Thermally Stable Perovskite Solar Cells by Systematic Molecular

Jan 10, 2019 - With metal halide perovskite solar cells (PSCs) now reaching device efficiencies >23%, more emphasis must now shift toward addressing t...
4 downloads 0 Views 840KB Size
Subscriber access provided by Access provided by University of Liverpool Library

Letter

Thermally Stable Perovskite Solar Cells by Systematic Molecular Design of the Hole-Transport Layer Tracy H. Schloemer, Timothy S. Gehan, Jeffrey A. Christians, Deborah G. Mitchell, Alex Dixon, Zhen Li, Kai Zhu, Joseph J. Berry, Joseph M. Luther, and Alan Sellinger ACS Energy Lett., Just Accepted Manuscript • DOI: 10.1021/acsenergylett.8b02431 • Publication Date (Web): 10 Jan 2019 Downloaded from http://pubs.acs.org on January 11, 2019

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 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

ACS Energy Letters

Thermally Stable Perovskite Solar Cells by Systematic Molecular Design of the Hole-Transport Layer T. H. Schloemer*1, Timothy S. Gehan*1,3, Jeffrey A. Christians4,7, Deborah G. Mitchell5, Alex Dixon6,8, Zhen Li4,9, Kai Zhu4, Joseph J. Berry3, Joseph M. Luther4, and Alan Sellinger1,2,4** Affiliations: 1Department of Chemistry, 2Materials Science Program, Colorado School of Mines, Golden, CO, USA. 3National Renewable Energy Laboratory, Materials Science Center, Golden, CO, USA. 4National Renewable Energy Laboratory, Chemistry and Nanoscience Center, Golden, CO, USA. 5Department of Chemistry, University of Denver, Denver, CO, USA. 6Renewable and Sustainable Energy Institute (RASEI), University of Colorado Boulder, Boulder, CO. 7Department of Engineering, Hope College, Holland, Michigan, USA. 8Laboratory of Organic Matter Physics, University of Nova Gorica, Vipavska 13, SI-5000 Nova Gorica, Slovenia. 9State Key Laboratory of Solidification Processing, Center for Nano Energy Materials, School of Materials Science and Engineering, Northwestern Polytechnical University and Shaanxi Joint Laboratory of Graphene (NPU), Xi’an, 710072, China. *These authors contributed equally. **Correspondence to: [email protected]

ACS Paragon Plus Environment

1

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 20

Abstract: With metal halide perovskite solar cells (PSC) now reaching device efficiencies >23%, more emphasis must now shift towards addressing their device stability. Recently, a triarylamine-based organic hole-transport material (HTM) doped with its oxidized salt analogue (EH44/EH44-ox) led to unencapsulated PSC with high stability in ambient conditions. Here we report criteria for triarylamine-based organic HTMs formulated with stable oxidized salts as hole-transport layer (HTL) for increased PSC thermal stability. The triarylamine-based dopants must contain at least two para-electron donating groups for radical cation stabilization to prevent impurity formation that leads to reduced PSC performance. The stability of unencapsulated devices prepared using these new HTMs stressed under constant load and illumination far outperform both EH44/EH44-ox and Li+-doped spiro-OMeTAD controls at 50 ºC. Furthermore, the ability to mix and match these dopants with a non-identical small-molecule-based HTL matrix broadens the design scope for highly stable and cost-effective PSC without sacrificing performance. Table of Content Graphic:

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

ACS Energy Letters

With rapid power conversion efficiency (PCE) improvements from 3.83%1 to 23.3%2 in less than a decade, perovskite solar cells (PSCs) have great potential to serve as a low-cost, renewable electricity source to meet increasing energy demands. Even so, devices durable enough for real world power generation is needed prior to commercialization, which is significantly impacted by hole-transport layer (HTL) formulation. Most organic-based HTLs for PSCs require dopants for improved charge-transport properties. The stateof-the-art organic hole-transport materials (HTMs) in PSCs, such as 2,2′,7,7′-tetrakis(N,N-di-pmethoxyphenylamine)-9,9′-spirobifluorene (spiro-OMeTAD) and poly(triarylamine) (PTAA), utilize bis(trifluoromethane)sulfonamide lithium salt (LiTFSI) as a dopant for improved fill factor (FF) and PCE.3–7 In the prototypical case of spiro-OMeTAD, LiTFSI contributes to the formation of spiro-OMeTAD radical cations in the presence of oxygen,8,9 which in turn promotes charge hopping10–12 and increases the hole mobility and conductivity of the HTL. However, the radical cation generation is typically inconsistent, as it is not directly related to dopant concentration and very dependent on atmospheric exposure conditions.8,13,14 While use of such dopants improves charge-transport properties to realize high PCE, the dopants can be detrimental to the overall stability of PSCs as the films need to be exposed to oxygen. Additionally, the hygroscopic Li+ cations have been shown to migrate through the entire device stack15 increasing moisture ingress.16,17 In contrast to the non-stoichiometric in situ radical cation generation achieved with LiTFSI doping, an HTM’s conductivity may also be increased by the addition of its oxidized salt analogue. Nguyen et al.18 combined spiro-OMeTAD with its oxidized salt doped analogue, 2,2′,7,7′-tetrakis(N,N-di-pmethoxyphenylamine)-9,9′-spirobifluorene di[bis(trifluoromethanesulfonyl)imide] (spiro(TFSI)2) and demonstrated improved HTL conductivity and PSC PCE without use of hygroscopic lithium salts. Moreover, this strategy allows for precise control over doping concentrations, eliminates the need for oxygen exposure, and is also far less susceptible to unwanted ion intercalation in the active layer. Leijtens et al.19,20 implemented this strategy with a low-cost, synthetically simple carbazole-cored HTM for comparable PCEs and improved hydrophobicity by using EH44 (9-(2-ethylhexyl)-N2,N2,N7,N7-tetrakis(4-methoxyphenyl)-9H-carbazole-2,7diamine) doped with its analogue TFSI- radical cation salt, EH44-ox (9-(2-ethylhexyl)-N2,N2,N7,N7-tetrakis(4methoxyphenyl)-9H-carbazole-2,7-diamine bis(trifluoromethanesulfonyl)imide). With further device engineering by Christians et al.21, unencapsulated PSCs utilizing EH44/EH44-ox as the HTL retained 88% average PCE over 1000 hours of continuous operation, dramatically outperforming LiTFSI-doped spiroOMeTAD with respect to stability and lifetime. This report currently represents the highest stability unencapsulated perovskite solar cell measured in ambient conditions. Despite these impressive findings with regard to PSC stability, PCEs suffer using this HTL motif as compared to devices utilizing Li+-doped spiro-OMeTAD. To this end, little is known about designing new HTMs that form stable cationic TFSI- salt doped analogues for efficient charge injection layers. Exchanging the HTM core has been investigated22,18,20,23,24, but modulating the triarylamine pendant substituents directly ACS Paragon Plus Environment

3

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 4 of 20

involved in charge-transport has not been reported. Herein we report the synthesis of three novel HTMs and their oxidized TFSI- salt analogues (HTM-ox) with different triarylamine pendant substituents to assess PSC performance and lifetime as a function of both HTM and oxidized salt dopant, as well as present design criteria for preparing HTMs with stable doped analogues for efficient PSCs. We find hole mobility and PSC performance, specifically fill factor (FF), are dependent upon dopant stability in conjunction with HTL matrix selection. With regard to stable dopant design, we find parasubstitution with two electron rich substituents on the oxidized salt’s triarylamine core25–28 is required for sufficient cation resonance stabilization. Importantly, HTMs that result in unstable radical cations can be doped with a stable organic HTM-ox variant (e.g., EH44-ox) and thus can still be utilized as Li+-free HTL matrices. When an unstable dopant is replaced with a stable dopant, we find increased HTL mobilities directly correlate with PSC FF and PCE. Stable, lithium-free HTL dopants can be designed independently of the HTL matrix, meaning that a hole-transport material candidate that yields an unstable oxidized salt dopant can still be a promising HTL matrix for efficient PSC. Stability of unencapsulated devices with new HTL formulations under constant resistive load and constant illumination outperform both state of the art EH44/EH44-ox21 and Li+doped spiro-OMeTAD at 50 ºC. This dopant interchange opens new avenues for Li+-free HTL design independent of the HTL matrix with the potential cost effective, mobile ion-free, stable PSC without compromising performance.

