Efficient Solution-Processed n-Type Small-Molecule Thermoelectric

Subscriber access provided by University of Rochester | River Campus & Miner Libraries

Article

Efficient Solution-Processed n-Type Small-Molecule Thermoelectric Materials Achieved by Precisely Regulating Energy Level of Organic Dopants Dafei Yuan, Dazhen Huang, Cheng Zhang, Ye Zou, Chong-an Di, Xiaozhang Zhu, and Daoben Zhu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b07282 • Publication Date (Web): 11 Aug 2017 Downloaded from http://pubs.acs.org on August 12, 2017

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

ACS Applied Materials & Interfaces 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 Applied Materials & Interfaces

Efficient Solution-Processed n-Type SmallMolecule Thermoelectric Materials Achieved by Precisely Regulating Energy Level of Organic Dopants. Dafei Yuan, †,‡,ǁ Dazhen Huang, †,‡,ǁ Cheng Zhang, †,ǁ Ye Zou, † Chong-an Di, *,†,ǁ Xiaozhang Zhu *,†,ǁ

†

and Daoben Zhu†,ǁ

Beijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Organic Solids,

Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, P. R. China. ǁ

University of Chinese Academy of Sciences, Beijing 100190, P. R. China.

KEYWORDS: organic thermoelectrics, n-type doping, small molecule, solution-processed, energy level.

ABSTRACT: To achieve efficient n-type doping, three dopants, 2-Cyc-DMBI-H, (2-CycDMBI)2, and (2-Cyc-DMBI-Me)2 with precisely regulated electron-donating ability were designed and synthesized. By doping with a small molecule 2DQTT-o-OD with high electron mobility, an unexpectedly high power factor of 33.3 µW m−1 K−2 was obtained with the new dopant (2-Cyc-DMBI-Me)2. Notably, with the intrinsically low lateral thermal conductivity of

ACS Paragon Plus Environment

1


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

0.28 W m−1 K−1, the figure of merit was determined to be 0.02 at room temperature. Thus, we have demonstrated that small molecules with high electron mobility and low-lying LUMO energy levels can achieve high doping efficiency and excellent thermoelectric properties by doping with n-type dopants featuring highly matched energy levels and excellent miscibility.

1. INTRODUCTION Because of their flexibility, abundance of sources, and potential in large-area production, organic semiconducting materials are receiving increasing attention in thermoelectric generators that transform waste heat into electricity and can be used as a new type of electronic devices, such as power sensors. 1-3 Thermoelectric (TE) efficiency is dominated by materials’ figure of merit, ZT = S2σT/κ, where σ is the electrical conductivity, S is the Seebeck coefficient, κ is the thermal conductivity, T is the absolute temperature, and the power factor (PF) = S2σ can also be used to describe TE efficiency. Because the organic materials intrinsically possess low thermal conductivities in the range of 0.1−1 W m−1 K−1, the thermoelectric efficiency can be improved through the optimization of the Seebeck coefficient and electrical conductivity .4,5 A working TE device should consist of both the p- and n-type modules. To date, p-type materials such as PEDOT have achieved high thermoelectric performance with high ZT values of 0.42 and 0.25.6,7 However, the development of n-type thermoelectric materials are lagging far behind, which can be attributed to the lack of high-performance n-type organic semiconductors. Among the highestn-type thermoelectric materials, the organometallic poly(Ni 1,1,2,2-ethenetrathiolate) derivatives, prepared by the electrochemical deposition, have obtained power factors over 400 µW m−1 K−2.8,9 Meanwhile, n-type thermoelectric materials based on solution-processed conjugated polymers have been developed quickly. Chabinyc et al. used the organic dopant NDMBI to dope the semiconducting polymer P(NDIOD-T2), which led to a power factor of 0.6


2

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


µW m−1 K−2.10 Later on, Pei et al. used the same dopant N-DMBI to dope BDPPV derivatives and achieved electrical conductivity as high as 14 S cm−1 and a power factor of 28 µW m−1 K−2.11 In comparison to conjugated polymers, small molecules have well-defined structures, which is beneficial for better repeatability and the establishment of structure-property relationship. There are several kinds of n-type small molecules which show both high Seebeck coefficient and good electrical conductivity after doping, such as vapor doped fullerenes12-14 and metal bismuth interfacial doped TDPPQ.15 However, small molecules for solution-processed n-type thermoelectric materials are quite rare. Very recently, Segalman et al. synthesized self-doped PDI small molecules for n-type thermoelectric applications and obtained power factors as high as 1.4 µW m−1 K−2.16 Zhu et al. have developed solution-processed Cu-TCNQ nanorod arrays giving a power factor of 2.5 µW m−1 K−2.17 To achieve high performance solution-processed n-type thermoelectric materials, doping is crucial. Nowadays, for n-type thermoelectric materials, chemical doping and electrochemical doping are commonly used to achieve high device performance. Taking the advantages of easy synthesis, convenient device processing, and potential of large-scale preparation, chemical doping is a promising way for industrial application. However, there are just a few kinds of ntype organic dopants such as organometallics and cationic dyes which can be utilized for n-type thermoelectric materials. 18-20 Recently, Bao et al. developed a series of n-type dopants based on 1,3-dimethyl-2,3-dihydro-1H-benzimidazoles (DMBI-H)21,22 and the dimer dopant (2-CycDMBI)2 giving an electrical conductivity of 12 S cm−1 when co-deposited with C60 in vacuum.23 Designing appropriate n-type dopants is challenging for the crucial requirements of high-lying highest occupied molecular orbital (HOMO) energy level and relative ambient stability in order to achieve high doping efficiency. Meanwhile, according to the definition of electrical


