doi.org/10.1063/1.2170424.
ACCEPTED MANUSCRIPT • •
AC C
EP
TE D
M AN U
SC
RI PT
•
The non-geminate recombination was investigated in TQ1:PC61BM:PC71BM bulk heterojunctions employing photo-CELIV and integral ToF techniques. The shapes of photo-CELIV and ToF transients point out purely non-reduced Langevin recombination while the decay of the photogenerated charge carrier density measured with photoCELIV technique indicates reduced third-order recombination. In case of the concentrated electric field, photo-CELIV fails to evaluate the recombination rate and order from the density decay of charge carriers.
S1566-1199(17)30188-X
DOI:
10.1016/j.orgel.2017.04.030
Reference:
ORGELE 4068
To appear in:
Organic Electronics
Received Date: 17 March 2017 Accepted Date: 27 April 2017
Please cite this article as: J. Važgėla, A. Galvelytė, G. Juška, Mismatch of non-geminate recombination in polymer TQ1/fullerene blends employing different techniques, Organic Electronics (2017), doi: 10.1016/j.orgel.2017.04.030. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Mismatch of Non-Geminate Recombination in Polymer TQ1/Fullerene Blends Employing Different Techniques J. Važgėla1,a), A. Galvelytė1, and G. Juška1 Department of Solid State Electronics, Vilnius University, Saulėtekio 9 III k., 10222 Vilnius, Lithuania
RI PT
1
The non-geminate recombination was investigated in TQ1:PC61BM:PC71BM (poly[[2,3-bis(3-
SC
octyloxyphenyl)-5,8-quinoxalinediyl]-2,5-thiophenediyl]:[6,6]-Phenyl-C61-butyric-acid-methylester:[6,6]-Phenyl-C71-butyric-acid-methyl-ester) bulk heterojunctions employing the extraction of the photogenerated charge carriers by linearly increasing voltage (photo-CELIV) and integral
M AN U
Time-of-Flight (ToF) techniques. The shapes of photo-CELIV and ToF transients point out purely non-reduced Langevin recombination while the decay of the photogenerated charge carrier density measured with photo-CELIV technique indicates reduced third-order recombination. A clear discrepancy could be explained by spatially separated holes and electrons during the delay time while employing photo-CELIV method.
EP
Electronic mail: [email protected].
AC C
a)
TE D
_____________________________
1
ACCEPTED MANUSCRIPT 1. Introduction The power conversion efficiency of single junction organic solar cells (OSCs) and modules have reached 11.2 % and 8.7 % respectively [1]. In order to reach higher
RI PT
efficiencies, it is necessary to enhance the extraction of charge carriers and thus to reduce the recombination. Photo-induced charge carrier extraction by linearly increasing voltage (photo-CELIV) technique has been widely used for the studies of
SC
charge carrier transport properties [2–6]. Photo-CELIV allows to evaluate the relaxation of the photogenerated charge carrier mobility and the recombination of
M AN U
charge carriers [2,5,6]. The recombination rate can be evaluated both from the decay of charge carrier density [3–5] and the shape of current transient [6]. Photo-CELIV results give the information about the recombination rate and order. The concentration decay of electrons in solar cell can be described as:
TE D
dn 1 d jn =− + G − R, dt e dx
(1)
where n is the concentration of electrons, t – time, e – elementary charge, jn – the
EP
current of electrons, G – generation rate, R – recombination rate. In photo-CELIV experiments photogenerated charge carriers are extracted after certain delay time tdel
AC C
(G = 0). In order to avoid the extraction of charge carriers during the delay time tdel, an offset voltage Uoffset (approximately equal to the open circuit voltage Uoc) is applied to the sample. In this case the first member on the right side of Eq. (1) is almost equal to zero. Taking this into account Eq. (1) can be simplified to:
dn = − R = −k0nλ +1, dt
(2)
where k0 is the recombination coeffiecient independent of charge carrier density, λ+1 – the apparent recombination order.
ACCEPTED MANUSCRIPT The integration of Eq. (2) gives the charge carrier density dependence on time:
(
n(t ) = n(0)−λ + λk0t
)
−1 / λ
(3)
.
