Reducing Surface Recombination Velocities at the ... - ACS Publications

Subscriber access provided by Columbia University Libraries

Letter

Reducing Surface Recombination Velocities at the Electrical Contacts Will Improve Perovskite Photovoltaics Jian Wang, Weifei Fu, Sarthak Jariwala, Irika Sinha, Alex K.-Y. Jen, and David S Ginger ACS Energy Lett., Just Accepted Manuscript • DOI: 10.1021/acsenergylett.8b02058 • Publication Date (Web): 10 Dec 2018 Downloaded from http://pubs.acs.org on December 11, 2018

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 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 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.

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 22 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

ACS Energy Letters

Reducing Surface Recombination Velocities at the Electrical Contacts Will Improve Perovskite Photovoltaics Jian Wang,1 Weifei Fu,2 Sarthak Jariwala,1,2 Irika Sinha,1 Alex K.-Y. Jen1,2,3,4 David S. Ginger1* 1Department

2Department

of Chemistry, University of Washington, Seattle, WA 98195, USA

of Materials Science and Engineering, University of Washington, Seattle, WA 98195, USA

3Department

4Department

of Chemistry, City University of Hong Kong, Kowloon, Hong Kong

of Materials Science and Engineering, City University of Hong Kong, Kowloon, Hong Kong

Corresponding Author * E-mail:[email protected]

1 ACS Paragon Plus Environment

ACS Energy 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 22

Abstract We explore the effects of non-radiative recombination at the extracting contacts on the achievable performance of halide perovskite photovoltaic cells.

First, we perform device

simulations using standard drift-diffusion models with experimental semiconductor parameters matching those of methylammonium lead triiodide (MAPbI3). We quantify the range of surface recombination velocities (SRVs) that would allow this archetypal perovskite to reach power conversion efficiencies of 27%. In particular, for contacts with well-aligned energy levels, SRVs of ~1-10 cm/sec should enable open circuit voltages of 1.30V, within 96% of the ShockleyQuiesser limit. Next, we use time-resolved photoluminescence to experimentally determine the SRVs on 14 different common electron- and hole-extracting contacts, including TiO2, SnO2, ZnO, PCBM, ITIC, ICBA, TPBi, PEDOT:PSS, PTAA, PVK, NiO, MoO3, WO3, and spiro-OMeTAD. These results point the way to the selection and rational engineering of better contacts as a means to achieve higher efficiencies in perovskite solar cells.

ToC


Page 3 of 22 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

ACS Energy Letters

Organic-inorganic halide perovskite photovoltaics (PVs), with a demonstrated power conversion efficiency of over 23%,1 offer a pathway for continued cost reduction and efficiency increases in thin-film solar cells. Champion cells reported in the literature already have demonstrated 97% of the theoretical limit for current density, and 98% for the fill factor – leaving improvement of the open-circuit voltage (VOC) as the largest remaining opportunity for continued increases in performance.2 Approaching theoretical VOC limits requires eliminating all competing non-radiative channels. Several studies have shown that processing additives and/or smallmolecule passivators can lead to remarkably long carrier lifetimes,3-5 internal photoluminescence quantum efficiencies (PLQE) even approaching 100%, and quasi-Fermi level splittings of over 97% of the radiative limit in the prototypical methylammonium lead triiodide perovskite (CH3NH3PbI3, MAPbI3) thin films.6 However, achieving these values in an operational device architecture has remained elusive, because contacting the perovskite with extracting contacts generally induces new, non-radiative loss pathways at the surface,7,8 resulting in a decrease in the PLQE and PL lifetime.9,10 Recently, a number of studies have discussed the roles that energy level alignment,11-13 conductivity/mobility,14 stability,15 and processing conditions,16 play in selecting optimum contacts for perovskite solar cells.

Surface recombination, as quantified by the surface

recombination velocity (SRV),17,18 despite being as an well-known important parameter that limits solar cell performance,17,19-21 has so far received less academic interest in the context of perovskite solar cells.22-24 When the atomic lattice is abruptly broken at a surface/interface, unsatisfied dangling bonds (or foreign bonds) introduce electronic energy levels inside the bandgap, which enhance electron-hole non-radiative recombination at the surface/interface by acting as stepping stones for