3


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 22

conductivity, σ = nqµ, where n is carrier concentration, q is carrier charge, and µ is carrier mobility. σ can be improved by doping to significantly increase the carrier concentration (n). Besides, carrier mobility (µ) is also crucial for semiconductors to obtain high electrical conductivity. Thus, many high electron mobility organic materials should be principally ideal candidate for n-type organic thermoelectric materials. We report herein the thermoelectric application of n-type small molecule 2DQTT-o-OD that was recently developed in our group by effective n-type doping. 2DQTT-o-OD is an excellent ntype semiconductor with high electron mobility over 5 cm2 V−1 s−1 attributing to the rigid and planar molecular structure and the favorable “brick-layer” arrangement in the solid state.24,25 Moreover, the deep lowest unoccupied molecular orbital (LUMO) energy level of −4.4 eV makes it much easier to be n-doped. Based on two dopants 2-Cyc-DMBI-H and (2-Cyc-DMBI)2 developed by Bao et al.19, we synthesized a new dopant (2-Cyc-DMBI-Me)2 with methyl groups on two benzimidazoles to further increase the HOMO energy level for higher doping efficiency. These three dopants are all utilized to dope 2DQTT-o-OD and investigated the thermoelectric properties systematically. In contrast with low power factor of 2-Cyc-DMBI-H and (2-CycDMBI)2, an unexpected high power factor of 33.3 µW m−1 K−2 was achieved by the new dopant (2-Cyc-DMBI-Me)2, which, to the best of our knowledge, is the highest among solutionprocessed n-type small-molecule thermoelectric materials. Meanwhile, the ZT value of (2-CycDMBI-Me)2-doped film was determined to be 0.02 at room temperature with the intrinsically low lateral thermal conductivity of 0.28 W m−1 K−1.

2. EXPERIMENTAL SECTION


4

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


2.1 Materials. All the reactions dealing with air- or moisture-sensitive compounds were carried out in a positive atmosphere of nitrogen. Unless stated otherwise, starting materials were obtained from Adamas, Aldrich and J&K and were used without any further purification. Anhydrous THF and toluene were distilled over Na/benzophenone prior to use. The small molecule semiconductor 2DQTT-o-OD25 and new dopant 2,2’-dicyclohexyl-1,1’,3,3’,5,5’hexamethyl-2,2’,3,3’-tetrahydro-1H,1’H-2,2’-bibenzo[d]imidazole

(2-Cyc-DMBI-Me)2

were

prepared according to the published procedures.23 2.2 Measurements and General Methods. Hydrogen nuclear magnetic resonance (1H NMR) and carbon nuclear magnetic resonance (13C NMR) spectra were measured on BRUKER DMX 300 and BRUKER DMX 400 spectrometers. Chemical shifts for hydrogens are reported in parts per million (ppm, δ scale) downfield from tetramethylsilane and are referenced to the residual protons in the NMR solvent and (13C NMR spectra were recorded at 100 MHz. Chemical shifts for carbons are reported in parts per million (ppm, δ scale) downfield from tetramethylsilane and are referenced to the carbon resonance of the solvent. All the NMR measurements were conducted at ambient condition. HR-ESI and HR-MALDI-TOF measurements were performed on an Applied Biosystems 4700 Proteomics Analyzer. The UPS and XPS measurements were carried out in a Kratos ULTRA AXIS DLD ultrahigh vacuum photoelectron spectroscopy connected to a custom-made high vacuum thermal evaporation system without exposure to air. The base pressure of the analysis chamber and the evaporation chamber were better than 5 × 10−10and 5 × 10−9 Torr, respectively. The pristine films and doped films were obtained by spin coated on 1 cm × 1 cm bare silicon wafer in glove box. After thermal annealing at 120 oC in N2 atmosphere for 2 hours, the films were transferred to the analysis chamber without breaking the vacuum for UPS and XPS measurements. An unfiltered He-discharge lamp (21.22 eV) and a monochromatic Al Kα X-ray (1486.6 eV) excitation sources were respectively equipped for UPS and XPS analysis. The energy resolution for UPS was 100 meV as estimated from the Fermi edge of an Ar+ sputtered clean Au film. The samples were negatively biased at 9.0 V with respect to the electron analyzer for obtaining the secondary electron cutoff (SECO) spectra. The Fermi edge was calibrated from a UPS spectrum of the cleaned Au substrate. UV-vis spectra of thin films in a sealed cuvette were recorded on a JASCO


5


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 22

V-570 spectrometer. Atomic force microscopy (AFM) images of the thin films were obtained on a NanoscopeIIIa AFM (Digital Instruments) operating in tapping mode. X-Ray difratio (XRD) measurements of thin films were performed in reflection mode at 40 kV and 200 mA with Cu Kα radiation using a 2 kW Rigaku X-ray diffractometer. Synchrotron-based Grazing-incidence wide-angle X-ray scattering (GIWAXS) were measured at the small-wide-angle X-ray scattering beamline 8ID-E at the Institute of High Energy Physics Chinese Academy of Sciences. The sample was placed in a helium chamber, using Pilatus 1 M detector. The sample-detector was 208 mm; the X-ray energy and wavelength are 7.35 KeV and 1.6868 Angstrom, respectively. The incidence angle was 0.2 degrees, with single acquisition time of 1 second.

