Molecular Engineering of Triphenylamine-Based Non-Fullerene

Subscriber access provided by UNIV OF LOUISIANA

Energy, Environmental, and Catalysis Applications

Molecular Engineering of Triphenylamine-Based Non-fullerene Electron Transport Materials for Efficient Rigid and Flexible Perovskite Solar Cells Cheng Chen, Hong-Ping Li, Xingdong Ding, Ming Cheng, Henan Li, Li Xu, Fen Qiao, Huaming Li, and Licheng Sun ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b15130 • Publication Date (Web): 19 Oct 2018 Downloaded from http://pubs.acs.org on October 19, 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 27 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 Applied Materials & Interfaces

Molecular

Engineering

of

Triphenylamine-Based

Non-fullerene

Electron Transport Materials for Efficient Rigid and Flexible Perovskite Solar Cells Cheng Chen,

a#

Hongping Li,

a#

Xingdong Ding, a Ming Cheng, a* Henan Li, c Li Xu, a

Fen Qiao, d Huaming Li, a Licheng Sun b* a

Institute for Energy Research, Jiangsu University, Zhenjiang 212013, P. R. China

* E-mail: [email protected]. b

Department of Chemistry, KTH Royal Institute of Technology, SE-10044 Stockholm,

Sweden * E-mail: [email protected] c

School of Chemistry and Chemical Engineering, Jiangsu University, Zhenjiang 212013,

P. R. China d

School of Energy and Power Engineering, Jiangsu University, Zhenjiang 212013, P. R.

China #

These two authors contribute equally.

Keywords: perovskite solar cell, non-fullerene, flexible device, electron transport material, planar structure

Abstract: There has been a growing interest in the design and synthesis of non-fullerene electron transport materials (ETMs) for perovskite solar cells (PSCs) which may overcome the drawbacks of the traditional fullerene derivatives. In this work, a novel donor-acceptor (D-A) structured ETM termed TPA-3CN is presented by molecular engineering to include triphenylamine (TPA) as donor group and (3-Cyano-4,5,5-trimethyl-2(5H)-furanylidene)

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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 27

malononitrile as acceptor group. To further improve the electron mobility and conductivity and achieve excellent photovoltaic performance, a solution processable n-type dopant is introduced during ETM spin-coating step. After device optimization, PSCs based on the doped TPA-3CN exhibit impressive power conversion efficiency (PCE) of 19.2% with negligible hysteresis. Benefitting from the low-temperature and good solution processability of ETM TPA-3CN, it was further applied in flexible inverted PSCs and achieved an impressive PCE of 13.2%, which is among the highest values reported for inverted flexible fullerene-free PSCs. Introduction Due to the easy preparation technology and high efficiency, perovskite solar cells (PSCs) are considered to be one of the biggest scientific developments during the past few years. 1-10

PSCs are typically fabricated with a layer-by-layer structure, in which the perovskite

light harvesting layer is sandwiched between electron transport layer (ETL) and hole transport layer (HTL). The ETL plays an important role in selectively extracting generated electrons at the perovskite/ETL interface, in turn, efficiently restrict the charge recombination. So far, the fullerene derivatives, such as C60, PC61BM, PC71BM and ICBA, are still predominant electron transport materials (ETMs) in highly efficient inverted planar PSCs, due to their high electron mobility and isotropic electron-transfer properties.

11-18

Nevertheless, the high production costs, limited energy level variability, and particularly the poor photochemical stability restricted their further large-scale application in PSCs. 1922

Therefore, researchers devote themselves to

developing new non-fullerene ETMs to


Page 3 of 27 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


overcome their aforementioned defects while replicating the favorable properties.