Page 4 of 22

charge carrier transitions between the conduction and valence bands. The surface recombination velocity, defined as the constant ratio of surface recombination rate (Rs) to excess minority carrier concentration at the surface/interface (ns), 𝑆𝑅𝑉 = 𝑅𝑠/∆𝑛𝑠, is the surface analogue of the minority carrier lifetime in a bulk semiconductor. This parameter provides a convenient homogeneous boundary condition for the excess minority carrier concentration, which can be used in both device modelling

calculations21

and

in

the

interpretation

of

photoluminescence

lifetime

measurements.17,19 Historically, many PV technologies, such as in Si,17,19 III-IV,20 CIGS,21 had been successfully optimized for higher performance by focusing on strategies to reduce SRV. In this study, we first perform drift-diffusion simulation to ascertain the likely impact of SRV on perovskite device efficiency at the highest levels of device performance, as other properties like intrinsic bulk and surface defects are eliminated. We then report experimental measurements of the SRVs at the perovskite (MAPI3) surface with or without surface passivation, as well as the SRVs at various perovskite/contact layer interfaces. Finally, we compare our simulations and experimental results with the reports of device performance achieved to date, and discuss possible future contact layer development strategies.


Page 5 of 22 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

ACS Energy Letters

Figure 1. (a) SCAPS simulated J-V curves for a 1000 nm-thick MAPbI3 cell with varying SRV values from 105 cm/sec to 10-2 cm/sec in logarithmic steps. Inset: Energy level diagram for the simulation (assuming 0.1 eV offset at both contacts. See SI Fig. S1 for the impact of energy barrier selection). Detailed simulation parameters are given in Table S1. (b) – (e) Summarized PV parameters (PCE, JSC, VOC, FF) as a function SRV values. (the simulated values plateau slightly below the Shockley-Quiesser limit 30.2% as a combined result of our selection of limited thickness, and bulk recombination coefficient, as discussed in SI, see Fig. S2). Figures 1 show device performance parameters we obtained using drift-diffusion simulations with the SCAPS software package25,26 for a 1000-nm-thick MAPI3 solar cell with the hypothetical architecture depicted in the inset of Fig. 1A, using varying assumptions about the SRV. As SRV is purely a 2D parameter, here we only consider the absorber/electron transport 5 ACS Paragon Plus Environment


Page 6 of 22

layer (ETL) and the absorber/hole transport layer (HTL) interface, but not the ETL and HTL volume, i.e. we assume no optical or transport loss beyond the interface. All other material/device parameters in the simulations are selected based on reported literature values (experimentally determined, whenever possible), as given in Table S1. Furthermore, for simplicity, we omit an explicit discussion of the contribution of ion/ion vacancy migration, which had been shown to impact the perovskite recombination properties,4,27 though these effects are treated indirectly through our use of TOPO-passivation, which is believed to passivate halide ion vacancies.3 Here, we selected the SRVs for simulation (from high to low) to span a wide range encompassing the following scenarios: (1) metallic contacts, which typically induce SRVs on the order of >105 cm/sec in many semiconductors.17,21 As a result, metallic contacts, other than establishing a built-in field, usually do not prevent surface recombination of the minority carriers. (2) In the range of 103~104 to represent “unpassivated” perovskite interfaces without induced defects. The few reported SRVs for native perovskite (MAPbBr3 single crystal, CsSnI3 ingot) surfaces have been on the order of 103~104 cm/sec for the free surface.22,23 (3) The reported lowest SRV for Si PV after passivation is on the order of 0.1~1 cm/sec.19 (4) We chose 10-2 cm/sec simply as an endpoint to represent a near ideal interface. Here we assume that the SRV is identical at both contacts, i.e. the hole recombination velocity at the electron collection interface and the electron recombination velocity at the hole collection interface, and vary them simultaneously in logarithmic steps. Fig. 1b shows that the simulated device efficiency increases from 20 % to 27%, as the SRV decreases from 105 cm/sec to ~1 cm/sec, and saturates at ~27% upon further decreasing the SRV to 10-2 cm/sec. The saturation at 27% PCE, below the Shockley-Quiesser limit of 30.2% for a 1.63 eV gap material, is the result primarily of our practical choice to use an experimental absorption profile, and a 1000-nm-thick