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

ACS Energy Letters

Chart 1: (a) HTM chemical structures for improved resonance stabilization of radical cations. Blue atoms represent para- or meta-electron donating groups with respect to triarylamine. The arrow highlights location of increased electron density based upon the electron donating group position. (b) Resonance contributor of EH44-ox depicting radical triarylamine delocalization via para-oxygen electron donation.

The three new HTMs demonstrated in this work were inspired by EH44 as outlined in Chart 1a but were modified in an effort to study the effect of electron donating group position with respect to triarylamine. In the context of this manuscript, HTM refers specifically to the different molecules under investigation while HTL refers to the films of HTMs which also contain functional additives and dopants. These HTMs, like many other reported HTMs, contain triarylamines.29–31 Oxidizing the triarylamine nitrogen to a radical cation can increase hole conductivity within the HTL.8 The substituents on each of the new HTMs were tailored with the intent to stabilize radical cations and thus promote charge hopping10,11 within the HTL while improving thermal properties by increasing HTM molecular weight.32,33 Each new HTM has an extended conjugation network affixed to the triarylamine, replacing one of the para-methoxy substituents in EH44, and together these HTMs have a range in electron donating ability with respect to triarylamine groups. The fluorene and 2-ethylcarbazole substituents in N2,N7-bis(9,9-dimethyl-9H-fluoren-2-yl)-9-(2-ethylhexyl)-N2,N7-bis(4-methoxyphenyl)-9HACS Paragon Plus Environment

5

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 6 of 20

carbazole-2,7-diamine, abbreviated as EHCz-MeFl, and N2,N7-bis(9-ethyl-9H-carbazol-2-yl)-9-(2-ethylhexyl)N2,N7-bis(4-methoxyphenyl)-9H-carbazole-2,7-diamine, abbreviated as EHCz-2EtCz, offer radical triarylamine stabilization via induction, while the latter has increased electron density for stabilization via induction due to the meta-nitrogen in the carbazole unit instead of a meta-carbon at the equivalent position in EHCz-MeFl. N2,N7-bis(9-ethyl-9H-carbazol-3-yl)-9-(2-ethylhexyl)-N2,N7-bis(4-methoxyphenyl)-9H-carbazole-2,7-diamine, abbreviated as EHCz-3EtCz, offers direct para-resonance stabilization for the radical triarylamine in addition to stabilization via induction. These subtle substituent modifications offer insight as to whether one parasubstituted methoxy group per radical triarylamine is sufficient to stabilize radical triarylamines in ambient atmosphere for stable PSC (Chart 1b). EHCz-MeFl, EHCz-2EtCz, and EHCz-3EtCz were synthesized via simple carbazole alkylations and two subsequent Buchwald-Hartwig aminations in good yields (Scheme S1). Full synthetic methods, yields, and characterization spectra (1H-NMR, 13C-NMR, Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), and UV-Vis absorption spectroscopy) are presented in the Supporting Information. Presently, cost is one of the biggest hurdles to widespread use of spiro-OMeTAD at the industrial scale. To understand how these new HTMs compare, the cost/mL of solution was calculated for the optimized formulation of each HTL. Per academic pricing rates,34 the $/mL of doped HTL solutions at optimized concentrations for device fabrication is approximately one quarter to one third the cost of doped spiroOMeTAD solution (Tables S1-S6). While extrapolation from such pricing estimates to terawatt scale PSC production is inherently problematic, this analysis does suggest that the $/mL of each new HTL formulation would most likely be lower-cost.

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

ACS Energy Letters

Figure 1: HTM and HTM-ox absorption spectra, 10-5 M in tetrahydrofuran (THF). HTM and HTM-ox spectra are normalized to λmax of each pristine HTM (EHCz-3EtCz and EHCz-3EtCz-ox at 376.2 nm (a), EHCz-2EtCz and EHCz-2EtCz-ox at 397.8 nm (b), and EHCz-MeFl and EHCz-MeFl-ox at 397.6 nm (c)). (d) Representative first derivative CW-EPR spectra for 1.5 mM HTM-ox in toluene.

To prepare the analogous oxidized salts (HTM-ox), a one-step simple room temperature reaction of each HTM with silver bis(trifluoromethanesulfonyl)imide (AgTFSI) afforded HTM-ox (HTM+TFSI-) in excellent yields (Scheme S1 and Supporting Information). Silver ions have been shown to be mobile throughout the device stack,35 so synthesis and purification to ensure HTM-ox is silver-free is a key component of stable HTL formulations (in contrast to in situ doping with AgTFSI36). As for purity characterization, common techniques (NMR, MALDI-TOF MS, etc.) are insufficient to discriminate between pristine HTM and HTM-ox18,20, UVVis absorption spectroscopy can be used because the methoxy-functionalized radical cation triarylamine gives three characteristic peaks at c.a. 500, 700, and 1100 nm, as seen in the spectra for each HTM-ox in Figure 1ac.8 ACS Paragon Plus Environment

7

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 8 of 20

In addition to observing this absorption signature, the radical cation in each HTM-ox salt was further characterized by continuous wave electron paramagnetic resonance spectroscopy (CW-EPR) in ambient atmospheric conditions (Figure 1d). The splitting pattern in a CW-EPR spectrum is dependent on the interaction between the electron density of the unpaired electron and neighboring nuclear spins. Because this is a hydrocarbon-based system, the primary spins available are from nitrogen (14N, I=1) and hydrogen (1H, I=1/2) as all other elements in the system have relatively low abundance of nuclear spins. The splitting patterns for the HTM-ox salts are similar due to the fact that these molecules have a similar ring system substructure, and therefore similar interactions between the unpaired electron and nitrogen and hydrogen nuclei. The thermal properties (decomposition onset, Td, and glass transition temperature, Tg) of each pristine HTM were measured and are summarized in Table 1 and Figures S1-S3, as the Tg can influence device stability at elevated temperatures. Per thermogravimetric analysis (TGA) in N2, the onset of thermal decomposition of EHCz-MeFl, EHCz-2EtCz, and EHCz-3EtCz are 433, 395, and 414 C respectively. Differential scanning calorimetry (DSC) data indicates that none of the new HTMs show a discernible melting point, but all have a glass transition (Tg) between 98-100 C, which is 30 C higher than EH44. While the Tg for spiro-OMeTAD is reported to be higher (124 C), crystallization from concentrated solutions at room temperature has been observed, suggesting that the Tg might be lower in practice.37 The higher Tg for this set of HTMs is attributed to increased molecular weight32,33,38 and replacement of anisole with more rigid fluorene and carbazole moieties.37 Table 1: Thermal properties of pristine HTM: decomposition temperature onset, (Td, onset) and glass transition temperature (Tg) Td, onset (C)a,b Tg (C)b,c Spiro-OMeTAD 449 124 EH44 349 69 EHCz-3EtCz 414 99 EHCz-2EtCz 395 98 EHCz-MeFl 433 100 aError is +/- 0.01 wt. % bThermal characterization for spiro-OMeTAD from Malinaskas et al.37 cerror is +/- 0.1 ˚C.