23-31

Generally, several important parameters should be considered to obtain high-efficiency non-fullerene ETMs, such as suitable energy level, good electron transport property, easy processability and simple purification of production. In last few years, a series of donoracceptor (D-A) and acceptor-donor-acceptor (A-D-A) structured ETMs have been widely designed and reported in the field of organic solar cells, achieving high power conversion efficiency (PCE) nearly 15%. 32-34 However, none of those materials can be used as ETM in PSCs due to the problem of energy level mismatching. The LUMO of those D-A and AD-A structured small molecular materials are mainly determined by the terminal electron withdrawing groups. 35-40 Thus, those D-A or A-D-A type small molecular could be used as ETM in PSCs if appropriate core building block and electron withdrawing end groups were involved. Herein, we reported an efficient D-A structured non-fullerene ETMs, termed TPA-3CN, by molecular engineering of small molecular material, in which triphenylamine (TPA) was employed as donor group and (3-Cyano-4,5,5-trimethyl-2(5H)-furanylidene) malononitrile (3CN) was employed as acceptor group. To further improve the electron mobility and conductivity of TPA-3CN, a solution processable n-type dopant was introduced to the electron transport layer. The PSCs based on doped TPA-3CN exhibited a promising PCE of 19.2% with negligible hysteresis, which is even comparable with PC61BM based PSCs (19.1%). Benefitting from the low-temperature and good solution processability of ETM TPA-3CN, flexible inverted PSCs were also fabricated and achieved an impressive PCE of 13.2%, which is among the highest values reported for inverted flexible fullerene-free PSCs.



Page 4 of 27

Result and Discussion NC O

CN CN N N

NC NC

NC

N

CN

O

O

CN

O

O

n-type dopant H2

CN

TPA-3CN

Figure 1. Molecular structure of non-fullerene ETM TPA-3CN and n-type dopant H2 The chemical structures of ETM TPA-3CN and n-type dopant H2 are shown in Figure 1. The detailed synthetic route and characterization of ETM TPA-3CN are shown in supporting information (SI, Scheme S1). TPA-3CN can be easily synthesized through Knoevenagel condensation reaction with the total yield of 68%. The synthesis cost was calculated to be 133.85 $/g according to previously published material cost model, which is much lower than that of PC61BM.

29, 41

For D-A structured materials, the highest

occupied molecular orbital (HOMO) is mainly determined by donor group, and the lowest unoccupied molecular orbital (LUMO) is mainly controlled by electron withdrawing end group. In our case, to render deep-lying HOMO and LUMO levels, a core building block (TPA) with weak donating ability and terminal group (3CN) with strong electron withdrawing ability were selected to construct ETM TPA-3CN


Page 5 of 27 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


Figure 2. a) UV-Vis absorption spectra of TPA-3CN in chloroform solution and in solid film state, b) Cyclic voltammetry of TPA-3CN in chloroform solution, c) ESR spectra of pure TPA-3CN and doped TPA-3CN, d) Energy level diagram of the components in PSCs The optoelectronic property of TPA-3CN were characterized and the UV-Vis absorption spectra of TPA-3CN in solution and in solid film state are depicted in Figure 2a. The relevant data were summarized and listed in Table 1. As shown in Figure 2 and Table 1, in chloroform solution, TPA-3CN exhibited very strong absorption with the maximum absorption peak (λmax) of 576 nm. In comparison with the absorption in solution, the spectrum was obviously broadened in solid film state, suggesting π-π



Page 6 of 27

stacking of TPA-3CN occurred. According to the equation of E0-0 = 1240/λonset (eV), the optical bandgaps of TPA-3CN was calculated to be 1.85 eV. To roughly evaluate energy levels of TPA-3CN, cyclic voltammetry was performed in chloroform solution. As displayed in Figure 2b, t he LUMO level of TPA-3CN was extracted from the first reduction redox potentials and listed in Table 1. The ELUMO was calculated to be –4.07 eV for TPA-3CN. The slightly lower LUMO level of TPA-3CN than the valence band (VB) of perovskite (MAPbI3 here) suggests that the electron extraction and transport could be favorable from the perspective of energy level (see Figure 2d). The HOMO level of TPA-3CN is calculated to be –5.92 eV, which is negative enough for hole blocking function (see Figure 2d). Table 1. Optical and electrochemical data of TPA-3CN b

ETM

TPA-3CN a

a

λmax film /

λmax solution / nm

576

nm 573

c

d

d

E0-0 / eV

EHOMO / eV

ELUMO / eV

1.85

–5.92

–4.07

Absorption spectra were recorded in chloroform solution (2×10 -5 M). b Film of TPA-3CN were

prepared by spin-coating chloroform solution (20 mg/mL) onto FTO substrates at a spin rate of 1500 rpm. c E0-0 was estimated from the film absorption onsets. d CV measurements were carried out in chloroform solutions with [TBA]PF6 (0.1 M) as electrolyte, ferrocene/ferrocenium (Fc/Fc+) as an internal reference. EHOMO= ELUMO–E0-0.