In disordered materials the recombination is commonly considered as a second-
RI PT
order Langevin type recombination that is limited by the probability for the electrons and holes to meet in space [7]:
R = kL n2 ,
(4)
SC
where kL is Langevin recombination coefficient, kL = e(µn+µp)/εε0, µn (µp) – electron (hole) mobility, ε (ε0) – relative (absolute) dielectric permittivity. However, it
M AN U
was shown that in some polymer blends the recombination is reduced compared with the one predicted by Langevin [3,4,8–14]. In such case the coefficient of the secondorder recombination can be described as:
k = ξkL = ξ
e(µn + µp )
εε0
,
(5)
TE D
where ξ is the Langevin prefactor, ξ = 1 in a pure Langevin recombination case and ξ < 1 if the recombination is reduced.
EP
Nonetheless, experimental results [15–22] and theoretical simulations [23–25] indicate the recombination with the order higher than 2. One of the explanations is
AC C
related to the dependence of non-geminate recombination rate along with the mobility on charge carrier density. This dependence has been assigned to the presence of an energetic tail of localized trap sites [21,22]:
R = k (n)n2 = k0n2+δ ,
(6)
µ (n) = µ0nδ ,
(7)
ACCEPTED MANUSCRIPT where µ0 is charge carrier density-independent mobility, δ – constant, δ > 0. In disordered organic materials charge transport is explained by charge carrier hopping between localized states of the density of states. With increasing disorder the mobility becomes more dependent on charge carrier density (constant δ increases). It was
RI PT
shown that in P3HT:PC61BM blend the recombination order is 2 in case P3HT is completely crystalline (a single crystal) and with the increasing fraction of amorphous phase the concentration of traps and trapped charges increases along with the
SC
recombination order, connected to a trap-assisted recombination [18].
Another explanation of the recombination order being higher than 2 in bulk
M AN U
heterojunction is related to donor–acceptor phase separation [19]. Due to the phase separation trapped charge carrier cannot recombine with the different type of charge carrier. The recombination occurs only when trapped charge carrier becomes mobile and meets a mobile charge carrier which is oppositely charged. To sum up, in the
TE D
presence of charge carrier trapping, the recombination order can be higher than 2 due to the mobility dependence on the charge carrier density and the separation of donor– acceptor phase.
EP
In this paper we have employed photo-CELIV and integral time-of-flight (ToF)
AC C
techniques to investigate the recombination processes in TQ1:PC61BM:PC71BM solar cells. While the shape of photo-CELIV and ToF results indicate a non-reduced Langevin recombination, charge carrier decay evaluated with photo-CELIV technique shows strongly reduced recombination processes. We have shown that in case of concentrated internal electric field, the decay of carrier density measured using photoCELIV technique is governed by strongly reduced recombination with the order
ACCEPTED MANUSCRIPT higher than 2+δ and is extremely approximate. In such a case, photo-CELIV method is not suitable for measuring the charge carrier decay. 2. Materials and methods
RI PT
The samples used for the following experiments were prepared as follows: the glass substrates covered with ITO were cleaned and dried out, then a thin layer of PEDOT:PSS was spin-coated at 3000 RPM for 40 s and heated at 120 ℃ for 15 min. (TQ1),
SC
Poly[[2,3-bis(3-octyloxyphenyl)-5,8-quinoxalinediyl]-2,5-thiophenediyl]
[6,6]-Phenyl-C61-butyric-acid-methyl-ester (PC61BM) and [6,6]-Phenyl-C71-butyric-
M AN U
acid-methyl-ester (PC71BM) with a mass ratio of 5:8:2 were dissolved in o-xylene (20 and 30 mg/ml) and spin-coated (the thickness was 80 nm and 190 nm, respectively) on top of ITO/PEDOT:PSS layers. Afterwards, the layers of LiF (0.6 nm) and Al (90 nm) were evaporated under the high vacuum conditions (5·10-4 Pa). Finally, the
TE D
devices were encapsulated. The active area of the device was 0.048 cm2. Photo-CELIV measurements were performed in three different setups: ordinary (I), ordinary setup with an offset voltage (II) and setup with a voltage switcher (III,
EP
