Hysteresis and Photoinstability Caused by Mobile Ions in Colloidal

Oct 10, 2017 - Organic–inorganic hybrid photovoltaics (PVs) have recently attracted considerable attention as their PV performance has rapidly impro...
0 downloads 9 Views 1MB Size
Subscriber access provided by UNIV OF CAMBRIDGE

Letter

Hysteresis and Photo-Instability Caused by Mobile Ions in Colloidal Quantum Dot Photovoltaics Jung Hoon Song, Xuan Dung Mai, Sohee Jeong, and Yong-Hyun Kim J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.7b02350 • Publication Date (Web): 10 Oct 2017 Downloaded from http://pubs.acs.org on October 10, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

Hysteresis and Photo-Instability Caused by Mobile Ions in Colloidal Quantum Dot Photovoltaics Jung Hoon Song,†,‡ Xuan Dung Mai,‡,∥ Sohee Jeong,*,‡,§ and Yong-Hyun Kim*,†



Graduate School of Nanoscience and Technology (GSNT) and Department of Physics, Korea

Advanced Institute of Science and Technology (KAIST), Daejeon 34141, Republic of Korea ‡

Nano-Convergence Systems Research Division, Korea Institute of Machinery and Materials

(KIMM), Daejeon 34103, Republic of Korea §

Department of Nanomechatronics, Korea University of Science and Technology (UST),

Daejeon 34113, Korea ∥Department

of Chemistry, Hanoi Pedagogical University No2, Vinh Phuc, Vietnam

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] *E-mail: [email protected]

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 20

ABSTRACT

Organic-inorganic hybrid photovoltaics (PVs) have recently attracted considerable attention as their PV performance has rapidly improved. Abnormal current-voltage (I-V) characteristics or I-V hysteresis, however, was occasionally observed in such systems with hampering the development of the PV technology. Here we study the hysteresis of organicinorganic hybrid colloidal quantum dot (CQD) PVs by analyzing I-V characteristics upon systematic modulation of organic components of CQDs. We demonstrate that an external bias stress to CQD films transiently prompts redistribution of mobile ions, particularly protons of surface ligands, thus leading to the formation of a transient space-charge region in the CQD films. The variable space-charge region causes I-V hysteresis and photo-instability of CQD PVs, which is closely correlated with the transient behavior of mobile ions. Our findings here could provide significant implications to the understanding of the influence of mobile ions on I-V hysteresis in other organic-inorganic hybrid PVs such as perovskites.

TOC GRAPHICS

ACS Paragon Plus Environment

2

Page 3 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

A hysteretic behavior in electronic devices was reported first time in 1962,1 and refers to the situation in which a transient internal state induces a dependency of system’s physical outputs on present and past inputs. The anomalous behaviors are often observed in thin-film based semiconducting systems with terminal-electrode configurations, where the motions of charge carriers including electrons, holes, and mobile ions are readily coupled.2,3 The hysteresis has been observed in thin-film photovoltaic (PV) devices including amorphous silicon, CIGS and CdTe solar cells,4-6 and featured by a fast transient response time less than one second. This sub-second hysteresis has been attributed to the capacitive charge accumulation or the trapping/de-trapping of electrons at defect states within the light-absorber or at interface. Very recently, hysteresis in emerging organic-inorganic hybrid PVs such as dyesensitized,7 quantum dot,8 and perovskite solar cells9 has been critically debated, mostly because their hysteretic response time is noticeably slow, e.g. over several hundreds of seconds. Their promising PV performance metrics [i.e., power-conversion efficiency (PCE), open-circuit voltage (VOC), short-circuit current density (JSC), and fill factor (FF)], if derived from the hysteretic current-voltage (I-V) characteristics, should be regarded as very unreliable or non-operational.10 The slow hysteretic response time could be originated from a plenty of mobile ions presented in the hybrid systems, as electronic hysteresis is typically very fast. Yet, the role of mobile ions has not been clearly resolved. Recently, it has been suggested that photo-instability of organic-inorganic hybrid PVs may associates with transient mobile ions.11 As an example of typical organic-inorganic hybrid system, colloidal quantum dot (CQD) solar cells offer the opportunity for low-cost high-performance PVs beyond the