Cyclic voltammetric analysis of each pristine HTM shows good oxidation reversibility (Figure S4). Space charged limited current (SCLC) measurements were used to determine the hole mobility for each HTL formulation (Table 2, Figure S5).39 All films included tBP (tert-butylpyridine) to be device relevant and obtain a realistic representation of hole mobility with the additive.23,40,41 While the exact functions of tBP in the HTL are under investigation,40,42–45 we have previously shown that tBP incorporation is beneficial for PSC stability with EH44/EH44-ox HTL formulation.21 The hole mobilities of HTLs containing only the pristine organic small molecule HTMs are typically low, on the order of 10-5 cm2 V-1s-1. The hole mobility of spiro-OMeTAD, EH44, and EHCz-3EtCz increased by an order of magnitude when the dopants (either LiTFSI in the case of spiroOMeTAD or the analogous HTM-ox) were added to the HTL. Surprisingly, little to no change in hole mobility 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

ACS Energy Letters

is observed for EHCz-2EtCz and EHCz-MeFl with the addition of their analogous HTM-ox. In general, the HTLs with highest hole mobilities, EH44 and EHCz-3EtCz, are composed of HTMs and dopants with 2 paraelectron donating groups with respect to the triarylamine. Table 2: HTL doped with analogous HTM-ox hole mobility values measured by space charge limited current (SCLC). HTM

Hole Mobility (10-4 cm2 V-1s-1) Pristine

Doped with Analogous HTM-ox

Spiro-OMeTAD

0.20

12.52 (LiTFSI)

EH44

1.30

18.95

EHCz-3EtCz

0.61

14.93

EHCz-2EtCz

0.46

1.01

EHCz-MeFl

0.22

0.17

Devices were tested in an ITO/PEDOT:PSS/HTL/Ag architecture. HTM layers were between 200-300nm thick. All samples contained tBP. A dielectric constant of 3 was used, common for organic semiconductors.39

Figure 2: Summary of frontier molecular orbital energetics for each HTM. Optical bandgap (Egap) were derived from absorbance onset of pristine HTM, EHOMO were derived from photoelectron spectroscopy in air (PESA) of thin films, and ELUMO were calculated using Egap,onset + EHOMO=ELUMO. The colors represent relative energetics as follows: ELUMO and HTM EHOMO in black, doped HTL EHOMO in red (spiro-OMeTAD doped with LiTFSI, all other HTMs doped with 14 wt% of analogous HTM-ox), and HTM-ox ESOMO in blue. All PESA samples include tert-butylpyridine (tBP).

Frontier molecular orbital alignment of the HTL with the perovskite active layer is important for efficient hole extraction and electron blocking. The UV-Vis absorption (Figure 1a-c) onsets of 425 nm for the three new HTMs correspond to a roughly 3 eV bandgap. The UV-Vis absorption spectra indicates the absorption onset for each new HTM are slightly red shifted from EH44; however, all three absorb less of the ACS Paragon Plus Environment

9

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 10 of 20

visible spectrum than spiro-OMeTAD.20 The highest occupied molecular orbital energetics (EHOMO) for pure HTM and HTM-ox films, as well as films of HTMs doped with HTM-ox, were measured by Photoelectron Spectroscopy in Air (PESA) and summarized in Figure 2, Table S7, and Figure S6. Once again, all films included tBP to be device relevant and obtain a realistic representation of energetics23,40,41 with the additive. The EHOMO of new pristine HTMs becomes less negative as electron donating capability increases (electron donating ability EHCz-3EtCz>EHCz-2EtCz>EH44=spiro-OMeTAD>EHCz-MeFl). The singly occupied molecular orbital energetics of all three new HTM-ox salt (ESOMO) are nearly equivalent to each other and less negative than that of EH44-ox. The EHOMO of all of the new doped HTLs in this study have a slightly higher energy (are less negative) than doped spiro-OMeTAD and are energetically aligned to easily accept holes from the perovskite active layer. Of the new HTMs, doped EHCz-3EtCz has the highest EHOMO, at -5.02 eV, which is consistent with paranitrogen in the carbazole with respect to triarylamine as described previously, and is consistent with a linear combination of energy states for 14 wt% doped film.18 In all cases, the lowest unoccupied molecular orbital (ELUMO) alignment for each HTL is sufficient for blocking electron injection from the conduction band of the perovskite active layer. To summarize, the EHOMO and ELUMO of doped HTLs are well aligned with the perovskite active layer in regards to hole transfer and electron blocking respectively. To investigate how each HTM affects the photovoltaic performance, devices were prepared with the following architecture: ITO/SnO2/Perovskite/HTL/MoOx/Al. The perovskite active layer (PAL) with a precursor solution stoichiometry of (FA0.79MA0.16Cs0.05)0.97Pb(I0.84Br0.16)2.97 (FAMACs) was used due to its improved operational stability compared to other PALs.46 This architecture was selected to maximize device stability without compromising device performance.21 Device fabrication methods are outlined in the Supporting Information. While the charge-transport and perovskite active layer were deposited via spincoating in this study, future work will incorporate more scalable techniques such as blade-coating.47 Devices using each new HTM were optimized by tuning the HTM concentration, spin rate, and HTM-ox concentration (Tables S8-S12). The optimized device performance for each HTL is shown in Table 3. Devices that were prepared using EHCz-3EtCz with EHCz-3EtCz-ox have comparable PCE to Li+-doped spiroOMeTAD. Devices prepared with EHCz-3EtCz/EHCz-3EtCz-ox display slightly increased PCE as compared to devices prepared with EH44/EH44-ox due to improved Voc. HTM frontier molecular orbital energetics do not correlate with Voc., which has been previously reported.48,49 This points to the importance of HTL matrix selection for PSC performance. Conversely, devices that were prepared using EHCz-2EtCz with EHCz-2EtCzox and EHCz-MeFl with EHCz-MeFl-ox as the HTL show much lower PCE. The majority of this diminished device performance is derived from lower FF within the EHCz-2EtCz and EHCz-MeFl devices.