Fig. 1. The schematic view of photo-CELIV experimental setup with a voltage switcher for the investigation of charge carrier recombination.). The measurements of the all setups
AC C
were conducted with a pulsed second-harmonic 532 nm Nd:YAG laser (EKSPLA PL2143), oscilloscope (Tektronix DPO4054B) and generator (Tektronix AFG3022B). The additional generator (Tektronix AFG3022B) and a controllable voltage switcher were used for the III setup. The laser pulse duration was 30 ps and the beam diameter 6 mm. At 532 nm the absorption coefficients of the samples were α = 4.1·104 cm-1 and α = 1.03·105 cm-1, giving αd = 0.33 and αd = 0.82 for 20 mg/ml and 30 mg/ml
ACCEPTED MANUSCRIPT sample respectively. The sample was excited through Al contact side. The laser was synchronized with triangle (Generator II) and square (Generator I, only III setup) voltage pulse generators. In order to avoid the extraction by inserted voltage during the delay time tdel, an offset voltage Uoffset was used in II setup. During the delay time
RI PT
tdel in III setup the extraction of photogenerated charge carriers was avoided by employing a voltage switcher. The sample’s contacts were disconnected from the electrical circuit and after a certain delay time tdel Generator I applied square voltage
SC
pulse to the controllable voltage switcher which connected the contacts to the circuit (III setup). Photogenerated charge carriers were extracted after the same delay time
M AN U
tdel by the application of a linearly increasing voltage.
In case of bulk photogeneration the mobility of charge carriers can be found as [5]:
2d 2 , 2 [1 + 0.36∆j / j(0)] 3 Atmax
TE D
µ=
(8)
where d is the thickness of the active layer, A – the voltage ramp, A = dU/dt, tmax – the time when photo-CELIV transient reaches its maximum (Fig. 2), j(0) – the
AC C
EP
displacement current density, ∆j – the extraction current density.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 1. The schematic view of photo-CELIV experimental setup with a voltage switcher for the investigation of charge carrier recombination. In a time window of 0 ≤ t ≤ tp, where tp is CELIV pulse duration, photo-CELIV transient in a sample consists of a displacement j(0) and extraction ∆j current
TE D
densities (Fig. 2). Applying long enough CELIV pulse, ∆j = 0 at the end of the pulse. Thus, by deducting a dark transient from the illuminated one and integrating the result of it, the extracted charge Qe can be estimated. By changing delay time tdel between
EP
the laser and CELIV pulses, tmax increases (charge carrier mobility decreases, Eq. (8))
AC C
and the extracted charge Qe decreases due to the recombination. Hence, the mobility and carrier density dependences on time (µ(t) and n(t) respectively), where t is the delay time tdel, can be measured [2]. Using the measured mobility dependence on time µ(tdel), Eq. (5), Eq. (3) and the recombination order of 2 (λ = 1), the decay n(tdel) of charge carrier density can be calculated employing the different values of Langevin prefactor ξ. These theoretically calculated n(tdel) trends are compared with the
ACCEPTED MANUSCRIPT measured n(tdel) decay. Langevin prefactor ξ can be found from the theoretically calculated n(tdel) curve that matches experimental n(tdel) results the best. Using photo-CELIV technique Langevin prefactor ξ can be found in a different way. If the recombination is of Langevin type, the recombination is fast and the
RI PT
extraction current density ∆j does not exceed the displacement current density j(0) (Fig. 2) even at the saturation light intensity [26]. If the recombination is reduced (ξ <
intensity Langevin prefactor ξ can be evaluated [6]:
β j(0) = βL ∆j
(9)
M AN U
ξ=
SC
1 in Eq. (5)), ∆j exceeds j(0) and saturates with L. From j(0)/∆j ratio at a high light
Time-of-flight (ToF) experiments were made in integral mode ToF (τRC >> ttr, where τRC – is the time constant of the setup, ttr – charge carrier transit time, ttr = d2/µU, U – applied voltage, d – the thickness of the active layer). Using this
TE D
technique, the coefficient ξ could be estimated as described in Ref. 27:
ξ=
β CU ttr , = βL Qe te
(10)
where C is the capacitance of the sample, Qe – the extracted charge that is equal to
EP
integral of current transient, te is the extraction time, te = t1/2(LQg >> CU) - t1/2(LQg << CU).