Page 7 of 22 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

ACS Energy Letters

active layer (which results in imperfect absorption and a reduced JSC), and our choice of an experimental value for the bulk recombination coefficient rather than a coefficient chosen to match detailed balance arguments (which results in a reduced VOC compared to the detailed balance limit). By increasing device thickness or reducing the bulk recombination value we can recover the theoretical limit (see SI, Fig. S2), however our focus here is to ascertain to what extent surface recombination is likely to be limiting state of the art perovskite solar cells using practical experimental parameters today. Next, we turn to examine the effects of varying the SRV on the specific photovoltaic performance parameters. Fig. 1c shows that, as the SRV decreases from ~ 105 cm/sec to ~ 103 cm/sec, the JSC increases rapidly to ~22.93 mA/cm2, which corresponds to a ~ 99.9 % internal quantum efficiency for this film thickness, and saturates thereafter. This result reflects the fact that JSC increases linearly with respect to the carrier extraction, i.e. the difference between the generation (G) and recombination (R) rates, where G is a constant only defined by absorption and R is proportional to SRV. It is also noted that, the JSC saturation value corresponds to a 92.9% collection yield of all photons above the selected bandgap (24.67 mA/cm2, 1.63 eV), which leaves behind a 1.74 mA/cm2 uncollected photocurrent due to the insufficient absorption near its band edge. On the other hand, as the SRV is decreased, VOC increases consistently up to 1.30 V at an SRV of ~1 cm/sec (Fig. 1d), with the VOC gaining approximately 45 mV per decade reduction in SRV over the range from 104 cm/sec to 10 cm/sec. The plateau at ~ 1.30V is within 96% of the Shockley-Quessier limit (1.355 V) based on the material bandgap (1.63 eV), and this plateau occurs in our simulations because of our adoption of an experimentally determined bimolecular recombination coefficient of the MAPbI3 material (410-11 cm3/sec).3 This bulk bimolecular recombination becomes the dominant loss when SRV is below 100 cm/sec, which ultimately limits



Page 8 of 22

the simulated device performance from approaching the detailed balance limit (See Figure S3 for details). Further improvements in perovskite thin film growth and passivation to produce films with even higher internal photoluminescence quantum yields would still benefit from reduction of the SRV to below 10 cm/sec. The VOC gain as the logarithm of the SRV is also expected, the slope of which is dependent on the selected energy offset (maximum slope ~ 60 mV per decade, see SI, Fig. S2), a direct result of the proportionality relation between VOC and non-radiative loss (kT/e  ln(1/PLQE)). As the FF at the maximum power point represents a recombination environment between the JSC and VOC scenarios, its dependence on SRV as well shows a mixed behavior between these limits, as expected (Fig. 1e). Comparing our simulation results to the current record efficiencies ~ 23%,1 we could estimate that a SRV of ~ 5103 cm/sec could be tolerated before becoming the limiting factor in the performance of today’s current cells. We emphasize that, while the leading efficiency cells possess slightly different perovskite material parameters due to the composition and processing variations, these simulation results based on prototypical MAPbI3 are meant to provide order of magnitude upper limits to help frame the search for better contact materials. In summary, the simulation results clearly show that interface engineering efforts to decrease the SRV down to the orders of ~ 1-10 cm/sec could open the door to the perovskite efficiencies up to ~ 27%, which would result mainly from an increase in the achievable VOC. In addition to generally benchmarking the required SRV values, we note two specific points. Firstly, there exists an interplay between the SRV values and the contact energy alignment on device performance. Figure S1 shows that the worse the contact alignment (higher energy offset) is, the lower the device performance is at a medium/high SRV value, while they all saturate to ~ 27% at very low SRV values. This result implies that a more poorly aligned contact has even


Page 9 of 22 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

ACS Energy Letters

more stringent requirements for reducing the SRV, which is a practical point as often extracting contacts don’t necessarily form an ideal energy alignment with the perovskite active layer (Figure S5). Second, the impact of the two interfaces are not identical, but depend on whether they are located at the front or the back contacts (Figure S2). Because of the higher generation rate at front (illuminated) contact, a high SRV at this interface would more significantly limit the device performance, yet the back interface remains important – both interfaces must be optimized to possibly achieve the predicted ~27 % efficiency. Nevertheless, results in Figure S2 suggest there is possibly a judicious sequence when optimizing the contacts – first the front contact then the back contact (see SI for more details). Having established a quantitative criterion for the SRV in perovskite devices, we next turn to the experimental determination of this parameter. In conventional PV technology such as Si, the free surface SRV can be determined by either time-resolved photoconductance (TRPC) or TRPL measurements.17,19 However, photoconductive approaches can be difficult to apply near interfaces, as the contact layers are usually more conductive than the semiconductor, making analysis of the signal more challenging. In any event, radiative efficiency is what ultimately matters in approaching the

maximum power conversion efficiency,28 leaving PL under open circuit

conditions as a straightforward probe. In TRPL measurements of a semiconductor slab with interfaces, an effective decay lifetime (τeff) is measured. We fit the transient PL decays (τeff) using a stretched exponential, consistent with distributed kinetics in a polycrystalline sample,3 which is then related to the bulk lifetime (τb) and surface lifetime (τs):18 1 𝜏𝑒𝑓𝑓

=

1 1 + 𝜏𝑏 𝜏𝑠

(1)