ACS Paragon Plus Environment

3

The Journal of Physical Chemistry Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 20

Shockley-Queisser limit12 through solution process and multiple exciton generation.13 The state-of-the-art CQD PVs show an impressive PCE of over 10%.14,15 A typical CQD PV has a built-in potential (Vbi) at the junction between an n-type metal oxide and a p-type CQD layers. The CQD layer consists of light-absorbing inorganic semiconductor nanocrystal cores and short organic16 or atomic surface ligands,17 typically formed via a layer-by-layer (LBL) assembly.18 It is obvious that light absorption characteristics of the CQD layer can be easily tuned by controlling the core size of inorganic nanocrystals.19 It has been also known that the electrical conductivity and the doping polarity of the CQD layer are predominantly determined not only by the size of surface ligands or the interparticle distance, but also by the surface chemistry of organic ligands, such as atomic coordination, dipole moment, and pKa (which is a negative logarithm of proton dissociation constant, pKa, representing proton dissociation power).20-23 As an electrical characteristic, the I-V hysteresis of CQD PVs should be directly associated with detailed ligand-surface chemistry of CQDs. Therefore, we note that this material system presents an opportunity to study the influence of mobile ions migration in a PV device with better control over mobile ion concentration than that may be achieved within a recentlyfascinating perovskite absorber system. To our knowledge, the I-V hysteresis issue observed in CQD PVs has not been yet seriously discussed in relation with the effects of surface ligands and their characteristics. In this report, we have analyzed electrical transient behaviors of ligand-exchanged lead sulfide (PbS) conductive CQD films in order to understand I-V hysteresis of CQD PVs. Specifically, we have investigated mobile-ion induced transient behaviors of metal/CQD/metal devices and CQD field-effect transistors, of which I-V characteristics critically vary depending

ACS Paragon Plus Environment

4

Page 5 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

on pre-applied external bias. The external bias stress to the CQD films redistributes mobile ions or protons originated from surface ligands, and doping polarity of individual CQDs is modulated upon protonation or deprotonation, consistent with density functional theory (DFT) analysis. We have found that transient behaviors of mobile ions non-trivially correlate with I-V hysteresis of CQD PVs. Based on these understandings, we could control the hysteresis and photo-instability of CQD PVs by modifying the concentration of mobile ions such as protons and lithium ions. First, to figure out the transient response of CQD films without being affected by any other factors, we devised a metal-semiconductor-metal (MSM) structure, in which a ligand-exchanged PbS CQD film semiconducting layer is sandwiched between indium tin oxide (ITO) and Au electrodes, as shown in Figure 1A. Various thiol-containing surface ligands, namely, 3-mercaptopropionic acid (MPA), 1,2-ethanedithiol (EDT), and ethanethiol (ET) with having pKa values of 5 (carboxyl), 10 (thiol), and 53 (methyl), respectively, were used to fabricate 250 nm-thick conductive CQD films on ITO-coated glass via LBL spin coating method. The effective ligand-exchange process with all thiolcontaining ligands was comfirmed by using Fourier transform infrared (FT-IR) analysis (see Figure S1 in Supporting Information). In addition, the morphology of the conductive CQD films did not change with the surface ligands (Figure S2). After a thermal annealing process, a thin MoO3 electron blocking layer and Au electrode were sequential vacuum deposited atop the CQD films. The relevant energy diagram is displayed in Figure 1A for the MSM device using PbS CQDs with a bandgap of 1.4 eV.23,24 Figure S3 and S4 show optical properties and transmission electron microscopy (TEM) images, respectively, of PbS CQDs used in this study.