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

ACS Energy Letters

Table 3: Average device performance using HTMs doped with analogous HTM-ox. HTL Scan Direction VOC (V) JSC (mA/cm2) Rev 1.118  0.004 Fwd 1.060  0.018 EH44 +EH44-ox Rev 1.077  0.003 Fwd 1.045  0.005 EHCz-3EtCz + EHCz-3EtCz-ox Rev 1.098  0.004 Fwd 1.073  0.004 EHCz-2EtCz + EHCz-2EtCz-ox Rev 1.055  0.005 Fwd 1.043  0.006 EHCz-MeFl + EHCz-MeFl-ox Rev 1.033  0.008 Fwd 1.017  0.010 The data presented is an average of 5-12 devices  the standard deviation. Spiro-OMeTAD + LiTFSI

FF

21.62  0.34 21.42  0.37 21.84  0.04 21.67  0.05 21.58  0.05 21.49  0.05 18.39  0.61 18.04  0.64 15.71  0.45 14.46  0.48

0.71  0.01 0.66  0.01 0.69  0.01 0.68  0.01 0.69  0.01 0.68  0.01 0.26  0.01 0.24  0.01 0.25  0.01 0.21  0.01

PCE (%) 17.23  0.51 14.90  0.55 16.29  0.22 15.28  0.25 16.33 0.35 15.76  0.21 5.10  0.45 4.56  0.43 3.98  0.20 3.04  0.20

Low FF for an HTL, as seen in EHCz-2EtCz and EHCz-MeFl, is likely due to low HTL hole mobility (Table 2). To determine whether the HTM matrix or HTM-ox is primarily responsible for poor HTL hole mobility in the case of EHCz-2EtCz and EHCz-MeFl, we performed SCLC measurements where the corresponding HTM-ox salts (viz., EHCz-2EtCz-ox or EHCz-MeFl-ox) were replaced with EH44-ox (Table 4). Indeed, when EHCz-2EtCz or EHCz-MeFl-ox were doped with EH44-ox the hole mobility increased in line with what was observed for the other HTL systems (e.g., EH44 doped with EH44-ox). By exchange of an ineffective dopant (viz., EHCz-2EtCz-ox, EHCz-MeFl-ox) with an effective dopant (viz., EH44-ox), the hole mobility increased almost by an order of magnitude for EHCz-2EtCz and EHCz-MeFl-based HTL. Whereas, in EHCz-3EtCz-based HTL, exchanging EHCz-3EtCz-ox for EH44-ox led to almost no change in hole mobility. These results indicate that exchanging HTM-ox dopants among these HTMs is a viable strategy to improved HTL mobility. Table 4: HTL doped with EH44-ox hole mobility values measured by space charge limited current (SCLC). HTL doped with analogous HTM-ox hole mobility values are presented again for clarity of comparison. HTM Hole Mobility (10-4 cm2 V-1s-1) Doped with Analogous HTM-ox

Doped with EH44-ox

EH44

18.95

-

EHCz-3EtCz

14.93

12.08

EHCz-2EtCz

1.01

8.54

EHCz-MeFl

0.17

2.44

Devices were tested in an ITO/PEDOT:PSS/HTL/Ag architecture. HTM layers were between 200-300nm thick. All samples contained tBP. A dielectric constant of 3 was used, which is commonly used for organic semiconductors.39

The low mobility of EHCz-2EtCz- and EHCz-MeFl-based HTLs can be attributed to a number of factors, such as the energetic difference of the frontier orbitals of the pristine and oxidized molecules.50,51 Since

ACS Paragon Plus Environment

11

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 12 of 20

the E between EHOMO and ESOMO is relatively consistent from system to system (Table S7), energetic mismatch between HTM and HTM-ox likely is not responsible for observed performance differences. The consistent E between EHOMO and ESOMO for each HTL formulation suggests that EHCz-2EtCz-ox and EHCz-MeFl-ox simply provide fewer radical cations for charge hopping as compared to EHCz-3EtCz-ox and EH44-ox and can thus not effectively serve as dopants. While it has been shown that each HTM-ox molecule can provide up to one radical cation upon equilibration in solution18,20, CW-EPR spectroscopy was used to assess HTM-ox stability by quantifying the actual radical cation concentration of each new HTM-ox in solution and comparing to the concentration by mass assuming one radical cation is produced per molecule.9 Using 4-hydroxy-2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPOL) as a standard, EH44-ox and EHCz-3EtCz-ox were determined to maintain 88% and 57% of their predicted radical cation concentration in solution, while EHCz-2EtCz-ox and EHCz-MeFl-ox maintained just 21% and 28%, respectively, of their predicted radical cation concentration (Table 5, Table S13). The CW-EPR data indicates that EH44-ox and EHCz-3EtCz-ox contain more stable radicals than EHCz-2EtCz-ox and EHCz-MeFl-ox. The loss of radical triarylamine absorption can be rationalized as follows. Since the triarylamine reduction rate in ambient conditions is similar among all HTM-ox molecules over 3 hours (Table 5, Figure S7, Table S14-15), there is likely another mechanism responsible for the relatively low radical cation concentrations observed in EHCz-MeFl-ox and EHCz-2EtCz-ox solutions. The low radical cation concentrations are observed without the addition of tBP, so pyridination is precluded as a reduction pathway.23 Therefore, we attribute quenching of the unstable radicals in EHCz-2EtCz-ox and EHCz-MeFl-ox solutions to dopant dimerization25–27 (Figure S8). Indeed, MALDI-TOF MS of EHCz-MeFl-ox and EHCz-2EtCz-ox shows an increased propensity for dimerization of these HTM-ox salts as compared to EH44-ox and EHCz-3EtCz-ox. (Figures S9-S12). Table 5: CW-EPR spectroscopy summary for radical cation quantification against TEMPOL standard in toluene. Actual Radical Cation Concentration (mM) 1.04  0.01

Ratio Actual Radical Cation Concentration to Concentration by Mass at t=0 0.88

EHCz-3EtCz-ox

0.79  0.02

0.57

93

EHCz-2EtCz-ox

0.22  0.01

0.21

83

EHCz-MeFl-ox

0.28  0.01

0.28

HTM-ox EH44-ox

% Radical Cation Absorption Retained at t=3 hr 93

92 The “Actual Radical Cation Concentration” data is presented as an average concentration  the standard deviation.

To understand the stability of the HTM-ox in presence of the HTL matrix, we measured the UV-Vis absorption spectra of the HTL solutions containing each HTM mixed with its own HTM-ox (without tBP), as shown in Figure 3d-f. The characteristic peak for a radical cation on a triarylamine (~500 nm) can be seen for all HTM-ox salts, as shown in Figure 1, when they are measured pristine in solution. When solutions of EHCz12 ACS Paragon Plus Environment