Page 10 of 22

In equation 1, τb is a factor that only depends on the bulk material properties, while τs is related to the SRV, sample thickness (W) and the diffusion constant (D) of the excess carriers. Given any thin film sample would have a top and bottom surface/interface, previous study has established that τs can be approximated analytically in two limiting cases.18 First, when the SRV at the top and bottom interfaces are identical, τs is given by Equation 2: 1 𝑊 2 𝜋

𝑊

𝜏𝑠≅2𝑆𝑅𝑉 + 𝐷

()

, when 𝑆𝑅𝑉1 = 𝑆𝑅𝑉2;

(2)

On the other hand, when the SRV at one contact (SRV2) is far greater than the SRV at the other contact (SRV1), or equivalently when SRV1 ~ 0, τs is approximately given by Equation 3: 𝑊

4 𝑊 2 𝜋

𝜏𝑠≅𝑆𝑅𝑉2 + 𝐷

()

, when 𝑆𝑅𝑉1≅ 0;

(3)

We note that sample thickness (W) can be determined, and the diffusion constant (D) can be estimated from measured carrier mobility values (μ) based on the Einstein relationship (𝐷 = 𝜇 𝑘𝐵𝑇), where kB and T are Boltzmann constant and temperature, respectively. Therefore, by measuring the bulk lifetime (τb) and knowing the boundary conditions (either SRV1= SRV2 or SRV1=0), the SRV of the interface in question can be assessed from measured TRPL lifetime (τeff). One approach to determine τb is to measure and extrapolate τeff from a very thick sample series, such as a Si wafer19 or Perovskite single crystal,29 where bulk recombination dominates over surface recombination. However, it is difficult to adopt this method in polycrystalline perovskite thin films, where the thickness is usually limited below 1 μm due to practical processing considerations. To address this difficulty, we adopt the strategy of Lewis base surface passivation using trioctylphosphine oxide (TOPO) passivation on MAPI3 thin films, which can yield internal PL quantum efficiencies over ~90%, and lifetimes exceeding ~8 μs.3,6 Since we have shown that TOPO molecules only bond to the external surface (passivating surface defects),3 we make the 10 ACS Paragon Plus Environment

Page 11 of 22 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

ACS Energy Letters

assumption here that the near unity internal PLQE of this processing route implies that remaining recombination occurs nearly all in bulk. In other words, this method allows us to establish a lower limit for the corresponding bulk lifetime (τb > 8 μs), since surface passivation alone can restore well-prepared samples to such levels of performance. With this estimate in hand, we can then determine the SRVs of MAPbI3(TOPO)/Air interfaces and the Glass/MAPbI3 interface, by plotting the experimentally determined τeff (Figure S4) as a function of thickness (W) on top of numerically calculated τeff-W relationships for different combinations of τeff and W (Figure 2). Comparing the two limiting scenarios (SRV1= SRV2, or SRV1  0) (eqn.2 & 3), we find that our data indicate that the SRV of our TOPO-passivated samples on glass lies between SRV2  10 cm/sec, if SRV1  0 (dashed lines), and SRV  6 cm/sec, if SRV1= SRV2 (solid lines). The τeff values are similar when measured from either excitation direction, so we can’t further resolve the SRV differences between these two interfaces. However, given that SRV can vary by orders of magnitude for different electrode interfaces, this relatively small level of uncertainty in our reference samples is sufficient to allow us to benchmark the SRVs of many perovskite/electrode contacts in use today, setting an upper limit ( 10 cm/sec). In other words, with a clean long-lived reference sample, we can resolve SRV numerically down to ~10 cm/sec for any perovskite/contact layer interface by measuring PL in this system. We also apply analysis to the unpassivated perovskite/air interface (without TOPO treatment). Doing so yields the data in open squares plotted in Figure 2, which suggests a ~ 1000 cm/sec SRV, which is consistent with SRV values reported for unpassivated perovskite crystals (MAPbBr3 and CsSnI3) in the literature,22,23 giving us confidence in the method.