ACS Paragon Plus Environment

5

The Journal of Physical Chemistry Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 20

Figures 1. (A) Device structure and energy band diagram of the MSM (ITO/PbS/MoO3/Au) device. (B-D) Current density-voltage (J-V) curves of MSM devices treated with (B) 3mercaptopropionic acid (MPA), (C) 1,2-ethanedithiol (EDT), and (D) ethanethiol (ET). The dashed and solid lines represented J-V curves under dark and AM 1.5 G illumination, respectively. Bias annealing was indicated with applied voltage and duration time (black: no bias annealing, orange: 1 V for 100 s, purple: 2 V for 100 s, red: 2.5 V for 50 s, blue: 2.5 V for 100 s, magenta: 3 V for 200 s, green: 3.5 V for 100 s, and cyan: 3.5 V for 200 s).

ACS Paragon Plus Environment

6

Page 7 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

We examined photovoltaic responses of the MSM device after applying various external bias stresses. Before pre-biasing, all MSM devises showed the ohmic behavior with a negligible PCEs < 0.1% under AM 1.5 G illumination. When we pre-applied the bias stress, calling bias annealing, the MSM devices showed various PV responses, as shown in Figures 1B-D, depending on the applied bias voltage, duration, and exchanged surface ligands. For the PVresponsive MPA-capped CQD films (Figure 1B) with the carboxy (-COOH) terminal, we could see that the bias annealing clearly built a space-charge region (SCR) or p-n junction in the film, which was quickly saturated for the strength and duration of the bias stress. The PCE of the MPA-capped CQD MSM device was raised from 0.1 to 3.6 % due to the bias annealing. With the less-acidic thiol (-SH) group of EDT, however, we found that the PV response, or the SCR formation was sensitively subject to the strength and duration of bias annealing. The EDTcapped CQD films showed smaller PV response for larger and longer bias annealing than the MPA-capped CQD films, as shown in Figure 1C (See Table S1 in Supporting Information for detailed photovoltaic parameters). In contrast, no PV response was observed for the ET-capped CQD films, regardless of bias annealing details (Figure 1D). No SCR may form from the methyl (-CH3) group with no acidic proton available. Clearly, we can claim that the formation of bias-induced SCR sensitively depends on the acidity (pKa) of exchanged surface ligands from -COOH and -SH to -CH3, or the proton concentration generated from the terminals in the CQD films, as schematically shown in Figure 2A. Upon bias annealing, the mobile ions (protons) are redistributed across the thickness of CQD films, resulting in an enlarged p-n junction.2 The proton conduction or migration should have different response times for the bias annealing, depending on proton concentration. The higher

ACS Paragon Plus Environment

7

The Journal of Physical Chemistry Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 20

proton concentration leads to the faster ion migration and thus the faster formation of the SCR, as we can see in the MPA-, EDT-, and ET-capped CQD films.

Figures 2. (A) Schematic diagrams showing proton distributions and SCR formation in PbS CQD films with different proton concentrations upon bias annealing. (B) Time-dependent current density-voltage (J-V) profiles of an MPA-capped CQD MSM device under AM 1.5 G illumination, after applying bias annealing of +3.5 V for 200 s. (C) Variation of short-circuit current density (Jsc) as a function of time for EDT-capped CQD MSM device and MPA-capped CQD MSM devices with varied proton concentrations, measured after applying bias annealing of +3.5 V for 200 s.