AC C
t1/2 is the time at which current transient reduces to the half of its maximum value, t1/2(LQg >> CU) is the time t1/2 at a high light intensity L where photogenerated charge Qg is much higher than the charge stored on the contacts CU and t1/2(LQg << CU) is the t1/2 at low L where photogenerated charge Qg is much lower than the charge stored on the contacts CU. The condition of t1/2 (LQg
>> CU)
is obtained when the input light
intensity L is high enough to saturate the ToF current transient, creating a reservoir of
ACCEPTED MANUSCRIPT photogenerated charges. If the dominant recombination is of pure Langevin type, the photogenerated charge Qg quickly recombines and the extracted charge Qe remains equal to the charge stored on the contacts CU, since t1/2(LQg >> CU) ≈ t1/2(LQg << CU). If the recombination is strongly reduced, both t1/2(LQg >> CU) and the extracted charge Qe
RI PT
increases with light intensity L until it ultimately saturates. This results in the condition t1/2(LQg >> CU) > t1/2(LQg << CU); therefore, the extracted charge Qe becomes
SC
higher than CU. 3. Results and discussion
M AN U
In order to evaluate Langevin prefactor ξ, TQ1:PC61BM:PC71BM sample was measured with an ordinary photo-CELIV technique using 2 V amplitude and 10 µs duration triangle voltage pulse. In Fig. 2a depicted photo-CELIV current transients at different laser intensity L. With increasing L due to the fast recombination, the
TE D
extraction current ∆j saturates to the displacement current j(0) (Fig. 2b). It is an
AC C
EP
explicit evidence of non-reduced Lagevin recombination (Eq. 9).
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 2. (a) TQ1:PC61BM:PC71BM 5:8:2 20 mg/ml sample’s photo-CELIV transients for the A = 2×105 V/s, tdel = 3.5 µs at a different light intensity L (arbitrary units) measured with an ordinary setup. (b) The comparison between experimentally measured (squares) and theoretically calculated (line) [26] dependence of ∆j/j(0) ratio on light intensity L, where ∆j and j(0) are the densities of extraction and displacement current respectively. Integral ToF measurements were performed additionally to assure that the recombination is of non-reduced Langevin type. The sample was excited at different
ACCEPTED MANUSCRIPT light intensities L; the time t1/2 and the extracted charge Qe were evaluated from the current transient. From Fig. 3. Extracted charge Qe (black squares) and extraction half-time t1/2 (blue circles) dependence on light intensity measured with integral mode ToF technique in TQ1:PC61BM:PC71BM 5:8:2 20 mg/ml sample for the E = 2.5·105
RI PT
V/cm, R = 1 MΩ. it is obvious that t1/2, the time when ToF transient decreases twice, slightly depends on light intensity L. With increasing L, t1/2 increases only slightly and the extracted charge Qe saturates to CU. It shows that due to fast non-reduced
EP
TE D
M AN U
SC
Langevin recombination, no charge carrier reservoir is created [14,28,29].
AC C
Fig. 3. Extracted charge Qe (black squares) and extraction half-time t1/2 (blue circles) dependence on light intensity measured with integral mode ToF technique in TQ1:PC61BM:PC71BM 5:8:2 20 mg/ml sample for the E = 2.5·105 V/cm, R = 1 MΩ. We have also employed photo-CELIV technique to analyse the charge carrier
density decay which gives valuable insights about the order and the rate of recombination. During photo-CELIV delay time tdel it is common to use offset voltage Uoffset for the investigation of recombination processes. We can see from Fig.