Page 7 of 27 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


To gain insight into the stationary structure and molecular orbitals of ETM TPA3CN, density functional theory (DFT) calculations were performed at the B3LYP/631G(d)* level with the Gaussian 09 program. As shown in Figure 3, the HOMO and LUMO of TPA-3CN both delocalize on the whole molecular backbone, while the HOMO exhibits higher electron density on TPA moiety and the electron partly transfer to electron-withdrawing terminal groups for LUMO. The overlaps of the HOMO and LUMO orbitals in TPA-3CN indicates the formation of neutral excitons and hole-transfer transition, due to the strong coulomb interaction.

Figure 3. The stationary structure and molecular orbitals of TPA-3CN



Figure 4. a) Current-voltage (J-V) characteristics of pristine and 0.03 wt% doped TPA3CN based electron-only devices; b) Electron conductivity of pristine and 0.03 wt% doped TPA-3CN; c) steady-state photoluminescence (PL) of perovskite, perovskite/pristine TPA-3CN and perovskite/doped TPA-3CN systems; d) time-resolved PL decay of perovskite, perovskite/pristine TPA-3CN and perovskite/doped TPA-3CN systems The electron mobility and conductivity of ETM have great impact on the charge transfer rate. Thus, space-charge-limited currents (SCLC) were conducted to determine the electron mobilities, and a two-contact setup was used to determine the conductivity of the TPA-3CN, 31, 42-44 the obtained data are shown in Figure 4a and 4b. The extracted electron mobility and electron conductivity of pristine TPA-3CN are 1.43×10-4 cm2·V-1·s-1 and 1.11×10-4 s·cm-1, respectively. To further improve the electron mobility and electron conductivity of this new ETM TPA-3CN, n-type doping technology, which is widely used for PC61BM, was employed here and a reported strong n-type organic dopant H2 was selected. 45-48 The doping effect was proved by UV–vis absorption spectra and electron spin resonance (ESR). The pristine TPA-3CN film shows very strong absorption in the


Page 8 of 27

Page 9 of 27 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


region of 400–750 nm with a maximum absorption peak located at 573 nm (see Figure 2a). Doped by H2, TPA-3CN shows much weaker absorption intensity compared with pristine TPA-3CN and simultaneously exhibits a new absorption band around 490 nm, which is assigned to the formation of radical anion TPA3CN•-. 47, 49-50 Moreover, ESR spectra in Figure 2c further confirmed the doping of TPA-3CN by the H2. The optimized doping ratio is 0.03 wt% (see Figure S2) in our case. For 0.03 wt% doped TPA-3CN, the electron mobility and electron conductivity are dramatically improved and calculated to be 6.27×10-4 cm2·V-1·s-1 and 1.24×10-3 s·cm-1, respectively. To investigate the charge transfer properties, steady-state photoluminescence and time-resolved photoluminescence (TRPL) spectra were further tested. In our case, the perovskite material was fabricated on FTO glass. Maybe due to this reason, TRPL decay is existing in the process of excitation. Obviously quenched emission peak can be detected when coated by pristine and 0.03 wt% doped TPA-3CN (see Figure 4c), indicating the possibility of electron extraction at perovskite/ETL interface. The lower PL intensity for sample with 0.03 wt% doped TPA3CN as ETM than that with pristine TPA-3CN as ETM suggests more efficient electron extraction and transfer for doped TPA-3CN, matching well with the higher electron mobility for doped TPA-3CN obtained above. Moreover, the PL lifetime for pristine perovskite film is fitted to be 76 ns. When coated with pristine and 0.03 wt% doped TPA3CN, the PL lifetimes of the samples substantially decrease to 20.1 ns and 7.8 ns, respectively. The decreased PL lifetimes indicate faster electron transfer rate, which



agrees well with the electron mobility test results too.