ACS Paragon Plus Environment

8

Page 9 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

To characterize the transient behaviors of the mobile ion migration in CQD films, we monitored time-dependent PV response of MPA-capped CQD MSM devices with controlled proton concentrations. In Figure 2B, we monitored time-dependent J-V profiles of an MPAcapped CQD MSM device under AM 1.5 G illumination following a bias annealing at +3.5 V for 200 s. The PV response decreased gradually and relaxed ultimately to the original state in a timescale of tens of minutes. Note that this type of time-dependent PV characteristics caused by mobile ions could be linked to slow hysteresis and long term photo-instability issues observed in emerging organic-inorganic hybrid thin-film solar cells.11 When the bias annealing was reapplied to the CQD films, the initial PV response as well as its relaxation were exactly reproduced. Figure 2C shows the variations of short-circuit current density (JSC) as a function of time for EDT-capped and MPA-capped CQD MSM devices. Obviously, EDT-capped CQD films show slower JSC relaxation in comparison with MPA-capped CQD films, which strongly supports that the lower proton concentration leads to the slower ion migration and thus the slower formation of the SCR. Furthermore, when protons in the MPA-capped CQD films were intentionally replaced partially with 5% mol or 20% mol bulkier tetramethylammonium (TMA) ions, as confirmed by X-ray photoelectron spectroscopy (XPS) analysis in Figure S5, JSC decay was slowed down more and more with increasing TMA content, as shown in Figure 2C (See Figure S6 in Supporting Information for time-dependent current density-voltage profiles and discussion on the effect of TMA on carrier transport). This retarded JSC decay is not surprising because the bulky TMA ion could not move as fast as proton. These results clearly indicate that the proton migration is responsible for the transient formation of SCR in CQD films upon bias annealing.

ACS Paragon Plus Environment

9

The Journal of Physical Chemistry Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 20

In order to monitor the variation of doping polarity of MPA-capped CQD films upon proton migration, we fabricated a bottom-gate CQD field-effect transistor (FET), i.e. Si++/SiO2/PbS CQD film/Au source-drain (S-D) electrodes, as shown as the inset in Figure 3A. Without gate-bias annealing, MPA-capped PbS CQD FET exhibited almost ambipolar transport characteristics25, as shown in Figure 3A. When a +70 V or -70V bias was applied for 2 minutes, the CQD film switched into p-type or n-type, respectively. When -70 V was applied for 2 minutes, it shifted to n-type. For a positive gate bias, MPA-capped CQDs near the gate channel could have more deprotonated species because of less proton concentration than before bias annealing. Similarly, negative bias annealing drives proton accumulation, and hence increased electron concentration near the gate performing n-type characteristics. Therefore, we speculate that the deprotonated surface ligands could be an origin accounting for the variation in doping polarity of the CQD films.

ACS Paragon Plus Environment

10

Page 11 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

Figures 3. (A) Doping polarity variation of a MPA-capped PbS CQD field-effect transistor (FET) upon gate-bias annealing. The inset shows a schematic structure of the CQD FET. The gate-bias annealing is applied for 2 min before I-V scan. (B) DFT-calculated vacuum-aligned electrostatic potentials for MPA-capped PbS (111) surface slab models before/after deprotonation. (C) Schematic diagrams of energy band edges and atomic structure for PbS CQDs capped with MPA (left) and deprotonated MPA (right) upon gate-bias annealing.

To supplement this scenario, we performed first-principles DFT calculations for MPAcapped PbS (111) surface slab models.15,23,26 For deprotonated MPA ligands on the surface, we removed the hydrogen atom of the carboxyl group and added one electron and compensating positive charge background for overall charge neutrality. Figure 3B shows DFT-calculated plane-averaged electrostatic potentials for the MPA-capped PbS (111) slabs along the out-ofplane z direction before and after deprotonation. When the vacuum level is set to zero, one can see that the deprotonated MPAs clearly up-shift the entire energy band by 0.46 eV. This shift may be due to the change of dipole moment of surface ligands after deprotonation. Therefore, as summarized in Figure 3C, less (more) proton-concentrated CQD regions show the p-type (ntype) character,27 which well complying with the formation of variable p-n junction or SCR due to the proton migration. We could be able to analyze PV hysteresis in views of mobile ion migration by evaluating PbS CQD PVs with various CQD surface modifications. The measured J-V curves of MPA-capped PbS CQD PVs exhibit noticeable hysteresis in PV operation depending on the scan direction [forward (i.e. JSC→VOC) and backward (i.e. VOC→JSC)] and scan rate, while the EDTcapped CQD PVs show negligible hysteresis, as shown in Figures 4A-4B. The PV performances

ACS Paragon Plus Environment

11

The Journal of Physical Chemistry Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 20

are summarized in Table S2. The magnitude of hysteresis can be quantified using the hysteresis index (HI)28 as marked in Figure 4. MPA-capped PbS CQD PVs resulted in significant hysteresis at all scan rates. Particularly, the PCEs measured in the backward scan always exhibit higher values than those measured in the forward scan. When J-V curves are measured in the backward scan, the higher bias of the previous measurement points may act as bias annealing to the CQD film and thus accumulatively leads to the enhanced photovoltaic effect. On the other hand, EDTcapped CQD PVs resulted in reduced hysteresis, which can be explained with the decreased changes of SCR upon applied external bias due to the less proton concentration.