2.3 Doping Procedure. ALL the doping procedure were conducted under N2 atmosphere. In the concentration of 5 mg/mL 2DQTT-o-OD in CHCl3 solution, different molar ration of dopnats in CHCl3 was added. The solutions were then spin-coated on glass and OTS-modified SiO2/Si substrate with a rotation rate of 2000 rpm and then thermal annealing at 120 oC for 2 hours under N2 atmosphere. The thickness of the thin films were obtained by Atomic force microscopy (AFM) method and testimated to be 40nm. 2.4 OFET Fabrication and Measurement. The Bottom gate Bottom contact device (BGBC) were fabricated to investigate OFET property. SiO2 with a thickness of 300 nm was used as the dielectric and n-type heavily doped Si was used as the gate. The patterned Au (30 nm) for drainsource electrode was deposited on the surface of SiO2/Si substrate by the shadow mask under vaccum. The channel length were 5-50 µm and the width was 1400 µm. The SiO2/Si substrate was

cleaned

by

distilled

water,

ethanol

and

acetone

and

then

modified

with

octadecyltrichlorosilane (OTS) at 120 oC under vaccum atmosphere. The pristine and doped solutions were spin-coated on the substrate and then thermal annealing at 120 oC for 2 hours


6

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


under N2 atmosphere. The measurement of OFET were conducted under N2 atmosphere by the use of Keithley 4200 SCS. 2.5 Measurements of Thermoelectric Properties. The glass substrates were cleaned by distilled water, ethanol and acetone .The Au electrodes (30 nm) were then deposited by a shadow mask with a channel length of 500 µm and channel width of 5000 µm for Seebeck coefficient and conductivity measurement. The pristine and doped solutions were spin-coated on the glass substrates and then thermal annealing at 120 oC for 2 hours under N2 atmosphere. The electrical conductivity was tested by the four-probe method with Keithley 4200 SCS. The Seebeck coefficient was measured in a vacuum chamber and calculated by the formula S = Vtherm/∆T, where Vtherm is the thermal voltage obtained by creating a temperature gradient (∆T) at the two ends of the device by two Peliter elements and controlled by utilizing an infrared camera FLIR A300 (thermal sensitivity < 50 mK). The measurement of Vtherm was conducted by Keithley 4200 SCS and the accuracy of the temperature measurements was verified by two resistive thermometers near the electrodes. The Seebeck coefficient and electrical conductivity were measured with the same device. 15 2.6 Measurements of thermal conductivity. The differential 3ω method6,7 was used in this work to measure the vertical thermal conductivity of the (2-Cyc-DMBI-Me)2 doped films. The FTS modified Si/SiO2 substrates with shadow mask were treated with the UVO (UV-ozne) for 10 min to create hydrophobic substrates with patterned hydrophilic regions. The doped solutions in o-DCB with a concentration of 10mg/mL were dip-casted on the hydrophilic regions. After thermal annealing at 120oC for 2 hours under N2 atmosphere, a thickness of 100 nm SiO was vacuum-deposited on the surface of the doped thin films to keep off oxygen and water. And then a thickness of 1µm insulating layer of SiNx was deposited by PECVD on the surface of the


7


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 22

samples. After that, gold electrode was patterned on the top of the samples by standard photolithography technique.

3. RESULTS 3.1 Molecular Structure and Energy level. The molecular structures of 2DQTT-o-OD and the three dopants, 2-Cyc-DMBI-H, (2-Cyc-DMBI)2, and (2-Cyc-DMBI-Me)2 are shown in Figure 1. According to the theoretical calculations at the B3LYP/TZP level with the Amsterdam Density Functional package (Figure 2), the HOMO energy level of (2-Cyc-DMBI-Me)2 is −4.59 eV, which is higher than those of the other two dopants, −4.98 eV for 2-Cyc-DMBI-H and −4.66 eV for (2-Cyc-DMBI)2. Meanwhile, the LUMO energy level of 2DQTT-o-OD is −4.68 eV that is very close to the HOMOs of dimer dopants, (2-Cyc-DMBI)2 and (2-Cyc-DMBI-Me)2. Considering the highest HOMO energy level of (2-Cyc-DMBI-Me)2 among the three dopants, the electron transfer from the HOMO of the dopant to the LUMO of 2DQTT-o-OD could be facilitated, which suggests the high doping efficiency. 2DQTT-o-OD radical anion (Figure S4) shows a fully delocalized singly occupied molecular orbital over the whole planar backbone after doping, which is highly preferred for electrical conductivity.26

Figure 1. Molecular structures of 2DQTT-o-OD and three dopants.