Figure 5. a)-c) Top-view images of the perovskite, perovskite/pristine TPA-3CN and perovskite/doped TPA-3CN, d)-f) Height images of the perovskite, perovskite/pristine TPA-3CN and perovskite/doped TPA-3CN, g) Cross-section SEM image of the TPA-3CN based PSC Furthermore, the PSC employing TPA-3CN as ETMs was fabricated with an inverted planar structure [fluorine doped tin oxide (FTO)/NiO/MAPbI3/ETM/ bathocuproine (BCP)/Al], and its corresponding cross-section SEM image was shown in Figure 5g. The device exhibits well defined layer-by-layer structure. The BCP based interfacial layer were introduced between ETM and Al electrodes to improve the device fill factor (FF). The thickness of BCP modification layer is 3 nm, monitored by film thickness gauge during evaporation process, which was too


Page 10 of 27

Page 11 of 27 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


thin to be detected from SEM. The perovskite utilized in this work was MAPbI3. As shown in the SEM images from Figure 5a, a flat dense perovskite film with the thickness of 500nm can be detected. When the perovskite film was covered with the pristine TPA-3CN and doped TPA-3CN, respectively, both of their surfaces look obscure. The morphology of the pristine TPA-3CN film was inhomogeneous as small, aggregate dots were observed (Figure 5b). The morphology of the doped TPA-3CN film become more uniform and smoother (Figure 5c) due to the strong interaction between the n-type dopant and TPA3CN molecules. The film-formation property of pristine TPA-3CN and doped TPA3CN was further studied using atomic force microscopy (AFM). Due to the crystallization effects, the perovskite film shows high the root mean square (RMS) roughness of 42.6 nm with a peak-to-peak boundary height larger than 200 nm. Coated by pristine TPA-3CN and doped TPA-3CN, the RMS roughness of perovskite/pristine TPA-3CN and perovskite/doped TPA-3CN films were sharply reduced to 22.4 nm and 15.3 nm, respectively, suggesting that the perovskite film surface covered by the ETLs become much more uniform and smoother.



Figure 6. Current density-voltage properties of PSCs (scan rate: 20 mV/s) containing a) pristine TPA-3CN and b) doped TPA-3CN, c) IPCE spectra of PSCs employing pristine TPA-3CN and doped TPA-3CN as ETMs, d) stabilized current densities and PCEs at max power output points (0.83V for pristine TPA-3CN based PSCs, 0.88 V for doped TPA-3CN based PSCs) Figure 6a–b present the current density–voltage (J-V) scans for PSCs employing pristine TPA-3CN and doped TPA-3CN as ETMs, respectively, and the relevant parameters are collected in Table 2. The pristine TPA-3CN based PSCs show slightly high hysteresis with PCEs of 14.4% and 13.4% for scan from opencircuit (OC) to short-circuit (SC) and reverse scan, respectively. The hysteresis index (HI) was calculated to be 0.074. In contrast, the doped TPA-3CN based PSCs


Page 12 of 27

Page 13 of 27 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


exhibit higher PCE of 19.2% with dramatically improved short-circuit current density (Jsc), open-circuit voltage (Voc) and fill factor (FF), which is even comparable with PC61BM based devices (19.1%, see Figure S3). Moreover, the hysteresis (HI = 0.011) of the doped TPA-3CN based PSCs is negligible. Considering the only difference is with or without n-type dopant, the less hysteresis of doped TPA-3CN based PSC can mainly ascribe to the improved electron mobility and conductivity of ETL after doping. From the series resistance (Rs) values for pristine TPA-3CN and doped TPA-3CN based PSCs (see Table 2), we deduced that the increasement of FF can be attributed to the improved electron conductivity for doped TPA-3CN film. As shown in the incident photon-to-electron conversion efficiency (IPCE) spectra (see Figure 6c), no obvious difference in light absorption range was detected between the pristine TPA-3CN and doped TPA3CN based devices, therefore, the higher Jsc of PSC employing doped TPA-3CN as ETM is mainly attributed to its higher electron selection and transport efficiency, which is confirmed by the higher IPCE values obtained in the whole response range. The integrated current densities for pristine TPA-3CN and doped TPA-3CN based PSCs are 19.3 mA·cm−2 and 22.2 mA·cm−2, respectively, matching well with the J-V measurements. Meanwhile, the stabilized PCEs and photocurrent densities at maximum point (0.83 V and 0.88 V for pristine and doped TPA3CN based PSCs, respectively) were tested to confirm the veracity of J-V test. As shown in Figure 6d, stable photocurrent density of 18.2 mA·cm−2 and 21.6 mA∙cm-2 and stable