ACCEPTED MANUSCRIPT 4 that the employment of an offset voltage Uoffset increases the lifetime of charge carriers. However, during the delay time even applying a large Uoffset (-0.9 V) we observed the extraction of photogenerated charge carriers caused by a not fully compensated internal electric field. It was impossible to use a larger Uoffset due to the
RI PT
commenced injection of charge carriers. Thus, when Uoffset = -0.9 V, the charge carrier density n decreased due to the recombination and extraction. In order to avoid
TE D
M AN U
SC
this extraction, the voltage switcher was used (Fig. 4, black squares).
AC C
EP
Fig. 4. Extracted charge carrier density dependences on delay time between the laser pulse and voltage ramp measured in TQ1:PC61BM:PC71BM 5:8:2 30 mg/ml sample using different photo-CELIV setups: ordinary setup (green triangles), ordinary setup with an offset voltage (red stars), setup with a voltage switcher (black squares). CELIV pulse with a voltage ramp of 200 V/s was used. Employing ordinary setup photo-CELIV and Eq. (8), the mobility dependence on
delay time µ(tdel) was evaluated (Fig. 5, black circles) and it has a slope of 0.14. It is clearly seen that the mobility is almost time-independent and thus trapping of charge carriers is obscure. Using Eq. (5), Eq. (3) and time-independent mobility, we calculated the decay of carrier density n(t) for the non-reduced (ξ = 1) second-order
ACCEPTED MANUSCRIPT Langevin recombination (Fig. 5, black line). Taking µ(tdel) dependence into account, it gives the recombination order of 2.14 (Fig. 5, blue line). Thus, µ(tdel) slightly increases the recombination order compared to the time (or carrier density) independent recombination. However, we can see that the recombination is much
RI PT
slower than Langevin using III photo-CELIV setup (Fig. 5, grey squares). We fitted the carrier density decay with k/kL = 0.0025 second-order recombination employing Eq. (3) and Eq. (5) (Fig. 5, red line). Strongly reduced Langevin recombination
SC
contradicts the previous photo-CELIV (Fig. 2) and ToF (Fig. 3) results. The reduced Langevin type recombination was explained by the separation of the electron and hole
M AN U
pathways (i) [30,31], the limitation of the minimum mobility (ii) [32], the extent of charge delocalization of opposite charges in a charge transfer (CT) state (iii) [3] or the discrepancy between the spatially dependent non-geminate recombination rate, proportional to the local product of electron and hole densities and the recombination
TE D
rate based on the average concentrations of charge carrier (iv) [8]. The (i), (ii) and (iii) explanations are related to the morphology and are not able to define the nonreduced Langevin recombination observed in the results of photo-CELIV (Fig. 2) and
EP
ToF (Fig. 3). Thus, the reduction of the recombination rate compared with Langevin
AC C
could be explained by spatially separated holes and electrons during the photoCELIV delay time tdel. Consequently, carrier density decay n(t) is strongly reduced.
SC
RI PT
ACCEPTED MANUSCRIPT
TE D
M AN U
Fig. 5. Experimentally measured mobility (black circles) of TQ1:PC61BM:PC71BM 5:8:2 20 mg/ml sample using photo-CELIV setup with Uoffset = 0 V, Α = 2×105 V/s; extracted charge density dependence on delay time using photo-CELIV technique with a voltage switcher (grey squares) with Α = 400 V/s and calculated charge density decay using measured time-dependent mobility in case of non-reduced (blue line) and reduced with k/kL = 0.0025 (red line) Langevin recombination. Additionally, the charge carrier decay dependence on time was calculated using time-independent mobility for non-reduced Langevin (black line) and the third-order recombination with the coefficient of k = 2×10-30 cm6/s (green line). While in a time range of µs-ms reduced Langevin recombination can explain an experimentally observed decay of charge carrier, it fails to define the third-order
EP
decay in ms timescale. It was shown that in systems with a partial phase separation,
AC C
trapped charge carriers can be protected from the recombination. Therefore, it increases the order of recombination [19]. As it was mentioned before, indicated pure Langevin recombination denies the morphological phase separation. However, due to the spatial separation, trapped electrons in an electron-rich region (near the cathode) and trapped holes in a hole-rich region (near the anode) are protected from the recombination and it could lead to the increased third-order recombination.