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

ACS Energy Letters

2EtCz with EHCz-2EtCz-ox and EHCz-MeFl with EHCz-MeFl-ox are prepared, the characteristic radical cation peak disappears; this indicates that the unstable radical has been quenched under ambient conditions (Figure 3d-f, Figure S14-S17). However, when EHCz-3EtCz is mixed with EHCz-3EtCz-ox, the characteristic radical cation peak remains, further validating that the EHCz-3EtCz-ox species is more stable than the EHCz2EtCz-ox or EHCz-MeFl-ox. Taken together, when mixing EHCz-2EtCz-ox or EHCz-MeFl-ox into the HTL, the radical cation of the HTM-ox species is unstable and quenches, lowering the hole mobility of the HTL which prevents efficient extraction of the holes from the active layer, decreasing the FF, and thereby decreasing the overall PCE of these devices. On the other hand, when the HTM-ox species is stable, as is the case for EH44-ox and EHCz-3EtCzox, high FF and PCE devices can be obtained. In light of the dopant stability data presented, we assessed PSC performance utilizing EH44-ox as universal dopant for all HTLs, as summarized in Figure 3 and Table 6. Stabilized power output data of optimized PSCs is presented in Figure S13.52 When EHCz-2EtCz was mixed with EH44-ox, a large jump in PCE from 5.10% (with EHCz-2EtCz-ox) to 15.63% (with EH44-ox) was observed. This improvement is primarily the result of an increase in the FF, with a less significant increase in VOC and JSC. A similar improvement was observed for EHCz-MeFl when mixed with EH44-ox (PCE = 14.52%) compared to when it was mixed with EHCz-MeFl-ox (PCE = 3.98%), again mainly attributable to increased FF. Conversely, there was minimal change in PCE of EHCz-3EtCz when mixed with EH44-ox (16.77%) compared to when it is mixed with EHCz-3EtCz-ox (16.33%). Indeed, these device-level results are well-correlated with the hole mobility results shown in Table 4 and the retention of EH44-ox radical cation species in solution, highlighted in Figure 3d- 3f and Figures S15-S18. Notably, EHOMO of each HTL doped with their analogous HTM-ox salts or with EH44-ox are nearly identical, and there is an increase in E between HTM EHOMO and HTM-ox ESOMO with EH44-ox as dopant as compared to analogous HTM-ox (Table S7), so we can conclude that the increased E between HTM and EH44-ox is not high enough to hinder hole mobility.

ACS Paragon Plus Environment

13

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 14 of 20

Figure 3: Representative reverse scan J-V curves interchanging analogous HTM-ox with EH44-ox (a)-(c) and corresponding absorption spectra of doped solutions (d)-(f). Absorbance is normalized to 500 nm. Table 6: Average device performance using HTLs doped with EH44-ox. Scan VOC Direction (V) Rev 1.118  0.004 Fwd 1.060  0.018 EH44 + EH44-ox Rev 1.077  0.003 Fwd 1.045  0.005 EHCz-3EtCz + EH44-ox Rev 1.093  0.004 Fwd 1.066  0.004 EHCz-2EtCz + EH44-ox Rev 1.092  0.011 Fwd 1.052  0.015 EHCz-MeFl + EH44-ox Rev 1.058  0.014 Fwd 1.021  0.021 The data is presented as the average of 8-12 devices  the standard deviation. Spiro + LiTFSI (Control)

JSC (mA/cm2)

FF

PCE (%)

21.62  0.34 21.42  0.37 21.84  0.04 21.67  0.05 21.84  0.05 21.74  0.05 21.59  0.09 21.44  0.08 21.58  0.10 21.06  0.17

0.71  0.01 0.66  0.01 0.69  0.01 0.68  0.01 0.70  0.01 0.69  0.01 0.66  0.01 0.65  0.01 0.64  0.02 0.58  0.02

17.23  0.51 14.90  0.55 16.29  0.22 15.28  0.25 16.77  0.35 16.09  0.21 15.63  0.42 14.76  0.48 14.52  0.56 12.44  0.77

These device results corroborate the hypothesis that stabilizing the HTM-ox radical cation species in HTLs is essential for preparing efficient Li+-free PSCs. Our results indicate that in triarylamine based HTMs, which account for a majority of HTMs developed for perovskite based devices,29–31 at least two electron donating groups in the para position with respect to the triarylamine are required to stabilize the doped radical cation species and prevent quenching. The series of HTMs presented in this work have side groups on the triarylamine with a range of electron donating ability. EHCz-3EtCz has the strongest electron donating ability due to its carbazole nitrogen that is para to the triarylamine nitrogen. Whereas, EHCz-2EtCz will have a weaker electron donating ability because its carbazole nitrogen is meta to the triarylamine nitrogen. Finally, EHCzMeFl should have the weakest electron donating ability because the fluorene has minimal electron donating 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

ACS Energy Letters

properties. This correlates with the PESA data, as EHCz-3EtCz has the highest (less negative) EHOMO and EHCz-MeFl has the lowest (more negative) EHOMO, since increasing the donating character within the molecule should increase the HOMO energy level. Without sufficient resonance stabilization, the radical will likely find alternative pathways for stabilization and quench, and thus not be an effective HTL dopant. Ultimately, with use of a stable Li+-free dopant, all HTL matrices in this study allow for excellent device-level performance as compared to the devices with Li+- or EH44-ox-doped (Figure S14) spiro-OMeTAD as the HTL.

Figure 4: The device PCE over time for devices with various HTLs tested at an elevated temperature of 50C under constant illumination, constant load, and unencapsulated and exposed to laboratory atmosphere.  = Spiro-OMeTAD,  = EH44,  = EHCz3EtCz + EH44-ox,  = EHCz-2EtCz + EH44-ox,  = EHCz-MeFl + EH44-ox.

Christians et al. replaced spiro-OMeTAD with a Li+-free HTL, EH44/EH44-ox, and demonstrated that it significantly increases the device stability with unencapsulated devices retained 88% PCE over 1000 hours of continuous operation at 30C.21 When devices were tested at elevated temperatures, devices with EH44/EH44ox as HTL were as unstable as were devices prepared with spiro-OMeTAD/LiTFSI,21 likely due to EH44’s low Tg of 69C (Table 1). The three new HTMs presented in this work have a much higher Tg, indicating the devices should be much more thermally stable. To test this, PSCs were prepared with the architecture of ITO/SnO2/FAMACs/HTL/MoOx/Al using each new HTL, as well as spiro-OMeTAD/LiTFSI and EH44/EH44ox as controls. The devices were unencapsulated, thus exposed to ambient oxygen and humidity, under a constant resistive load and constant illumination from a sulfur plasma lamp over 500 hours at 50C. These devices were tested periodically for 500 hours as seen in Figures 4 and S19-S22. Within the first 100 hours EH44/EH44-ox, EHCz-3EtCz/EH44-ox, EHCz-2EtCz/EH44-ox, and EHCz-MeFl/EH44-ox behave very similarly and show an initial burn-in followed by a leveling out of performance. The initial drop in performance ACS Paragon Plus Environment