8

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


Figure 2. The Lowest unoccupied molecular orbital (LUMO) energy level of 2DQTT-o-OD and highest occupied molecular orbital (HOMO) energy level of three dopants. (B3LYP/TZP). 3.2 Spectroscopy Analysis of Doping Efficiency. To determine the doping efficiency of the three dopants, UV-vis-NIR absorption spectroscopy, ultraviolet photoemission spectroscopy (UPS), and X-ray photoelectron spectroscopy (XPS) were performed on doped thin films. Chloroform solutions containing 2DQTT-o-OD and dopants were spin-coated on glass for UVvis-NIR measurements and on bare Si wafers for UPS and XPS investigations. The rotation rate for spin-coating was 2000 rpm and thermal annealing was performed for 2 h at 120 oC under N2 atmosphere. Compared with the pristine film, the measured UV-vis-NIR spectroscopy (Figure S5) showed that all the doped films at the same molar ratio had a weakened absorption at around 450 nm and 750 nm and the new peaks at around 540 nm and 1250 nm. These changes of UVvis-NIR absorptions agreed well with the calculated spectra (Figure S6) obtained at B3LYP/TZP level, from which we can conclude that the doping actually work effectively between 2DQTT-oOD and three dopants.11,26 In addition, electron spin resonance (ESR) experiment was conducted for (2-Cyc-DMBI-Me)2 doped film as an example to show the doping and the generation of free electron (Figure S7). UPS spectra (Figure 3a, b) are used to accurately describe the doping level.


9


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 22

The work functions were shifted upward from 4.8 eV for the pristine film to 4.7 [2-Cyc-DMBIH], 4.3 [(2-Cyc-DMBI)2], and 4.2 eV [(2-Cyc-DMBI-Me)2], respectively. Moreover, the characteristic HOMO feature onset of pristine 2DQTT-o-OD (around 1.50 eV) was also shifted toward a higher binding energy level, 1.58 [2-Cyc-DMBI-H], 1.64 [(2-Cyc-DMBI)2], and 1.67 eV [(2-Cyc-DMBI-Me)2]. As a result, the Fermi levels of the doped films were shifted toward the LUMO energy level of 2DQTT-o-OD. Among the three doped films, the Fermi level of the film doped by (2-Cyc-DMBI-Me)2 was the highest, which means that it had the highest doping level among the three dopants. As shown in the XPS spectra (Figure 3c, d), all the carbon peaks C (1s) of the doped films were shifted toward the higher binding-energy side. As for the nitrogen peaks N (1s), the N signal of TPD (400.5 eV) and cyano (399 eV)15 became obviously weaker for the two dimers, whereas nearly no change could be observed for 2-Cyc-DMBI-H. The signal of N (1s) about 401.5 eV of the new generated DMBI+,11 can be easily recognized in the three doped films. Moreover, the area of the DMBI+ peak belonging to (2-Cyc-DMBI-Me)2 was slightly larger than that of (2-Cyc-DMBI)2, which agreed well with the higher doping level of the former. The area of the new peak for the 2-Cyc-DMBI-H-doped film was relatively small, and thus we consider its doping level would be comparatively low. Therefore, through regulating the energy level by electronic effects, more effective doping can be achieved with the (2-CycDMBI-Me)2 dopant for its highest HOMO energy level.


10

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


Figure 3. a) The low kinetic energy region and b) low binding energy region (HOMO) of UPS spectra; c) experimental C 1s and d) N 1s XPS spectra for the pristine film (black curve) and doped films with 2-Cyc-DMBI-H (green curve), (2-Cyc-DMBI)2 (red curve), and (2-Cyc-DMBIMe)2 (blue curve) at the same molar ratio of 10mol%. 3.3 OFET Properties of Doped Films. To further investigate the doping effect, organic thin film transistors (OTFTs) were used to explore the transistor properties of the pristine 2DQTT-o-OD film and films doped by the three dopants, 2-Cyc-DMBI-H, (2-Cyc-DMBI)2, and (2-Cyc-DMBIMe)2 at the same molar ratio. Bottom gate/bottom contact (BGBC) devices were thus fabricated to characterize the field effect transistor properties of pristine and doped films. After mixing the dopants and 2DQTT-o-OD in solutions, the mixtures were spin-coated on the OTS-modified