Figures 4. J-V curves of (A) MPA- and (B) EDT-capped PbS CQD PVs for forward and backward voltage scans. The inset shows the structure of PbS CQD heterojunction PVs. J-V

ACS Paragon Plus Environment

12

Page 13 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

curves were measured using slow (0.04 V/s, red) and fast (20 V/s, black) scan rates. The arrows indicate the scan direction. The hysteresis index (HI) was defined as the current density difference at a half VOC between sweep directions. (C) Typical hysteresis loops with fast (black) and slow (red) transient responses, where X and Y are input stimulation and output signal of hysteretic systems, respectively. (D) Non-trivial hysteresis loops with different magnitudes of saturated output signals (built-in potential, Vbi) depending on fast (black) and slow (red) transient responses of SCR formation in PbS CQD PVs, where V is the external bias stimulation.

This PV hysteresis appears characteristically different from conventional hysteretic systems such as magnetization29 and battery reaction,30 as schematically drawn in Figure 4C-4D. In conventional hysteretic systems, a slow transient response typically leads to a large hysteresis loop, indicating large energy dissipation through the slow response (see phase lag in hysteresis in SI for detailed discussion). This is because it takes a longer time for saturation of output signal. For CQD PVs, however, the saturation amplitude of the Vbi is much smaller than for conventional systems when transient response is slow. Therefore, the overall hysteresis loop should have a small area although the coercivity, a typical measure of hysteresis, is large. This is why EDT-capped CQD PVs with slow transient responses non-trivially show negligible hysteresis, compared to hysteretic MPA-capped CQD PVs with fast transient response. Based on these understandings, it is well expected for us to modulate the magnitude of hysteresis in CQD PVs by controlling the concentration of mobile ions. To demonstrate this, we fabricated PbS CQD PVs using tetrabutylammonium chloride (TBACl) pre-treated PbS CQDs, as shown in Figure 5A. This TBACl-pretreated PbS CQD films did not show any noticeable J-V

ACS Paragon Plus Environment

13

The Journal of Physical Chemistry Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 20

hysteresis, which might be attributed to the decreased concentration of mobile ions after exchanging native OA ligands by chloride during the TBACl pretreatment, as previously discussed.31,32 As proton concentration appears too small, we decided to add another mobile ions, Li, to check our mobile-ion hypothesis. Indeed, adding mobile Li ions to the CQD films reinitiated J-V hysteresis, as shown in Figure 5B. Therefore, we clearly demonstrated that either proton or Li ion could directly effect on the J-V hysteresis of PV devices. Because Li ion is heavier than proton, J-V hysteresis is noticeably observed at the high voltage sweep region. The PV performance metrics of both the CQD PVs are summarized in Table S3. We further monitored the transient PCE at maximum power point (Pmax) of PbS CQD PVs, as shown in the inset of Figure 5. Clearly, it turns out that photo-instability of CQD PVs correlates with hysteretic behaviors; while the hysteresis-free PbS CQD PV was quite photo-stable, the hysteretic CQD PV with Li ions was extremely unstable upon photo-excitation. The photo-bias is generated inside PVs under illumination and causes the redistribution of mobile ions or SCR. This photo-bias induced SCR opposites to the inherent SCR of the p-n junction. As time passes, the photo-bias induced SCR gradually cancels the VOC of the hysteretic CQD PV (see Figure S7). This strongly implies that the photo-bias induced SCR formation after mobile ion migration could cause photo-instability and hysteresis of organic-inorganic hybrid thin-film solar cells.11

ACS Paragon Plus Environment

14

Page 15 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

Figures 5. (A) J-V curves of PbS CQD PVs using TBACl pre-treated PbS CQDs. (B) The same as (A), but 5% bis (trifluoromethane) sulfonamide lithium salt (LiTFSI) was added in the CQD film. The inset shows the photo-stability at maximum power point (Pmax) under AM 1.5 G illumination.