15

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 16 of 20

correlates with a small drop in Voc as well as a drop in FF (Figures S21 and S22). Whereas, devices prepared with spiro-OMeTAD/LiTFSI show rapid degradation even within the first 100 hours due to a significant initial drop in FF, as well as a constant decrease in Jsc . After the first 100 hours, devices prepared with EH44/EH44-ox start to degrade and correlate with a consistent drop in Jsc. Conversely, devices prepared with each of the new HTLs (EHCz-3EtCz, EHCz-2EtCz, EHCz-MeFl) are very stable from 100 to 500 hours. The Jsc is very consistent over the 500 hours for each of the new HTMs indicating they are a better barrier for the perovskite active layer and they are much more thermally stable. Currently, experiments are underway to understand the extent of the thermal stability of these HTLs. Most HTMs developed for perovskite photovoltaics rely on the use of LiTFSI to oxidize the triarylamine, even though Li+ has been shown to make the device less stable.15,17,21 Further, designing a device architecture that is stable at elevated temperatures which solar cells are routinely exposed to has been very challenging. In this paper, we prepared three new HTMs and their respective HTM-ox salts to elucidate design criteria for Li+-free triarylamine-based oxidized salts for dopants for improved HTL hole mobilities, PSC performance, and device stability. With CW-EPR and UV-Vis absorption spectroscopy, we demonstrate that at least two para electron donating groups are required for the preparation of stable HTM-ox molecules. Stabilizing the HTM-ox is required for obtaining HTLs with high hole mobilities, that in turn leads to optimized device performance. We also displayed the versatility in this approach by mixing HTMs that are known to not form stable oxidized species with a stable HTM-ox salt, EH44-ox, to improve HTL charge-transport properties, and in turn, PSC device performance. This allows for a wider tuneability of this system than previously appreciated. Moreover, HTMs that previously had poor performance due to their unstable oxidized species, like EHCz-2EtCz and EHCz-MeFl, but display promising characteristics (e.g., low synthetic cost, high thermal stability, etc), may now be effectively doped using this approach. With this strategy, PSC utilizing EHCz3EtCz-based HTL outperform state-of-the-art Li+-free PSC with EH44-based HTL, and demonstrate that incorporation of stable, Li+-free doped HTL does not always result in PSC performance penalty. Unencaspulated PSCs prepared with each new HTL formulation outperform devices with state of the art EH44/EH44-ox and Li+-doped spiro-OMeTAD when subjected to elevated temperature, constant illumination, and constant load. Further mechanistic studies of degradation processes are ongoing. We anticipate that the general design rules outlined are extendible to other HTM systems beyond the triarylamine HTMs studies, and that this versatile strategy will help to broaden Li+-free HTL design for the realization of inexpensive, efficient, and stable electricity generation with PSCs. Conflicts of Interest: There are no conflicts to declare. Electronic Supplementary Information (ESI) available: Synthesis and Characterization of all HTM and HTM-ox, PSC Fabrication methods, 500 MHz 1H-NMR, 500 MHz 13C-NMR, MALDI-TOF, Cyclic Voltammetry, Cost Analysis, Photoelectron Spectroscopy in Air, TGA curves, DSC curves, cyclic voltammetry, 16 ACS Paragon Plus Environment

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

ACS Energy Letters

cost analysis, PSC device optimization, HTL hole mobilities, CW-EPR of HTM-ox, Time Dependent UV-Vis Absorption Spectra of HTM-ox, PSC Stabilized Power Output, UV-Vis absorption spectra of HTM/HTM-ox. Acknowledgements: This work was authored by in part Alliance for Sustainable Energy, LLC, the manager and operator of the National Renewable Energy Laboratory for the U.S. Department of Energy (DOE) under Contract No. DEAC36-08GO28308. Funding was provided by U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Solar Energy Technologies Office under the Hybrid Perovskite Solar Cell Program and based upon work under the Agreement Number DE-EE0008174. J.A.C. was supported by the DOE Office of Energy Efficiency and Renewable Energy (EERE) Postdoctoral Research Award through SETO under DOE contract number DE-SC00014664. T.H.S. acknowledges the Department of Chemistry at the Colorado School of Mines for financial support through teaching assistantships and a Graduate Research Fellowship award. We thank Chuck Hitzman, Director, Surface Analysis Lab, Stanford Nano Shared Facility, Stanford University for providing the photoelectron spectroscopy in air (PESA) data described in this work, Gareth and Sandra Eaton from the University of Denver for thoughtful CW-EPR discussions, Al Hicks and Don Gwinner from NREL for the table of contents graphic, and Francis Bélanger from PCAS for supplying 2,7-dibromocarbazole. References: (1) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Organometal Halide Perovskites as Visible-Light Sensitizers for Photovoltaic Cells. J. Am. Chem. Soc. 2009, 131 (17), 6050–6051. https://doi.org/10.1021/ja809598r. (2) NREL Photovoltaic Efficiency Chart. https://www.nrel.gov/pv/assets/images/efficiency-chart.png. (3) Ko, Y.; Kim, Y.; Lee, C.; Kim, Y.; Jun, Y. Investigation of Hole-Transporting Poly(Triarylamine) on Aggregation and Charge Transport for Hysteresisless Scalable Planar Perovskite Solar Cells. ACS Appl. Mater. Interfaces 2018, 10 (14), 11633–11641. https://doi.org/10.1021/acsami.7b18745. (4) Hawash, Z.; Ono, L. K.; Qi, Y. Recent Advances in Spiro-MeOTAD Hole Transport Material and Its Applications in Organic-Inorganic Halide Perovskite Solar Cells. Adv. Mater. Interfaces 2018, 5 (1), 1700623. https://doi.org/10.1002/admi.201700623. (5) Jeon, N. J.; Na, H.; Jung, E. H.; Yang, T.-Y.; Lee, Y. G.; Kim, G.; Shin, H.-W.; Seok, S. I.; Lee, J.; Seo, J. A Fluorene-Terminated Hole-Transporting Material for Highly Efficient and Stable Perovskite Solar Cells. Nat. Energy 2018, 3 (8), 682–689. https://doi.org/10.1038/s41560-018-0200-6. (6) Luo, J.; Xia, J.; Yang, H.; Chen, L.; Wan, Z.; Han, F.; Malik, H. A.; Zhu, X.; Jia, C. Toward High-Efficiency, Hysteresis-Less, Stable Perovskite Solar Cells: Unusual Doping of a Hole-Transporting Material Using a Fluorine-Containing Hydrophobic Lewis Acid. Energy Environ. Sci. 2018, 11 (8), 2035–2045. https://doi.org/10.1039/C8EE00036K. (7) Hou, Y.; Du, X.; Scheiner, S.; McMeekin, D. P.; Wang, Z.; Li, N.; Killian, M. S.; Chen, H.; Richter, M.; Levchuk, I.; et al. A Generic Interface to Reduce the Efficiency-Stability-Cost Gap of Perovskite Solar Cells. Science 2017, 358 (6367), 1192–1197. https://doi.org/10.1126/science.aao5561. (8) Abate, A.; Leijtens, T.; Pathak, S.; Teuscher, J.; Avolio, R.; E. Errico, M.; Kirkpatrik, J.; M. Ball, J.; Docampo, P.; McPherson, I.; et al. Lithium Salts as “Redox Active” p-Type Dopants for Organic Semiconductors and Their Impact in Solid-State Dye -Sensitized Solar Cells. Phys. Chem. Chem. Phys. 2013, 15 (7), 2572–2579. https://doi.org/10.1039/C2CP44397J. (9) Namatame, M.; Yabusaki, M.; Watanabe, T.; Ogomi, Y.; Hayase, S.; Marumoto, K. Direct Observation of Dramatically Enhanced Hole Formation in a Perovskite-Solar-Cell Material Spiro-OMeTAD by Li-TFSI Doping. Appl. Phys. Lett. 2017, 110 (12), 123904. https://doi.org/10.1063/1.4977789. (10) Walzer, K.; Maennig, B.; Pfeiffer, M.; Leo, K. Highly Efficient Organic Devices Based on Electrically Doped Transport Layers. Chem. Rev. 2007, 107 (4), 1233–1271. https://doi.org/10.1021/cr050156n.