11


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 22

SiO2/Si substrate and thermally annealed at 120 oC under N2 atmosphere for 2 h. The pristine film showed typical n-type transfer and output features with a large average on-off ratio of 106 (Figure S8 and S9). After doping with the three dopants, no off state could be observed and the drain-source current IDS increased from 10−5 A in the pristine film to 10−3 A in doped films, which indicates the generation of free electrons after doping. Among the three doped films at the same molar ratio, the higher IDS of (2-Cyc-DMBI-Me)2 also implies a higher doping level than the other two dopants. 3.4 Thermoelectric Properties. To investigate the thermoelectric properties, solutions of 2DQTT-o-OD mixed with the three dopants in a series of molar ratios were spin-coated on glass substrates with pre-patterned Au electrode. After thermal annealing at 120 °C for 2 h, the fourpoint probe method was used to measure the electrical conductivity. The Seebeck coefficient was obtained by measuring the thermovoltage when applying a temperature difference between the two sides of the film. The electrical conductivity of doped films increases dramatically by increasing the dopant molar ratio up to the maxima at 15mol% for 2-Cyc-DMBI-H, 10mol% for (2-Cyc-DMBI)2 and (2-Cyc-DMBI-Me)2 (Figure 4). The highest electrical conductivities of three doped films were 0.18 [2-Cyc-DMBI-H], 0.43[(2-Cyc-DMBI)2], and 1.1 S cm−1 [(2-Cyc-DMBIMe)2], respectively. By contrast, the Seebeck coefficients decreased dramatically with increasing dopant ratios, which is in inverse correlation with the behavior of electrical conductivity. The Seebeck coefficients of the 2-Cyc-DMBI-H-doped films were higher than for the films doped by the other two dopants under the same dopant molar ratios consistent with its comparatively low doping level. When the molar ratios of dopants were increased to over 20mol% for 2-CycDMBI-H and 10mol% for (2-Cyc-DMBI)2 and (2-Cyc-DMBI-Me)2, the electrical conductivity decreased dramatically, which suggests the excess dopants actually acted as impurities that


12

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


disrupted the molecular packing in the doped films and decreased electron mobility. The power factor (PF) reaches its maximum at a specific molar ratio following the equation: PF = σS2. The maximum of PF was 2.7 µW m−1 K−2 for 15mol% 2-Cyc-DMBI-H and 7.2 µW m−1 K−2 for 10 mol% (2-Cyc-DMBI)2. Among the three dopants, the film doped by (2-Cyc-DMBI-Me)2 showed the highest PF of 17.2 µW m−1 K−2 at the molar ratio of 10mol%. At elevated temperatures, both the electrical conductivity and the Seebeck coefficient increased, which is in agreement with the thermally assisted Mott’s variable range hopping transport mode27. The highest power factor of 33.3 µW m−1 K−2 for the (2-Cyc-DMBI-Me)2-doped film was achieved at 90 °C.

Figure 4. The thermoelectric properties of 2DQTT-o-OD doped by different dopants: 2-CycDMBI-H (black point), (2-Cyc-DMBI)2 (red point), and

(2-Cyc-DMBI-Me)2 (blue point) in

different molar ratios: a) Electrical conductivity, b) Seebeck coefficient, c) Power factor.


13


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 22

Temperature dependence of d) Electrical conductivity, e) Seebeck coefficient, f) Power factor for doped films with the optimized 10mol% (2-Cyc-DMBI-Me)2. 3.5 Thermal Conductivity and Figure of Merit. To calculate the figure of merit (ZT), the differential 3ω method6,7 were used to measure the thermal conductivity of the n-doped thin film, which is recognized as a great challenge and was not reported in the very recent papers (Table 1). Here, o-dichlorobenzene solutions containing 2DQTT-o-OD and (2-Cyc-DMBI-Me)2 at the molar ratio of 10mol% were dip-casted on the patterned hydrophilic region of SiO2/Si substrate (Figure S1). As shown in Figure S2, by comparing the thermal response to the heater (50 µm) of the doped film with the reference, the vertical thermal conductivity was determined to be 0.18 W m−1 K−1 at 298 K. The anisotropic property was observed in the thermal conductivity of (2-CycDMBI-Me)2-doped film with a ratio of 1.55 (lateral versus vertical), when comparing the thermal response to line heaters with a width of 50 and 3.9 µm (Figure S3). The lateral thermal conductivity was determined to be 0.28 W m−1 K−1 at 298 K, which is in consistent with the intrinsically low thermal conductivity of organic thermoelectric materials. Together with the power factor measured at room temperature, the ZT value was determined to be 0.02. Furthermore, as illustrated in Figure 4f, the temperature dependence of thermoelectric properties, the ZT value may be further enhanced at elevated temperature in future study. Table 1. Thermoelectric Properties of n-Type Solution-Processed Polymer and Small-Molecule Thermoelectric Materials Reported Recently and in This Work. Table 1. Materials

Power factor Vertical (µWm−1 K−2) thermal conductivity (W m−1 K−1)

Lateral thermal conductivity (W m−1 K−1)