In conclusion, we observed that an external bias stress temporarily altered the photovoltaic effect, as well as the carrier transport polarization due to the redistribution of mobile ionic species in PbS CQD films. This redistribution in mobile ionic species induced a variable SCR in CQD assemblies, which caused hysteretic I-V characteristics and photoinstability in optoelectronic devices. The hysteretic behavior of CQD PVs can be sensitively affected by surface chemistry of individual CQDs. AUTHOR INFORMATION Notes The authors declare no competing financial interests. ACKNOWLEDGMENT J. S. and X. M. contributed equally to this work.

ACS Paragon Plus Environment

15

The Journal of Physical Chemistry Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 20

This work was supported by the Global Frontier R&D (NRF-2017M3A6A7051087), National Research Foundation (NRF: 2016R1A2B3014182 and 2015R1A2A2A05027766), Vietnam National Foundation for Science & Technology Development (NAFOSTED: 103.99-2016.32), Science Research Center (2016R1A5A1008184), and the Global R&D program (1415134409) funded by KIAT. ASSOCIATED CONTENT Supporting Information. Detailed experimental procedures (synthesis of PbS CQD and fabrication of MSM, PVs, and FET devices), characterizations (absorption spectra, emission spectra, XPS spectra, J-V curves), and computational method. FT-IR analysis and AFM mapping of PbS CQD films. Crosssectional SEM images, XPS spectra, performance metrics of CQD PVs.

REFERENCES (1)

Hickmott, T. W. Low-Frequency Negative Resistance in Thin Anodic Oxide Films. J. Appl. Phys. 1962, 33, 2669-2682.

(2)

Strukov, D. B.; Snider, G. S.; Stewart, D. R.; Williams, R. S. The Missing Memristor Found. Nature 2008, 453, 80-83.

(3)

Riess, I. Mixed Ionic-Electronic Conductors-material Properties and Applications. Solid State Ionics 2003, 157, 1-17.

(4)

del Cueto, J. A.; Deline, C. A.; Albin, D. S.; Rummel, S. R.; Anderberg, A. Striving for a Standard Protocol for Preconditioning or Stabilization of Polycrystalline Thin Film Photovoltaic Modules. Proc. SPIE 2009, 7412, 741204-1-741204-12.

ACS Paragon Plus Environment

16

Page 17 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

(5)

Herman, M.; Jankovec, M.; Topic, M. Optimal I-V Curve Scan Time of Solar Cells and Modules in Light of Irradiance Level. Int. J. Photoenergy 2012, 2012, 1-11.

(6)

Graham, R. L.; Clark, L. A.; Albin, D. S. A Novel Approach for Correlating Capacitance Data with Performance during Thin-Film Device Stress Studies. Proc. SPIE 2011, 8112, 81120V-1-81120V-8.

(7)

Koide, N.; Han, L. Measuring Methods of Cell Performance of Dye-Sensitized Solar Cells. Rev. Sci. Instrum. 2004, 75, 2828-2831.

(8)

Liu H.; Tang J.; Kramer I. J.; Debnath R.; Koleilat G. I.; Wang X.; Fisher A.; Li R.; Brzozowski L.; Levina L.; et al. Electron Acceptor Materials Engineering in Colloidal Quantum Dot Solar Cells. Adv. Mater. 2011, 23, 3832-3837.