ACS Paragon Plus Environment

17

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 18 of 20

(11) Lüssem B.; Riede M.; Leo K. Doping of Organic Semiconductors. Phys. Status Solidi A 2012, 210 (1), 9–43. https://doi.org/10.1002/pssa.201228310. (12) Lüssem, B.; Keum, C.-M.; Kasemann, D.; Naab, B.; Bao, Z.; Leo, K. Doped Organic Transistors. Chem. Rev. 2016, 116 (22), 13714–13751. https://doi.org/10.1021/acs.chemrev.6b00329. (13) Snaith, H. J.; Grätzel, M. Enhanced Charge Mobility in a Molecular Hole Transporter via Addition of Redox Inactive Ionic Dopant: Implication to Dye-Sensitized Solar Cells. Appl. Phys. Lett. 2006, 89 (26), 262114. https://doi.org/10.1063/1.2424552. (14) Schölin, R.; Karlsson, M. H.; Eriksson, S. K.; Siegbahn, H.; Johansson, E. M. J.; Rensmo, H. Energy Level Shifts in Spiro-OMeTAD Molecular Thin Films When Adding Li-TFSI. J. Phys. Chem. C 2012, 116 (50), 26300–26305. https://doi.org/10.1021/jp306433g. (15) Li, Z.; Xiao, C.; Yang, Y.; Harvey, S. P.; Kim, D. H.; Christians, J. A.; Yang, M.; Schulz, P.; Nanayakkara, S. U.; Jiang, C.-S.; et al. Extrinsic Ion Migration in Perovskite Solar Cells. Energy Env. Sci 2017, 10, 1234–1242. https://doi.org/10.1039/C7EE00358G. (16) Hawash, Z.; Ono, L. K.; Raga, S. R.; Lee, M. V.; Qi, Y. Air-Exposure Induced Dopant Redistribution and Energy Level Shifts in Spin-Coated Spiro-MeOTAD Films. Chem. Mater. 2015, 27 (2), 562–569. https://doi.org/10.1021/cm504022q. (17) Christians, J. A.; Miranda Herrera, P. A.; Kamat, P. V. Transformation of the Excited State and Photovoltaic Efficiency of CH3NH3PbI3 Perovskite upon Controlled Exposure to Humidified Air. J. Am. Chem. Soc. 2015, 137 (4), 1530–1538. https://doi.org/10.1021/ja511132a. (18) Nguyen, W. H.; Bailie, C. D.; Unger, E. L.; McGehee, M. D. Enhancing the Hole-Conductivity of SpiroOMeTAD without Oxygen or Lithium Salts by Using Spiro(TFSI)2 in Perovskite and Dye-Sensitized Solar Cells. J. Am. Chem. Soc. 2014, 136 (31), 10996–11001. https://doi.org/10.1021/ja504539w. (19) Snaith, H., Leijtens, T., Abate, A., Sellinger, A. Organic Semiconductor Doping Process. US 9,818,944 B4, November 14, 2017. (20) Leijtens, T.; Giovenzana, T.; Habisreutinger, S. N.; Tinkham, J. S.; Noel, N. K.; Kamino, B. A.; Sadoughi, G.; Sellinger, A.; Snaith, H. J. Hydrophobic Organic Hole Transporters for Improved Moisture Resistance in Metal Halide Perovskite Solar Cells. ACS Appl. Mater. Interfaces 2016, 8 (9), 5981–5989. https://doi.org/10.1021/acsami.5b10093. (21) Christians, J. A.; Schulz, P.; Tinkham, J. S.; Schloemer, T. H.; Harvey, S. P.; Villers, B. J. T. de; Sellinger, A.; Berry, J. J.; Luther, J. M. Tailored Interfaces of Unencapsulated Perovskite Solar Cells for >1,000 Hour Operational Stability. Nat. Energy 2018, 3 (1), 68. https://doi.org/10.1038/s41560-017-0067-y. (22) Hsieh, B.; Li, X.C.; Sellinger, A. Semiconducting Hole Injection Materials for Organic Light Emitting Devices. U.S. Patent No. 6,830,830, December 4, 2004. (23) Kasparavicius, E.; Magomedov, A.; Malinauskas, T.; Getautis, V. Long-Term Stability of the Oxidized Hole Transporting Materials Used in Perovskite Solar Cells. Chem. - Eur. J. 2018, 24 (99), 9910–9918. https://doi.org/10.1002/chem.201801441. (24) Qu, J.; Jiang, X.; Yu, Z.; Lai, J.; Zhao, Y.; Hu, M.; Yang, X.; Sun, L. Improved Performance and Air Stability of Perovskite Solar Cells Based on Low-Cost Organic Hole-Transporting Material X60 by Incorporating Its Dicationic Salt. Sci. China Chem. 2018, 61 (2), 172–179. https://doi.org/10.1007/s11426-017-9141-9. (25) Seo, E. T.; Nelson, R. F.; Fritsch, J. M.; Marcoux, L. S.; Leedy, D. W.; Adams, R. N. Anodic Oxidation Pathways of Aromatic Amines. Electrochemical and Electron Paramagnetic Resonance Studies. J. Am. Chem. Soc. 1966, 88 (15), 3498–3503. https://doi.org/10.1021/ja00967a006. (26) Nelson, R. R.; Adams, R. N. Anodic Oxidation Pathways of Substituted Triphenylamines. II. Quantitative Studies of Benzidine Formation. J. Am. Chem. Soc. 1968, 90 (15), 3925–3930. https://doi.org/10.1021/ja01017a004. (27) Ambrose, J. F.; Nelson, R. F. Anodic Oxidation Pathways of Carbazoles. J. Electrochem. Soc. 1968, 115 (11), 1159. https://doi.org/10.1149/1.2410929. 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