Page 12 of 22

Figure 2. Calculated (lines) and experimental (squares), effective PL lifetimes, τeff, plotted vs. perovskite layer thickness for varying SRV values. Two conditions are assumed for the calculation to provide a bracket of SRV estimates: equal recombination at both contacts, SRV1=SRV2 (solid lines), and zero recombination at one contact, SRV1=0, (dashed lines). These assumptions allow us to determine SRV to within an order of magnitude. Next, we apply this approach to quantify the SRV at the interface between the perovskite layer and 14 common contact layers in both bottom and top contact configurations, including spiroOMeTAD, poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) (PEDOT:PSS), poly[bis(4phenyl)(2,4,6-trimethylphenyl)amine (PTAA), poly(9-vinylcarbazole) (PVK), NiO, WO3, MoO3 as hole transport layers (HTLs), and TiO2, ZnO, SnO2, phenyl-C61-butyric acid methyl ester (PC61BM), 1,3,5-tris(1-phenyl- 1H-benzimidazol-2-yl)benzene (TPBi), 3,9-bis(2-methylene-(3(1,1-dicyanomethylene)-indanone))-5,5,11,11-tetrakis(4-hexylphenyl)-dithieno[2,3-d:2’,3’-d’]-sindaceno[1,2-b:5,6-b’]dithiophene (ITIC), and indene-C60 bisadduct (ICBA) as electron transport layers (ETLs). The thicknesses and band structures of these contact layers are summarized in supporting information (Experimental Section & Figure S5). For contact layers usually processed underneath the perovskite, we process TOPO treatment on the perovskite; for contact layers processed on top of the perovskite, we omit the TOPO treatment. In either case the interface to be evaluated is identical to that in standard devices, while the other interface (either glass/perovskite 12 ACS Paragon Plus Environment

Page 13 of 22 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

ACS Energy Letters

or perovskite/TOPO) possess a low SRV ( 10 cm/sec) value that would not limit this evaluation process. Fig. 3(a) & (c) show the raw PL decays for several representative contacts, along with the fits used to extract the lifetimes. It can be seen that all contacts reduce the PL lifetime relative to the glass/perovskite/TOPO structure, with some, such as PEDOT:PSS, spiro-OMeTAD, or PC61BM, reducing the lifetime by factors of 1000 or more. We note that occasionally authors will interpret reductions in PL intensity and lifetime at open circuit upon contacting the active layer as indicative of a “good” contact that facilitates charge extraction. However, we believe that such an interpretation is incorrect. At VOC conditions, there is no net charge extraction, and any PL quenching is indicative of new non-radiative loss pathways created by the interface and should be avoided. Therefore, combined PL analyses at both VOC and JSC conditions are underway to discern these processes toward a thorough evaluation of contact layer materials. Figure 3(b) & (d) summarizes the extracted SRVs on various ETLs and HTLs either at bottom (labeled as ‘B’) or on top (labeled as ‘T’) the perovskite layer, all in relevant regular n-i-p or inverted p-i-n configurations. We find that the measured SRVs vary from ~10 cm/sec to ~5000 cm/sec with different contact layer interfaces. Some contact layers that are prevalently seen in leading efficient devices,30 such as PTAA,

TiO2, and NiO, which interestingly all reside

underneath the perovskite, show low SRVs < 30 cm/sec. Some other contact layers, such as PEDOT:PSS, spiro-OMeTAD, MoO3 and PC61BM, show large SRVs on the order of 5000 cm/sec. We note that the extracted SRVs should depend on the exact processing of specific contact layer material. For example, while we observe a low SRV with TiO2, consistent with the report how it was processed,31 a different TiO2 processing could result with a strong PL quenching hence a higher SRV.32 Nevertheless these results suggest that while some contact layers are already in the



Page 14 of 22

range that would allow perovskite solar cells to exceed 23%, the majority are not, pointing the way to improve performance subject to further interface engineering. Another trend we observe in the data from Fig. 3 is that, when using electrodes commonly used in electroluminescence applications,33 such as PVK and TPBi,34 we observe lower SRVs compared to other typical contact layers. This result is consistent with the reciprocity relation,28 which states that a good LED should be a good solar cell, and vice versa. As an ideal selective electron (hole) contact should only exchange electrons (holes) with the conduction (valence) band of the absorber, regardless whether the exchange is an ‘extraction’ or an ‘injection’, further investigation of a variety of carrier injection layer materials from the OLED library could be a promising route for perovskite PV contact optimization, particularly those with low extraction barriers.35 Finally, we note that the surface recombination rate is proportional to both SRV and excess minority carrier density near the contact. In some studies, engineering efforts are devoted to limiting the minority carrier density near the contact, via creating a bandgap gradient9 or passivating the perovskite minority traps near the interface.36 We recognize that the overall assessment of an electrode involves not only a low SRV, but also a good energy alignment, the facilitation of carrier extraction,8 as well the ease of processing. Nevertheless SRV, is convenient parameter to help screen contacts and their potential performance.