(9)

Snaith, H. J.; Abate, A.; Ball J. M.; Eperon G. E.; Leijtens T.; Noel N. K.; Stranks S. D.; Wang J. T.-W.; Wojciechowski K.; Zhang W. Anomalous Hysteresis in Perovskite Solar Cells. J. Phys. Chem. Lett. 2014, 5, 1511-1515.

(10)

Christians, J. A.; Manser, J. S.; Kamat, P. V. Best Practices in Perovskite Solar Cell Efficiency Measurements. J. Phys. Chem. Lett. 2015, 6, 852-857.

(11)

Tress, W.; Marinova, N.; Moehl, T.; Zakeeruddin, S. M.; Nazeeruddin, M. K.; Grätzel, M. Understanding the Rate-Dependent J-V Hysteresis, Slow Time Component, and Aging in CH3NH3PbI3 Perovskite Solar Cells: the Role of a Compensated Electric Field. Energy Environ. Sci. 2015, 8, 995-1004.

(12)

Shockley, W.; Queisser, H. J.; Detailed Balance Limit of Efficiency of p-n Junction Solar Cells. J.Appl. Phys. 1961, 32, 510-519.

ACS Paragon Plus Environment

17

The Journal of Physical Chemistry Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(13)

Page 18 of 20

Semonin, O. E.; Luther, J. M.; Choi, S.; Chen, H. Y.; Gao, J.; Nozik, A. J.; Beard, M. C. Peak External Photocurrent Quantum Efficiency Exceeding 100% via MEG in a Quantum Dot Solar Cell. Science 2011, 334, 1530-1533.

(14)

Liu, M.; Voznyy, O.; Sabatini, R.; de Arquer, F. P.; Munir, R.; Bakawi, A. H.; Lan, X.; Fan, F.; Walters, G.; Kirmani, A. R.; et al. Hybrid Organic-Inorganic Inks Flatten the Energy Landscape in Colloidal Quantum Dot Solids. Nat. Mater. 2016, 16, 258263.

(15)

Song, J. H.; Choi, H.; kim, Y.-H.; Jeong, S. High Performance Colloidal Quantum Dot Photovoltaics by Controlling Protic Solvents in Ligand Exchange. Adv. Energy Mater. 2017, 1700301-1-1700301-6.

(16)

Beak, S.-W.; Song, J. H.; Choi, W.; Song, H.; Jeong, S.; Lee, J.-Y. A ResonanceShifting Hybrid n-Type Layer for Boosting near-Infrared Response in Highly Efficient Colloidal Quantum Dots Solar Cells. Adv. Mater. 2015, 27, 8102-8108.

(17)

Kim, S.; Marshall, A. R.; Kroupa, D. M.; Miller, E. M.; Luther, J. M.; Jeong, S.; Beard, M. C. Air-Stable and Efficient PbSe Quantum-Dot Based upon ZnSe to PbSe Cation-Exchanged Quantum Dots. ACS Nano 2015, 9, 8157-8164.

(18)

Luther, J. M.; Law, M.; Song, Q.; Perkins, C. L.; Beard, M. C.; Nozik, A. J. Structural, Optical, and Electrical Properties of Self-Assembled Films of PbSe Nanocrystals Treated with 1,2-Ethanedithiol. ACS Nano 2008, 2, 271-280.

(19)

Jasieniak, J.; Califano, M.; Watkins, S. E. Size-Dependent Valence and Conduction Band-Edge Energies of Semiconductor Nanocrystals. ACS Nano 2011, 5, 5888-5902.

ACS Paragon Plus Environment

18

Page 19 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

(20)

Hughes, B. K.; Ruddy, D. A.; Blackburn, J. L.; Smith, D. K.; Bergren, M. R.; Nozik, A. J.; Johnson, J. C.; Beard, M. C. Control of PbSe Quantum Dot Surface Chemistry and Photophysics Using an Alkylselenide ligand. ACS Nano 2012, 6, 5498-5506.

(21)

Liu, Y.; Gibbs, M.; Puthussery, J.; Gaik, S.; Ihly, R.; Hillhouse, H. W.; Law, M. Dependence of Carrier Mobility on Nanocrystal Size and Ligand Length in PbSe Nanocrystal Solids. Nano Lett. 2010, 10, 1960-1969.