ACS Energy Letters

(28) Ambrose, J. F. Electrochemical and Spectroscopic Properties of Cation Radicals. J. Electrochem. Soc. 1975, 122 (7), 876. https://doi.org/10.1149/1.2134365. (29) Teh, C. H.; Daik, R.; Lim, E. L.; Yap, C. C.; Ibrahim, M. A.; Ludin, N. A.; Sopian, K.; Mat Teridi, M. A. A Review of Organic Small Molecule-Based Hole-Transporting Materials for Meso-Structured Organic– Inorganic Perovskite Solar Cells. J Mater Chem A 2016, 4 (41), 15788–15822. https://doi.org/10.1039/C6TA06987H. (30) Wang, J.; Liu, K.; Ma, L.; Zhan, X. Triarylamine: Versatile Platform for Organic, Dye-Sensitized, and Perovskite Solar Cells. Chem. Rev. 2016, 116 (23), 14675–14725. https://doi.org/10.1021/acs.chemrev.6b00432. (31) Bakr, Z. H.; Wali, Q.; Fakharuddin, A.; Schmidt-Mende, L.; Brown, T. M.; Jose, R. Advances in Hole Transport Materials Engineering for Stable and Efficient Perovskite Solar Cells. Nano Energy 2017, 34 (Supplement C), 271–305. https://doi.org/10.1016/j.nanoen.2017.02.025. (32) Wirth, H. O. Organische Gläser Mit Hohen Glasumwandlungstemperaturen Auf Basis Niedermolekularer Verbindungen. Angew. Makromol. Chem. 1991, 185 (1), 329–334. https://doi.org/10.1002/apmc.1991.051850132. (33) Shirota, Y. Organic Materials for Electronic and Optoelectronic DevicesBasis of a Presentation given at Materials Chemistry Discussion No. 2, 13–15 September 1999, University of Nottingham, UK. J. Mater. Chem. 2000, 10 (1), 1–25. https://doi.org/10.1039/A908130E. (34) Osedach, T. P.; Andrew, T. L.; Bulović, V. Effect of Synthetic Accessibility on the Commercial Viability of Organic Photovoltaics. Energy Environ. Sci. 2013, 6 (3), 711. https://doi.org/10.1039/c3ee24138f. (35) Boyd, C. C.; Cheacharoen, R.; Bush, K. A.; Prasanna, R.; Leijtens, T.; McGehee, M. D. Barrier Design to Prevent Metal-Induced Degradation and Improve Thermal Stability in Perovskite Solar Cells. ACS Energy Lett. 2018, 3 (7), 1772–1778. https://doi.org/10.1021/acsenergylett.8b00926. (36) Xu, B.; Huang, J.; Ågren, H.; Kloo, L.; Hagfeldt, A.; Sun, L. AgTFSI as P-Type Dopant for Efficient and Stable Solid-State Dye-Sensitized and Perovskite Solar Cells. ChemSusChem 2014, 7 (12), 3252–3256. https://doi.org/10.1002/cssc.201402678. (37) Malinauskas, T.; Tomkute-Luksiene, D.; Sens, R.; Daskeviciene, M.; Send, R.; Wonneberger, H.; Jankauskas, V.; Bruder, I.; Getautis, V. Enhancing Thermal Stability and Lifetime of Solid-State DyeSensitized Solar Cells via Molecular Engineering of the Hole-Transporting Material Spiro-OMeTAD. ACS Appl. Mater. Interfaces 2015, 7 (21), 11107–11116. https://doi.org/10.1021/am5090385. (38) Huang, X.-L.; Zou, J.-H.; Liu, J.-Z.; Jin, G.; Li, J.-B.; Yao, S.-L.; Peng, J.-B.; Cao, Y.; Zhu, X.-H. A High T g SmallMolecule Arylamine Derivative as a Doped Hole-Injection/Transport Material for Stable Organic LightEmitting Diodes. Org. Electron. 2018, 58, 139–144. https://doi.org/10.1016/j.orgel.2018.04.012. (39) Leijtens, T.; Ding, I.-K.; Giovenzana, T.; Bloking, J. T.; McGehee, M. D.; Sellinger, A. Hole Transport Materials with Low Glass Transition Temperatures and High Solubility for Application in Solid-State DyeSensitized Solar Cells. ACS Nano 2012, 6 (2), 1455–1462. https://doi.org/10.1021/nn204296b. (40) Habisreutinger, S. N.; Noel, N. K.; Snaith, H. J.; Nicholas, R. J. Investigating the Role of 4-Tert Butylpyridine in Perovskite Solar Cells. Adv. Energy Mater. 2017, 7 (1), 1601079. https://doi.org/10.1002/aenm.201601079. (41) Magomedov, A.; Kasparavičius, E.; Rakstys, K.; Paek, S.; Gasilova, N.; Genevičius, K.; Juška, G.; Malinauskas, T.; Nazeeruddin, M. K.; Getautis, V. Pyridination of Hole Transporting Material in Perovskite Solar Cells Questions the Long-Term Stability. J. Mater. Chem. C 2018, 6 (33), 8874–8878. https://doi.org/10.1039/C8TC02242A. (42) Juarez-Perez, E. J.; Leyden, M. R.; Wang, S.; Ono, L. K.; Hawash, Z.; Qi, Y. Role of the Dopants on the Morphological and Transport Properties of Spiro-MeOTAD Hole Transport Layer. Chem. Mater. 2016, 28 (16), 5702–5709. https://doi.org/10.1021/acs.chemmater.6b01777.

ACS Paragon Plus Environment

19

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 20 of 20

(43) Wang, S.; Huang, Z.; Wang, X.; Li, Y.; Günther, M.; Valenzuela, S.; Parikh, P.; Cabreros, A.; Xiong, W.; Meng, Y. S. Unveiling the Role of TBP–LiTFSI Complexes in Perovskite Solar Cells. J. Am. Chem. Soc. 2018, 140 (48), 16720–16730. https://doi.org/10.1021/jacs.8b09809. (44) Ono, L. K.; Hawash, Z.; Juarez-Perez, E. J.; Qiu, L.; Jiang, Y.; Qi, Y. The Influence of Secondary Solvents on the Morphology of a Spiro-MeOTAD Hole Transport Layer for Lead Halide Perovskite Solar Cells. J. Phys. Appl. Phys. 2018, 51 (29), 294001. https://doi.org/10.1088/1361-6463/aacb6e. (45) Wang, S.; Sina, M.; Parikh, P.; Uekert, T.; Shahbazian, B.; Devaraj, A.; Meng, Y. S. Role of 4-TertButylpyridine as a Hole Transport Layer Morphological Controller in Perovskite Solar Cells. Nano Lett. 2016, 16 (9), 5594–5600. https://doi.org/10.1021/acs.nanolett.6b02158. (46) Saliba, M.; Matsui, T.; Seo, J.-Y.; Domanski, K.; Correa-Baena, J.-P.; Nazeeruddin, M. K.; Zakeeruddin, S. M.; Tress, W.; Abate, A.; Hagfeldt, A.; et al. Cesium-Containing Triple Cation Perovskite Solar Cells: Improved Stability, Reproducibility and High Efficiency. Energy Env. Sci 2016, 9 (6), 1989–1997. https://doi.org/10.1039/C5EE03874J. (47) Yang, M.; Kim, D. H.; Klein, T. R.; Li, Z.; Reese, M. O.; Tremolet de Villers, B. J.; Berry, J. J.; van Hest, M. F. A. M.; Zhu, K. Highly Efficient Perovskite Solar Modules by Scalable Fabrication and Interconnection Optimization. ACS Energy Lett. 2018, 3 (2), 322–328. https://doi.org/10.1021/acsenergylett.7b01221. (48) 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. https://doi.org/10.1021/acsenergylett.6b00270. (49) Fakharuddin, A.; Schmidt-Mende, L.; Garcia-Belmonte, G.; Jose, R.; Mora-Sero, I. Interfaces in Perovskite Solar Cells. Adv. Energy Mater. 2017, 7 (22), 1700623. https://doi.org/10.1002/aenm.201700623. (50) Marcus, R. A. On the Theory of Oxidation-Reduction Reactions Involving Electron Transfer. III. Applications to Data on the Rates of Organic Redox Reactions. J. Chem. Phys. 1957, 26 (4), 872–877. https://doi.org/10.1063/1.1743424. (51) Heun, S.; Borsenberger, P. M. A Comparative Study of Hole Transport in Vapor-Deposited Molecular Glasses of N,N′,N″,N‴-Tetrakis(4-Methylphenyl)-(1,1′-Biphenyl)-4,4′-Diamine and N,N′-Diphenyl-N,N′Bis(3-Methylphenyl)-(1,1′-Biphenyl)-4,4′-Diamine. Chem. Phys. 1995, 200 (1), 245–255. https://doi.org/10.1016/0301-0104(95)00192-Q. (52) Christians, J. A.; Manser, J. S.; Kamat, P. V. Best Practices in Perovskite Solar Cell Efficiency Measurements. Avoiding the Error of Making Bad Cells Look Good. J. Phys. Chem. Lett. 2015, 6 (5), 852– 857. https://doi.org/10.1021/acs.jpclett.5b00289.

ACS Paragon Plus Environment

20