Page 15 of 22 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

ACS Energy Letters

Figure 3. Experimental PL decays for perovskite films in contact with various for HTLs (a) and various ETLs (c). Inset: shorter time windows for low lifetime stacks. Extracted SRV values for HTLs (b), and ETLs (d). The labels (B) and (T) represent that the contact layer is placed at bottom (B) or on top (T) of the perovskites, respectively. In conclusion, we have used numerical simulations to examine the impact of a range of SRV values on perovskite solar cell performance. We find that going beyond today’s champion cell performance will likely require reducing recombination losses at the perovskite/electrode interfaces, in some cases by orders of magnitude. We find that efficiencies up to ~ 27% could be achieved with today’s methylammonium lead triiodide perovskite thin films if SRV values in the range of ~10 cm/sec can be achieved at both the electron- and hole-extracting contacts. We also propose that photoluminescence-based will be increasingly useful in the search for new contacts and interface passivation strategies for perovskites as a means to probe SRVs without the need to optimize full device stacks. Finally, Perovskite solar cells should stand to benefit from exploration



Page 16 of 22

of new passivation strategies or new extraction layers, perhaps by exploring a wider range of the OLED literature, and perhaps by exploration of new device geometries, similar to the designs used for Si to reduce the amount of surface area between the absorbing Si and the extracting contact. Doing so provides a clear pathway to reaching single-layer perovskite efficiencies of 27% and beyond.

Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Experimental details, drift-diffusion simulation parameters, impact of energy alignment, comparison to Shockley-Quessier limit, raw TRPL data on different thickness, and PLQE data.

Acknowledgements This project acknowledges support from institutional, intellectual, and human resources built up over the years from the Office of Naval Research (ONR) photovoltaics program and current award N00014-17-1-2201. The authors also acknowledge support from University of Washington Clean Energy Institute, and the Washington Research Foundation. D.S.G acknowledges support from the University of Washington, Department of Chemistry Kwiram Endowment. J.W. acknowledges the support from Washington Research Foundation innovation fellowship and Mistletoe Foundation research fellowship, and also acknowledges Dr. Yeechi Chen’s help for the ToC art. F. W. and A.K.-Y.J. also acknowledges the financial support from the Asian Office of Aerospace R&D (FA2386-15-1-4106) and ONR N00014-17-1-2260. S.J.


Page 17 of 22 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

ACS Energy Letters

acknowledges the National Science Foundation Research Traineeship under award NSF DGE1633216.

References (1) NREL

Best

Research-Cell

Efficiencies

(Accessed

September

5th,

2018).

https://www.nrel.gov/pv/assets/images/efficiency-chart.png. (2) Tress, W. Perovskite Solar Cells on the Way to Their Radiative Efficiency Limit – Insights Into a Success Story of High Open-Circuit Voltage and Low Recombination. Adv. Energy Mater. 2017, 7 (14), 1602358. (3) DeQuilettes, D. W.; Koch, S.; Burke, S.; Paranji, R. K.; Shropshire, A. J.; Ziffer, M. E.; Ginger, D. S. Photoluminescence Lifetimes Exceeding 8 μs and Quantum Yields Exceeding 30% in Hybrid Perovskite Thin Films by Ligand Passivation. ACS Energy Lett. 2016, 1 (2), 438–444. (4)

Abdi-Jalebi, M.; Andaji-Garmaroudi, Z.; Cacovich, S.; Stavrakas, C.; Philippe, B.; Richter,

J. M.; Alsari, M.; Booker, E. P.; Hutter, E. M.; Pearson, A. J.; et al. Maximizing and Stabilizing Luminescence From Halide Perovskites with Potassium Passivation. Nature 2018, 555 (7697), 497–501. (5)

Wang, F.; Geng, W.; Zhou, Y.; Fang, H. H.; Tong, C.-J.; Loi, M. A.; Liu, L.-M.; Zhao, N.

Phenylalkylamine Passivation of Organolead Halide Perovskites Enabling High-Efficiency and Air-Stable Photovoltaic Cells. Adv. Mater. 2016, 28 (45), 9986–9992.



(6)

Page 18 of 22

Braly, I. L.; DeQuilettes, D. W.; Pazos-Outón, L. M.; Burke, S.; Ziffer, M. E.; Ginger, D.

S.; Hillhouse, H. W. Hybrid Perovskite Films Approaching the Radiative Limit with Over 90% Photoluminescence Quantum Efficiency. Nat. Photonics 2018, 12 (6), 355–361. (7)

Eperon, G. E.; Moerman, D.; Ginger, D. S. Anticorrelation Between Local

Photoluminescence and Photocurrent Suggests Variability in Contact to Active Layer in Perovskite Solar Cells. ACS Nano 2016, 10 (11), 10258–10266. (8)

Moerman, D.; Eperon, G. E.; Precht, J. T.; Ginger, D. S. Correlating Photoluminescence

Heterogeneity with Local Electronic Properties in Methylammonium Lead Tribromide Perovskite Thin Films. Chem. Mater. 2017, 29 (13), 5484–5492. (9)

Luo, D.; Yang, W.; Wang, Z.; Sadhanala, A.; Hu, Q.; Su, R.; Shivanna, R.; Trindade, G.