Page 14 of 27

PCEs of 15.1% and 19.0% were obtained for pristine TPA-3CN and doped TPA-3CN based PSCs, respectively. The values agree well with J-V test results, indicating our devices are not overestimated. In addition, pristine TPA-3CN and doped TPA-3CN based PSCs show good reproducibility of the PCEs (see Figure S4). Table 2. Photovoltaic performance of PSCs employing pristine TPA-3CN and doped TPA-3CN as ETMs Scan ETM

directio n

Voc / V

Jsc / mA·cm2

hyste FF

PCE

resis

/%

/%

index

Rsr /

Rsh /

Ω·cm2 Ω·cm2

/%

From Pristine TPA3CN

OC to

1.00

19.7

73.3

14.4

SC From SC to

1.00

19.8

67.7

13.4

1.05

22.5

81.1

19.2

0.074

10.2

4328

0.011

5.6

7381

0.024

15.6

3249

OC From Doped

OC to

TPA-

SC

3CN

From

(rigid)

SC to

1.05

22.5

80.3

19.0

0.98

19.7

68.6

13.2

OC Doped

From

TPA-

OC to

3CN

SC


Page 15 of 27 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


(flexible)

From SC to

0.98

19.8

67.9

13.1

OC

Figure 7. Light intensity dependence of a) Jsc and b) Voc for PSCs with pristine and doped TPA-3CN as ETM For doped TPA-3CN based PSC, the improvement of Voc can be ascribe to the lower recombination kinetics at the perovskite/ETL interface. To further verify our hypothesis, we investigated the light intensity dependence of the Jsc and Voc. Figure 7a presents the power law dependence of the Jsc on light intensities (J = Iα). The α for pristine TPA-3CN and doped TPA-3CN based PSCs was extracted to be 0.856 and 0.934, respectively. The higher α for the doped TPA3CN based PSCs suggests that the enhanced electron mobility by doping strategy significantly restricts the recombination loss at the perovskite/ETL interface. Moreover, as shown in Figure 7b, the light intensity dependence of Voc indicats interfacial trap-assisted Shockley−Read−Hall (SRH) recombination is involved



in both pristine and doped TPA-3CN based devices. The weaker Voc dependence on the light intensity (1.37kT/q) for doped TPA-3CN based device manifest the SRH recombination is better restricted, which is critical to achieve good photovoltaic performance. The long-term stability is also an important key performance indicator for the future applications. Therefore, the aging test of pristine TPA-3CN and doped TPA-3CN based PSCs were performed under controlled aging conditions (30-35 oC

and ⁓40% humidity in air atmosphere without encapsulation). The PSCs

containing pristine TPA-3CN as ETM maintained 93.6% of the champion photovoltaic performance by showing a PCE of 14.1%, after 480 h (see Figure S5). In comparison, the PSCs employing doped TPA-3CN as ETM showed slightly serious decrease of PCEs, from the initial 18.4% to 15.4%. The unideal stability of doped TPA-3CN based devices can be attributed to the decreased hydrophobicity of doped TPA-3CN film (see Figure S6). The pristine TPA-3CN film can more efficiently prevent the perovskite layer from being degraded by water vapor, leading to better device stability. Under the same conditions, the PCE of PC61BM based PSCs dropped from 18.8% to 5.7% due to poor photochemical stability of ETL in air. Therefore, TPA-3CN is a promising alternative in terms of device stability.


Page 16 of 27

Page 17 of 27 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


Figure 8. Current density-voltage curves of doped TPA-3CN based flexible PSCs with device photo inserted Benefiting from the low-temperature solution processability of TPA-3CN based ETL (

Molecular Engineering of Triphenylamine-Based Non-Fullerene

Recommend Documents