ZT

Refs

P(NDIODT2)a

0.6

—

—

10

—


14

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


BBLa

0.43

—

—

—

26

BDPPVa

28

—

—

—

11

PDIb

1.4

—

—

—

16

Cu-TCNQb

2.5

—

—

—

17

This workc

33

—

—

—

This workd

17.2

0.18

0.28

0.02

a

Polymer; bSmall molecule; Small molecule 2DQTT-o-OD doped by (2-Cyc-DMBI-Me)2 at the

optimized molar ratio at 363Kc and 298Kd. 3.6 Thin Film Morphology. AFM, X-ray diffraction (XRD) and Grazing-incidence wide-angle X-ray scattering (GIWAXS) were used to investigate the morphology and molecular packing of the pristine and doped films. From the AFM height image (Figure S10) of pristine and doped films at different doping molar ratios, the root-mean-square roughness values of the doped films, 1.39 nm [10mol% 2-Cyc-DMBI-H], 1.75 nm [10mol% (2-Cyc-DMBI)2], and 2.15 nm [10mol% 2-Cyc-DMBI-Me)2], were smaller than that for the pristine film (2.38 nm). Smaller roughness for the doping films with the dopants meant that the dopants could be well dispersed in 2DQTT-oOD with excellent miscibility. Meanwhile, the higher crystallinity of the (2-Cyc-DMBI-Me)2doped film with larger domain would facilitate better electron transport. When the dopant molar ratio was increased to 20mol%, larger roughness values and lower crystallinity were found for the three dopants, leading to the larger disruption of the 2DQTT-o-OD packing. As a result, the introduction of dopants will induce electron transfer to 2DQTT-o-OD, which would make the dopants fill into the space between adjacent 2DQTT-o-OD molecules. Thus, the high quality 2D stacking for the pristine film will be somehow disrupted. The XRD images along out-of-plane (Figure S11), showed that a higher dopant molar ratio (20mol%) would lead to lower


15


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 22

crystallinity than with an optimized molar ratio of 10mol%, which would hinder the transport of free electrons and lead to decreased electron mobility. Taking in-plane GIWAXS patterns (Figure S12) of the (2-Cyc-DMBI-Me)2 doped films as an example, the pristine 2DQTT-o-OD film showed very high crystallinity. With the increasing molar ratio of dopants to 10mol%, the intensity of the main diffraction peaks decreased slowly and two new peaks generated at around 5 degree and 26 degree, which demonstrated that the relatively high crystallinity can be retained with the optimized dopant molar ratio. However, with an excess of dopant molar ratio of 20mol%, the diffraction peaks dropped sharply, which can be attributed to the interference of π-π stacking. 4. DISCUSSION Based on the systematic investigations above, the high thermoelectric performance of 2DQTT-oOD can be attributed to its high electrical conductivity and intrinsically excellent Seebeck coefficient with a well-matched dopant. The possible mechanisms to molecular n-doping are shown in Figure 5.23 In mechanism (a), the n-type doping works between 2-Cyc-DMBI-H and 2DQTT-o-OD can be ascribed to the hydride or hydrogen atom transfer from the dopant to 2DQTT-o-OD. Meanwhile, in mechanism (b), dimer dopants (2-Cyc-DMBI)2 and (2-CycDMBI-Me)2 transfer an electron from their HOMO to the LUMO of 2DQTT-o-OD and then form cation radical M2•+ and anion radical A•–, followed by rapid irreversible cleavage of the dimer cation radical to give radical M• and cation M+. Then, the unpaired electron of M• continuously proceeds charge transfer to another 2DQTT-o- OD molecule, and finally two M+ and two A•–are achieved. Thus, under the same circumstances, the doping efficiencies of (2-CycDMBI)2 and (2-Cyc-DMBI-Me)2 are twice as much as that of 2-Cyc-DMBI-H, which leads to the lowest power factor of 2-Cyc-DMBI-H-doped film among the three dopants. Besides,


16

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


compared with (2-Cyc-DMBI)2, the higher lying HOMO energy level of (2-Cyc-DMBI-Me)2 is more facilitated to the charge transfer and result in higher doping level in the first doping process, which could further ensure the second doping process proceed smoothly to achieve higher performance, as proved by the UPS and XPS. With the relatively high crystallinity of (2Cyc-DMBI-Me)2 doped film (10mol%), the generated carrier is able to transport through the doped films, and a high electrical conductivity over 1.0 S cm−1 is obtained. Thus, a power factor as high as 33.3 µW m−1 K−2 is achieved together with the high Seebeck coefficient of 2DQTT-oOD. As a result, to achieve efficient solution-processed n-type small-molecule thermoelectric materials, both high electron-transport organic semiconductors and energy matched n-type dopants should be taken into consideration.

Figure 5. The possible doping mechanisms of a) monomer dopant and b) dimer dopant with small molecule 2DQTT-o-OD. 5. CONCLUSION We used three n-type dopants, 2-Cyc-DMBI-H, (2-Cyc-DMBI)2, and (2-Cyc-DMBI-Me)2 with varied HOMO energy levels, to dope a high-performance n-type small-molecule semiconductor, 2DQTT-o-OD, for solution-processed organic thermoelectric materials. By proper selection of dopants, excellent doping efficiency and thin-film morphology can be achieved, which results in a high electrical conductivity and Seebeck efficient. We obtained a high power factor of 33.3 µW m−1 K−2, which is the highest among solution-processed n-type small-molecule thermoelectric


17


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 22

materials, by applying the (2-Cyc-DMBI-Me)2 dopant featuring the most matched electronic energy level. The ZT value of (2-Cyc-DMBI-Me)2-doped film was determined to be 0.02 at 298 K with the low lateral thermal conductivity of 0.28 W m−1 K−1. Besides the energy-level regulation, we think that thin-film morphology being affected by dopants should be also carefully considered for high thermoelectric performance.