F.; Watts, J. F.; Xu, Z.; et al. Enhanced Photovoltage for Inverted Planar Heterojunction Perovskite Solar Cells. Science 2018, 360 (6396), 1442–1446. (10) Stolterfoht, M.; Caprioglio, P.; Wolff, C. M.; Márquez, J. A.; Nordmann, J.; Zhang, S.; Rothhart, D.; Hörmann, U.; Redinger, A.; Kegelmann, L et al. The Perovskite/Transport Layer Interfaces Dominate Non-Radiative Recombination in Efficient Perovskite Solar Cells. The Perovskite/Transport Layer Interfaces Dominate Non-Radiative Recombination in Efficient Perovskite Solar Cells. 2018, arXiv:1810.01333 [physics.app-ph]. (11) Xu, L.; Molaei Imenabadi, R.; Vandenberghe, W. G.; Hsu, J. W. P. Minimizing Performance Degradation Induced by Interfacial Recombination in Perovskite Solar Cells


Page 19 of 22 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

ACS Energy Letters

Through Tailoring of the Transport Layer Electronic Properties. APL Mater. 2018, 6 (3), 036104. (12) Wolff, C. M.; Zu, F.; Paulke, A.; Toro, L. P.; Koch, N.; Neher, D. Reduced InterfaceMediated Recombination for High Open-Circuit Voltages in CH3NH3PbI3 Solar Cells. Adv. Mater. 2017, 29 (28), 1700159. (13) Belisle, R. A.; Jain, P.; Prasanna, R.; Leijtens, T.; Mcgehee, M. D. Minimal Effect of the Hole-Transport Material Ionization Potential on the Open-Circuit Voltage of Perovskite Solar Cells. ACS Energy Lett. 2016, 1 (3), 556–560. (14) Crossland, E. J. W.; Noel, N.; Sivaram, V.; Leijtens, T.; Alexander-Webber, J. A.; Snaith, H. J. Mesoporous TiO2 Single Crystals Delivering Enhanced Mobility and Optoelectronic Device Performance. Nature 2013, 495 (7440), 215–219. (15) Christians, J. A.; Schulz, P.; Tinkham, J. S.; Schloemer, T. H.; Harvey, S. P.; Tremolet de Villers, B. J.; Sellinger, A.; Berry, J. J.; Luther, J. M. Tailored Interfaces of Unencapsulated Perovskite Solar Cells for >1,000 Hour Operational Stability. Nature Energy 2018, 3 (1), 68–74. (16) Dunlap-Shohl, W. A.; Daunis, T. B.; Wang, X.; Wang, J.; Zhang, B.; Barrera, D.; Yan, Y.; Hsu, J. W. P.; Mitzi, D. B. Room-Temperature Fabrication of a Delafossite CuCrO2 Hole Transport Layer for Perovskite Solar Cells. J Mater. Chem. A 2018, 6 (2), 469–477. (17) Black, L. E. New Perspectives on Surface Passivation: Understanding the Si-Al2O3 Interface; Springer: Switzerland, 2016.



Page 20 of 22

(18) Sproul, A. B. Dimensionless Solution of the Equation Describing the Effect of Surface Recombination on Carrier Decay in Semiconductors. J. Appl. Phys. 1994, 76 (5), 2851–2854. (19) Grant, N. E.; Niewelt, T.; Wilson, N. R.; Wheeler-Jones, E. C.; Bullock, J.; Al-Amin, M.; Schubert, M. C.; van Veen, A. C.; Javey, A.; Murphy, J. D. Superacid-Treated Silicon Surfaces: Extending the Limit of Carrier Lifetime for Photovoltaic Applications. IEEE J. Photovoltaics 2017, 7 (6), 1574–1583. (20) Woodall, J. M.; Hovel, H. J. High-Efficiency Ga1-xAlxAs Single - GaAs Solar Cells. Appl. Phys. Lett. 1972, 21 (8), 379–381. (21) Vermang, B.; Fjällström, V.; Pettersson, J.; Salomé, P.; Edoff, M. Development of Rear Surface Passivated Cu(In,Ga)Se2 Thin Film Solar Cells with Nano-Sized Local Rear Point Contacts. Sol. Energy Mater. Sol. Cells 2013, 117, 505–511. (22) Yang, Y.; Yan, Y.; Yang, M.; Choi, S.; Zhu, K.; Luther, J. M.; Beard, M. C. Low Surface Recombination Velocity in Solution-Grown CH3NH3PbBr3 Perovskite Single Crystal. Nat. Commun. 2015, 6 (1), 183. (23) Wu, B.; Zhou, Y.; Xing, G.; Xu, Q.; Garces, H. F.; Solanki, A.; Goh, T. W.; Padture, N. P.; Sum, T. C. Long Minority-Carrier Diffusion Length and Low Surface-Recombination Velocity in Inorganic Lead-Free CsSnI3 Perovskite Crystal for Solar Cells. Adv. Funct. Mater. 2017, 27 (7), 1604818.