ACCEPTED MANUSCRIPT 4. Conclusions The shape of standard setup photo-CELIV and the results of ToF experiments confirm the non-reduced Langevin recombination of the second-order in
RI PT
TQ1:PC61BM:PC71BM blends. The recombination processes were also investigated by photo-CELIV technique from charge carrier density decay. However, in case of the concentrated electric field at the contacts, the density of photogenerated charge
SC
carrier decreases due to both the recombination and the extraction by the internal electric field in standard photo-CELIV setup even with applied offset voltage. A
M AN U
voltage switcher that disconnects sample’s contacts from the electrical circuit during delay time can also be used in order to avoid the charge carrier extraction. Employing photo-CELIV setup with a voltage switcher, the decay of the charge carrier density was measured, and it shows the non-geminate strongly reduced recombination of the
TE D
third-order. Due to the concentrated electric field, the electrons are concentrated at the cathode and the holes – at the anode. This leads to the spatially dependent recombination rate that is proportional to the local product of electron and hole
EP
concentrations and this recombination is strongly reduced compared to Langevin recombination. To sum up, in case of the concentrated electric field, photo-CELIV
AC C
fails to evaluate the recombination rate and order from the density decay of charge carriers.
Acknowledgements
This research was funded by a grant MIP-091/2015 from the Research Council of Lithuania.
ACCEPTED MANUSCRIPT References [1]
[2]
RI PT
[3]
M. A. Green, K. Emery, Y. Hishikawa, W. Warta, E. D. Dunlop, D. H. Levi, and A. W. Y. HoBaillie, Solar cell efficiency tables (version 49), Prog. Photovoltaics Res. Appl., 25 (2016), 3–13, http://dx.doi.org/10.1002/pip.2855. R. Österbacka, A. Pivrikas, G. Juška, K. Genevičius, K.Arlauskas, and H. Stubb, Measuring the time relaxation of photogenerated charge carriers in organic materials, Current Appl. Phys. 415 (2004) 534–538, http://dx.doi.org/10.1016/j.cap.2004.01.013. D. H. K. Murthy, A. Melianas, Z. Tang, G. Juška, K. Arlauskas, F. Zhang, L. D. A. Siebbeles, O. Inganäs, T. J. Savenije, Origin of Reduced Bimolecular Recombination in Blends of Conjugated Polymers and Fullerenes, Adv. Funct. Mater. 23 (2013) 4262–4268, http://dx.doi.org/10.1002/adfm.201203852. C. Deibel, A. Baumann, and V. Dyakonov, Polaron recombination in pristine and annealed bulk heterojunction solar cells, Appl. Phys. Lett. 93 (2008) 163303, http://dx.doi.org/10.1063/1.3005593. A. J. Mozer, G. Dennler, N. S. Sariciftci, M. Westerling, A. Pivrikas, R. Österbacka, G. Juška, Time-dependent mobility and recombination of the photoinduced charge carriers in conjugated polymer/fullerene bulk heterojunction solar cells, Phys. Rev. B 72 (2005) 035217, http://dx.doi.org/10.1103/PhysRevB.72.035217. N. Nekrašas, K. Genevičius, M. Viliūnas, G. Juška, Features of current transients of photogenerated charge carriers, extracted by linearly increased voltage, Chem. Phys. 404 (2012) 56–59, http://dx.doi.org/10.1016/j.chemphys.2012.01.008. P. Langevin, Ann. Chim. Phys. 28 (1903) 433. C. Deibel, A. Wagenpfahl, and V. Dyakonov, Origin of reduced polaron recombination in organic semiconductor devices, Phys. Rev. B 80 (2009) 07520, http://dx.doi.org/10.1103/PhysRevB.80.075203. J.