ASSOCIATED CONTENT Supporting Information Detailed synthesis and characterization of dopants, fabrications and measurements of field-effect transistor, thermoelectric devices. This material is available free of charge via the Internet at http://pubs.acs.org.” AUTHOR INFORMATION Corresponding Author *[email protected], [email protected] Author Contributions ‡These authors contributed equally. Notes The authors declare no competing financial interests. ACKNOWLEDGMENT We thank the National Basic Research Program of China (973 Program) (No. 2014CB643502) for ﬁnancial support, the Strategic Priority Research Program of the Chinese Academy of


18

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


Sciences (XDB12010200), and National Natural Science Foundation of China (91333113, 21572234) REFERENCES (1) Park, T.; Park, C.; Kim, B.; Shin, H.; Kim, E. Flexible PEDOT Electrodes with Large Thermoelectric Power Factors to Generate Electricity by the Touch of Fingertips. Energy Environ. Sci. 2013, 6, 788-792. (2) Mai, C.-K.; Russ, B.; Fronk, S. L.; Hu, N.; Chan-Park, M. B.; Urban, J. J.; Segalman, R. A.; Chabinyc, M. L.; Bazan, G. C. Varying the Ionic Functionalities of Conjugated Polyelectrolytes Leads to Both p- and n-Type Carbon Nanotube Composites for Flexible Thermoelectrics. Energy Environ. Sci. 2015, 8, 2341-2346. (3) Kim, B.; Shin, H.; Park, T.; Lim, H.; Kim, E. NIR-Sensitive Poly(3,4ethylenedioxyselenophene) Derivatives for Transparent Photo-Thermo-Electric Converters. Adv. Mater. 2013, 25, 5483-5489. (4) Bubnova, O.; Berggren, M.; Crispin, X. Tuning the Thermoelectric Properties of Conducting Polymers in an Electrochemical Transistor. J. Am. Chem. Soc. 2012, 134, 16456-16459. (5) Zhang, F.; Zang, Y.; Huang, D.; Di, C.-a.; Gao, X.; Sirringhaus, H.; Zhu, D. Modulated Thermoelectric Properties of Organic Semiconductors Using Field-effect Transistors. Adv. Funct. Mater. 2015, 25, 3004-3012. (6) Bubnova, O.; Khan, Z. U.; Malti, A.; Braun, S.; Fahlman, M.; Berggren, M.; Crispin, X. Optimization of the Thermoelectric Figure of Merit in the Conducting Polymer Poly(3,4ethylenedioxythiophene). Nature Mater. 2011, 10, 429-433. (7) Kim, G. H.; Shao, L.; Zhang, K.; Pipe, K. P. Engineered Doping of Organic Semiconductors for Enhanced Thermoelectric Efficiency. Nature Mater. 2013, 12, 719-723. (8) Sun, Y.; Sheng, P.; Di, C.; Jiao, F.; Xu, W.; Qiu, D.; Zhu, D. Organic Thermoelectric Materials and Devices Based on p- and n-Type Poly(metal 1,1,2,2-ethenetetrathiolate)s. Adv. Mater. 2012, 24, 932-937. (9) Sun, Y.; Qiu, L.; Tang, L.; Geng, H.; Wang, H.; Zhang, F.; Huang, D.; Xu, W.; Yue, P.; Guan, Y. S.; Jiao, F.; Sun, Y.; Tang, D.; Di, C. A.; Yi, Y.; Zhu, D. Flexible n-Type HighPerformance Thermoelectric Thin Films of Poly(nickel-ethylenetetrathiolate) Prepared by An Electrochemical Method. Adv. Mater. 2016, 28, 3351-3358. (10) Schlitz, R. A.; Brunetti, F. G.; Glaudell, A. M.; Miller, P. L.; Brady, M. A.; Takacs, C. J.; Hawker, C. J.; Chabinyc, M. L. Solubility-Limited Extrinsic n-Type Doping of a High Electron Mobility Polymer for Thermoelectric Applications. Adv. Mater. 2014, 26, 2825-2830.