Page 21 of 22 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

ACS Energy Letters

(24) Wen, X.; Chen, W.; Yang, J.; Ou, Q.; Yang, T.; Zhou, C.; Lin, H.; Wang, Z.; Zhang, Y.; Conibeer, G.; et al. Role of Surface Recombination in Halide Perovskite Nanoplatelets. ACS Appl. Mater. Interfaces 2018, 10 (37), 31586–31593. (25) Burgelman, M.; Nollet, P.; Degrave, S. Modelling Polycrystalline Semiconductor Solar Cells. Thin Solid Films 2000, 361-362, 527–532. (26) Burgelman, M.; Decock, K.; Khelifi, S.; Abass, A. Advanced Electrical Simulation of Thin Film Solar Cells. Thin Solid Films 2013, 535, 296–301. (27) Birkhold, S. T.; Precht, J. T.; Liu, H.; Giridharagopal, R.; Eperon, G. E.; Schmidt-Mende, L.; Li, X.; Ginger, D. S. Interplay of Mobile Ions and Injected Carriers Creates Recombination Centers in Metal Halide Perovskites Under Bias. ACS Energy Lett. 2018, 3 (6), 1279–1286. (28) Miller, O. D.; Yablonovitch, E.; Kurtz, S. R. Strong Internal and External Luminescence as Solar Cells Approach the Shockley-Queisser Limit. IEEE J. Photovoltaics 2012, 2 (3), 303– 311. (29) Alarousu, E.; El-Zohry, A. M.; Yin, J.; Zhumekenov, A. A.; Yang, C.; Alhabshi, E.; Gereige, I.; AlSaggaf, A.; Malko, A. V.; Bakr, O. M.; et al. Ultralong Radiative States in Hybrid Perovskite Crystals: Compositions for Submillimeter Diffusion Lengths. J. Phys. Chem. Lett. 2017, 8 (18), 4386–4390. (30) Saliba, M.; Correa-Baena, J.-P.; Wolff, C. M.; Stolterfoht, M.; Phung, N.; Albrecht, S.; Neher, D.; Abate, A. How to Make Over 20% Efficient Perovskite Solar Cells in Regular (N-I-P) and Inverted (P-I-N) Architectures. Chem. Mater. 2018, 30 (13), 4193–4201. 21 ACS Paragon Plus Environment


Page 22 of 22

(31) Zhou, H.; Chen, Q.; Li, G.; Luo, S.; Song, T. B.; Duan, H.-S.; Hong, Z.; You, J.; Liu, Y.; Yang, Y. Interface Engineering of Highly Efficient Perovskite Solar Cells. Science 2014, 345 (6196), 542–546. (32) Wong, K. K.; Fakharuddin, A.; Ehrenreich, P.; Deckert, T.; Abdi-Jalebi, M.; Friend, R. H.; Schmidt-Mende, L. Interface-Dependent Radiative and Nonradiative Recombination in Perovskite Solar Cells. J. Phys. Chem. C 2018, 122 (20), 10691–10698. (33) Wang, J.; Wang, N.; Jin, Y.; Si, J.; Tan, Z.-K.; Du, H.; Cheng, L.; Dai, X.; Bai, S.; He, H.; et al. Interfacial Control Toward Efficient and Low-Voltage Perovskite Light-Emitting Diodes. Adv. Mater. 2015, 27 (14), 2311–2316. (34) Chen, P.; Xiong, Z.; Wu, X.; Shao, M.; Ma, X.; Xiong, Z.-H.; Gao, C. Highly Efficient Perovskite Light-Emitting Diodes Incorporating Full Film Coverage and Bipolar Charge Injection. J. Phys. Chem. Lett. 2017, 8 (8), 1810–1818. (35) Kim, B. S.; Lee, J. Y. Engineering of Mixed Host for High External Quantum Efficiency Above 25% in Green Thermally Activated Delayed Fluorescence Device. Adv. Funct. Mater. 2014, 24 (25), 3970–3977. (36) Fu, W.; Wang, J.; Zuo, L.; Gao, K.; Liu, F.; Ginger, D. S.; Jen, A. K. Y. Two-Dimensional Perovskite Solar Cells with 14.1% Power Conversion Efficiency and 0.68% External Radiative Efficiency. ACS Energy Lett. 2018, 3, 2086–2093.


Reducing Surface Recombination Velocities at the ... - ACS Publications

Recommend Documents