-M. Nunzi, G. Juška, F. Jean and K.Arlauskas, Slowup of Bimolecular Recombination in Organic Polymer Solar Cells. Acta Phys. Pol. 107 (2) (2005) 377–380, http://dx.doi.org/10.12693/APhysPolA.107.377. B. Philippa, M. Stolterfoht, R. D. White, M. Velusamy, P. L. Burn, P. Meredith, A. Pivrikas, Molecular weight dependent bimolecular recombination in organic solar cells, J. Chem. Phys. 141 (2014) 054903, http://dx.doi.org/10.1063/1.4891369. G. Juška, K. Arlauskas, G. Šliaužys, A. Pivrikas, A. J. Mozer, N. S. Sariciftci, M. Scharber, R. Österbacka, Double injection as a technique to study charge carrier transport and recombination in bulkheterojunction solar-cells, Appl. Phys. Lett. 87 (2005) 222110, http://dx.doi.org/10.1063/1.2137454. S. Roland, M. Schubert, B. A. Collins, J. Kurpiers, Z. Chen, A. Facchetti, H. Ade, and Dieter Neher, Fullerene-Free Polymer Solar Cells with Highly Reduced Bimolecular Recombination and Field-Independent Charge Carrier Generation, J. Phys. Chem. Lett. 5 (2014) 2815–2822, http://doi.org/10.1021/jz501506z. P. Zalar, M. Kuik, N. A. Ran, J. A. Love, and T.-Q. Nguyen, Effects of Processing Conditions on the Recombination Reduction in Small Molecule Bulk Heterojunction Solar Cells, Adv. Energy Mater. 4 (2014) 1400438, http://doi.org/10.1002/aenm.201400438. N. Nekrašas, G. Šliaužys, G. Juška, K. Arlauskas, J. Stuchlik, J. Kočka, Ultrafast Bimolecular Recombination in Nanocrystalline Hydrogenated Silicon, Acta Phys. Pol. A 107 (2005) 373–376, http://dx.doi.org/10.12693/APhysPolA.107.373. G. Juška, K. Genevičius, N. Nekrašas, G. Sliaužys, and G. Dennler, Trimolecular recombination in polythiophene: fullerene bulk heterojunction solar cells, Appl. Phys. Lett. 93 (2008) 143303, http://dx.doi.org/10.1063/1.2996588. C. G. Shuttle, B. O’Regan, A. M. Ballantyne, J. Nelson, D. D. C. Bradley, J. de Mello, and J. R. Durrant, Experimental determination of the rate law for charge carrier decay in a polythiophene:Fullerene solar cell, Appl. Phys. Lett. 92 (2008) 093311, http://doi.org/10.1063/1.2891871. C. Deibel, A. Wagenpfahl, V. Dyakonov, phys. stat. sol. (RRL) 2 (4) (2008) 175–177, http://dx.doi.org/10.1002/pssr.200802110.
[4]
[9]
[10]
[11]
[12]
AC C
[13]
TE D
[7] [8]
EP
[6]
M AN U
SC
[5]
[14]
[15]
[16]
[17]
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[18] D. Spoltore, W. D. Oosterbaan, S. Khelifi, J. N. Clifford, A. Viterisi, E. Palomares, M. Burgelman, L. Lutsen, D. Vanderzande, J. Manca, Effect of Polymer Crystallinity in P3HT:PCBM Solar Cells on Band Gap Trap States and Apparent Recombination Order, Adv. Energy Mater. 3 (4) (2012) 466–471, http://dx.doi.org/10.1002/aenm.201200674. [19] D. Rauh, C. Deibel, V. Dyakonov, Charge Density Dependent Nongeminate Recombination in Organic Bulk Heterojunction Solar Cells, Adv. Funct. Mater. 22 (2012) 3371–3377, http://dx.doi.org/10.1002/adfm.201103118. [20] J. Gorenflot, M. C. Heiber, A. Baumann, J. Lorrmann, M. Gunz, A. Kämpgen, V. Dyakonov, C. Deibel, Nongeminate recombination in neat P3HT and P3HT:PCBM blend films, J. Appl. Phys. 115 (2014) 144502, http://dx.doi.org/10.1063/1.4870805. [21] C. G. Shuttle, B. O’Regan, A. M. Ballantyne, J. Nelson, D. D. C. Bradley, and J. R. Durrant, Bimolecular recombination losses in polythiophene: Fullerene solar cells. Phys. Rev. B 78 (2008) 113201, http://dx.doi.org/10.1103/PhysRevB.78.113201. [22] C. G. Shuttle, R. Hamilton, J. Nelson, B. C. O’Regan, and J. R. Durrant, Measurement of ChargeDensity Dependence of Carrier Mobility in an Organic Semiconductor Blend. Adv. Funct. Mater. 