19


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 22

(11) Shi, K.; Zhang, F.; Di, C. A.; Yan, T. W.; Zou, Y.; Zhou, X.; Zhu, D.; Wang, J. Y.; Pei, J. Toward High Performance n-Type Thermoelectric Materials by Rational Modification of BDPPV Backbones. J. Am. Chem. Soc. 2015, 137, 6979-6982. (12) Sumino, M.; Harada, K.; Ikeda, M.; Tanaka, S.; Miyazaki, K.; Adachi, C. Thermoelectric Properties of n-Type C60 Thin Films and Their Application in Organic Thermovoltaic Devices. Appl. Phys. Lett. 2011, 99, 093308. (13) Menke, T.; Ray, D.; Meiss, J.; Leo, K.; Riede, M. In-situ Conductivity and Seebeck Measurements of Highly Efficient n-Dopants in Fullerene C60. Appl. Phys. Lett. 2012, 100, 093304. (14) Barbot, A.; Di Bin, C.; Lucas, B.; Ratier, B.; Aldissi, M. n-Type Doping and Thermoelectric Properties of Co-sublimed Cesium-Carbonate-Doped Fullerene. J. Mater. Sci. 2012, 48, 27852789. (15) Huang, D.; Wang, C.; Zou, Y.; Shen, X.; Zang, Y.; Shen, H.; Gao, X.; Yi, Y.; Xu, W.; Di, C.-a.; Zhu, D. Bismuth Interfacial Doping of Organic Small Molecules for High Performance nType Thermoelectric Materials. Angew. Chem., Int. Ed. 2016, 55, 10672-10675. (16) Russ, B.; Robb, M. J.; Brunetti, F. G.; Miller, P. L.; Perry, E. E.; Patel, S. N.; Ho, V.; Chang, W. B.; Urban, J. J.; Chabinyc, M. L.; Hawker, C. J.; Segalman, R. A. Power Factor Enhancement in Solution-Processed Organic n-Type Thermoelectrics through Molecular Design. Adv. Mater. 2014, 26, 3473-3477. (17) Sun, Y.; Zhang, F.; Sun, Y.; Di, C.-a.; Xu, W.; Zhu, D. N-Type Thermoelectric Materials Based on CuTCNQ Nanocrystals and CuTCNQ Nanorod Arrays. J. Mater. Chem. A. 2015, 3, 2677-2683. (18) Lussem, B.; Keum, C. M.; Kasemann, D.; Naab, B.; Bao, Z.; Leo, K. Doped Organic Transistors. Chem. Rev. 2016, 116, 13714-13751. (19) Qi, Y.; Mohapatra, S. K.; Bok Kim, S.; Barlow, S.; Marder, S. R.; Kahn, A. Solution Doping of Organic Semiconductors Using Air-Stable n-Dopants. Appl. Phys. Lett. 2012, 100, 083305. (20) Guo, S.; Kim, S. B.; Mohapatra, S. K.; Qi, Y.; Sajoto, T.; Kahn, A.; Marder, S. R.; Barlow, S. n-Doping of Organic Electronic Materials Using Air-stable Organometallics. Adv. Mater. 2012, 24, 699-703. (21) Wei, P.; Menke, T.; Naab, B. D.; Leo, K.; Riede, M.; Bao, Z. 2-(2-Methoxyphenyl)-1,3dimethyl-1H-benzoimidazol-3-ium Iodide as a New Air-stable n-Type Dopant for VacuumProcessed Organic Semiconductor Thin Films. J. Am. Chem. Soc. 2012, 134, 3999-4002. (22) Naab, B. D.; Guo, S.; Olthof, S.; Evans, E. G.; Wei, P.; Millhauser, G. L.; Kahn, A.; Barlow, S.; Marder, S. R.; Bao, Z. Mechanistic Study on the Solution-Phase n-Doping of 1,3Dimethyl-2-aryl-2,3-dihydro-1H-benzoimidazole Derivatives. J. Am. Chem. Soc. 2013, 135, 15018-15025.


20

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


(23) Naab, B. D.; Zhang, S.; Vandewal, K.; Salleo, A.; Barlow, S.; Marder, S. R.; Bao, Z. Effective Solution- and Vacuum-processed n-Doping by Dimers of Benzimidazoline Radicals. Adv. Mater. 2014, 26, 4268-4272. (24) Zhang, C.; Zang, Y.; Gann, E.; McNeill, C. R.; Zhu, X.; Di, C. A.; Zhu, D. TwoDimensional π-Expanded Quinoidal Terthiophenes Terminated with Dicyanomethylenes as nType Semiconductors for High-performance Organic Thin-film Transistors. J. Am. Chem. Soc. 2014, 136, 16176-16184. (25) Zhang, C.; Zang, Y.; Zhang, F.; Diao, Y.; McNeill, C. R.; Di, C. A.; Zhu, X.; Zhu, D. Pursuing High-mobility n-Type Organic Semiconductors by Combination of "MoleculeFramework" and "Side-Chain" Engineering. Adv. Mater. 2016, 28, 8456-8462. (26) Wang, S.; Sun, H.; Ail, U.; Vagin, M.; Persson, P. O.; Andreasen, J. W.; Thiel, W.; Berggren, M.; Crispin, X.; Fazzi, D.; Fabiano, S. Thermoelectric Properties of SolutionProcessed n-Doped Ladder-Type Conducting Polymers. Adv. Mater. 2016, 28, 10764-10771. (27) Bubnova, O.; Crispin, X. Towards Polymer-based Organic Thermoelectric Generators. Energy Environ. Sci. 2012, 5, 9345-9362


21


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 22 of 22

Table of Contents

Through precisely regulating the energy level of n-type dopants, high-performance solutionprocessed organic thermoelectric materials are achieved with small-molecule n-type semiconductor, 2DQTT-o-OD. An unexpectedly high power factor of 33.3 µW m−1 K−2 was achieved with (2-Cyc-DMBI-Me)2 featuring the most matched electronic energy level and proper miscibility.


22

Efficient Solution-Processed n-Type Small-Molecule Thermoelectric

Recommend Documents