(22)

Oh, S. J.; Berry, N. E.; Choi, J. -H.; Gaulding, E. A.; Lin, H.; Paik, T.; Diroll, B. T.; Muramoto, S.; Murray, C. B.; Kagan, C. R. Designing High-Performance PbS and PbSe Nanocrystal Electronic Device through Stepwise, Post-Synthesis, Colloidal Atomic Layer Deposition. Nano Lett. 2014, 14, 1559-1566.

(23)

Brown, P. R.; Kim, D.; Lunt, R. R.; Zhao, N.; Bawendi, M. G.; Grossman, J. C.; Bulovic, V. Energy Level Modification in Lead Sulfide Quantum Dot Thin Films through Ligand Exchange. ACS Nano 2014, 8, 5863.

(24)

Gao, J.; Perkins, C. L.; Luther, J. M.; Hanna, M. C.; Chen, H. -Y.; Semonin, O. E.; Nozik, A. J.; Ellingson, R. J.; Beard, M. C. n-Type Transition Metal Oxide as a Hole Extraction Layer in PbS Quantum Dot Solar cells. Nano Lett. 2011, 11, 3263-3266.

(25)

Balazs, D. M.; Nugraha, M. I.; Bisri, S. Z.; Sytnyk, M.; Heiss, W.; Loi, M. A. Increased Efficiency in pn-Junction PbS QD Solar Cell via NaHS Treatment of the pType Layer. Appl. Phys. Lett. 2014, 104, 112104-1-112104-5.

(26)

Choi, H.; Ko, J.-H.; Kim, Y.-H.; Jeong, S. Steric-Hindrance-Driven Shape Transition in PbS Quantum Dots: Understanding Size-Dependent Stability. J. Am. Chem. Soc. 2013, 135, 5278-5281.

ACS Paragon Plus Environment

19

The Journal of Physical Chemistry Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(27)

Page 20 of 20

Yun, J. M.; Park, S.; Hwang, Y. H.; Lee, E.-S.; Maiti, U.; Moon, H.; Kim, B.-H.; Bae, B.-S.; Kim, Y.-H.; Kim, S. O. Complementary p- and n-Type Polymer Doping for Ambient Stable Graphene Inverter. ACS Nano 2014, 8, 650.-656.

(28)

Sanchez, R. S.; Gonzalez-Pedro, V.; Lee, J.-W.; Park, N.-G.; Kang, Y. S.; Mora-Sero, I.; Bisquert, J. Slow Dynamic Processes in Lead Halide Perovskite Solar Cells. Characteristic Times and Hysteresis. J. Phys. Chem. Lett. 2014, 5, 2357-2363.

(29)

Farrow, R. F. C.; Marks, R. F.; Carey, M. J.; Gurney, B. A. Ultrafast Time Resolved Photoinduced

Magnetization

Rotation

in

a

Ferromagnetic/Antiferromagnetic

Exchange Coupled System. Phys. Rev. Lett. 1999, 82, 3705-3708. (30)

Kumar, A.; Ciucci, F.; Morozovska, A. N.; Kalinin, S.; Jesse, S. Measuring Oxygen Reduction/Evolution reactions on the Nanoscale. Nat. Chem. 2011, 3, 707-713.

(31)

Owen, J. S.; Park, J.; Trudeau, P.-E.; Alivisatos, A. P. Reaction Chemistry and Exchange at Cadmium-Selenide Nanocrystal Surfaces. J. Am. Chem. Soc. 2008 130, 12279-12281.

(32)

Kim, S.; Noh, J.; Choi, H.; Ha, H.; Song, J. H.; Shim, H. C.; Jang, J.; Beard, M. C.; Jeong, S. One-Step Deposition of Photovoltaic Layers Using Iodide Terminated PbS Quantum Dots. J. Phys. Chem. Lett. 2014, 5, 4002-4007.

ACS Paragon Plus Environment

20