20 (2010) 698–702, http://dx.doi.org/10.1002/adfm.200901734. [23] R. C. I. MacKenzie, T. Kirchartz, G. F. A. Dibb, J. Nelson, J. Phys. Chem. C 115 (19) (2011) 9806–9813, http://dx.doi.org/10.1021/jp200234m. [24] A. V. Nenashev, M. Wiemer, A. V. Dvurechenskii, F. Gebhard, M. Koch, S. D. Baranovskii, Why the apparent order of bimolecular recombination in blend organic solar cells can be larger than two: A topological consideration, Appl. Phys. Lett. 109 (2016) 033301, http://dx.doi.org/10.1063/1.4959076. [25] J. Szmytkowski, Electrostatic interface recombination in the system of disordered materials characterized by different permittivities, Synth. Met. 206 (2015) 120–123, http://dx.doi.org/10.1016/j.synthmet.2015.05.001. [26] G. Juška, N. Nekrašas, V. Valentinavičius, P. Meredith, and A. Pivrikas, Extraction of photogenerated charge carriers by linearly increasing voltage in the case of Langevin recombination, Phys. Rev. B 84 (2011) 155202, http://dx.doi.org/10.1103/PhysRevB.84.155202. [27] A. Pivrikas, G. Juška, R. Österbacka, M. Westerling, M. Viliūnas, K. Arlauskas, and H. Stubb, Langevin recombination and space-charge-perturbed current transients in regiorandom poly(3hexylthiophene), Phys. Rev. B 71 (2005) 125205, http://dx.doi.org/10.1103/PhysRevB.71.125205. [28] G. Juška, M. Viliūnas, and K. Arlauskas, Space-charge-limited photocurrent transients: The influence of bimolecular recombination, Phys. Rev. B 51 (1995) 16668, http://dx.doi.org/10.1103/PhysRevB.51.16668. [29] G. Juška, J. Kočka, M. Viliūnas, and K. Arlauskas, Subnanosecond bimolecular non-radiative recombination in a-Si:H, J. Non-Cryst. Solids 164–166 (1993) 579–582, http://dx.doi.org/10.1016/0022-3093(93)90618-8. [30] I. Balberg, R. Naidis, M. K. Lee, J. Shinar, L. F. Fonseca, Bipolar phototransport in p-conjugated polymer/C60 composites, Appl. Phys. Lett. 79 (2001) 197–199, http://dx.doi.org/10.1063/1.1383801. [31] G. J. Adriaenssens, V. I. Arkhipov, Non-Langevin recombination in disordered materials with random potential distributions, Solid State Commun. 103 (1997) 541–543, http://dx.doi.org/10.1016/S0038-1098(97)00233-0. [32] L. J. A. Koster, V. D. Mihailetchi, and P. W. M. Blom, Bimolecular recombination in polymer/fullerene bulk heterojunction solar cells, Appl. Phys. Lett. 88 (2006) 052104, http://dx.doi.org/10.1063/1.2170424.
ACCEPTED MANUSCRIPT • •
AC C
EP
TE D
M AN U
SC
RI PT
•
The non-geminate recombination was investigated in TQ1:PC61BM:PC71BM bulk heterojunctions employing photo-CELIV and integral ToF techniques. The shapes of photo-CELIV and ToF transients point out purely non-reduced Langevin recombination while the decay of the photogenerated charge carrier density measured with photoCELIV technique indicates reduced third-order recombination. In case of the concentrated electric field, photo-CELIV fails to evaluate the recombination rate and order from the density decay of charge carriers.