Novel Triphenylamines as Hole-Transport Materials in OLEDs &Cathode Materials in Organic Batteries

Summary:

Novel Triphenylamines as Hole-Transport Materials in OLEDs & Cathode Materials in Organic Batteries

Eva-Maria Gatzka, Ronald Alle, Claudia Dillmann, Stephanie Rüth, Robert Herzhoff, Daniele Fazzi, Lennard Höfer and Klaus Meerholz*
Department of Chemistry, University of Cologne, Greinstraße 4-6, 50939 Köln

Introduction
Novel and especially more sustainable materials for organic electronics are of great interest due to rapidly and globally increasing demand for energy.
Triphenylamines (TPAs) have been commonly used as hole-transport materials (HTM) in organic light-emitting diodes (OLEDs).1 Here, they benefit from excellent chemical stability of the “holes”, i.e. radical cations, as evidenced from electrochemical performance. Their HOMO energy level (redox potential) can be precisely tuned in a certain range.
Furthermore, polymeric TPAs enable the use of inexpensive solution based processing techniques like spincoating or ink-jet printing.1,2 TPAs containing oxetane-groups can be crosslinked via cationic ring-opening polymerization (CROP) to insoluble polymer networks, enabling solution processing of multilayer devices.3
Here, three novel crosslinkable TPAs are synthesized, characterized via optical and electrochemical methods and tested as HTMs in an OLED stack. Moreover, via spectroelectrochemistry the absorption spectra of the TPA radical cations are recorded and compared with density functional theory (DFT) calculations. Additionally, one of the TPAs is tested as organic cathode material in a battery coin cell setup.
 
Results & Discussion
Synthesis
The novel compounds, consisting of different triarylamine-based motifs, were synthesized via palladium-catalyzed Hartwig-Buchwald cross-coupling reactions.4–7 So far, most oxetane-functionalized HTMs in our group were synthesized based on triphenyldimer (TPD) systems. These systems possess hexyl or hexyloxyl groups between the functional aromatic core and the oxetane functional group to ensure both, high solubility in organic solvents and high diffusional mobility for the oxetane groups during the crosslinking reaction.1,2,8 The novel triphenylamine-based molecules have the advantage of an simplified scalable synthesis route compared to the TPD systems. They can be divided in TPAs containing one oxetane functionalized chain and TPAs containing two oxetane functionalized chains. Furthermore, the variation of the para-positions of the phenyl moieties leads to a HOMO energy level tunability of the novel TPAs. TPAs containing one oxetane chain can be copolymerized with different epoxy- or oxetan-containing resins like resin SU8, 165 or 1,6-bis{(3-ethyl-3-oxetanyl)methoxy}hexane (HBO) to yield insoluble polymer networks. Up to now, three target compounds were synthesized:
1. 4-({6-[(3-ethyloxetan-3-yl)methoxy]hexyl}oxy)-N,N-bis(4-methoxyphenyl)aniline; 4-({6-[(3-ethyloxetan-3-yl)methoxy]hexyl}oxy)-N-[4-({6-[(3-ethyloxetan-3-yl)methoxy]hexyl}oxy)cyclohexyl]-N-(4-methoxycyclohexyl)cyclohexan-1-amine (1)
2. 4-({6-[(3-ethyloxetan-3-yl)methoxy]hexyl}oxy)-N,N-bis(4-methylphenyl)aniline; 4-({6-[(3-ethyloxetan-3-yl)methoxy]hexyl}oxy)-N-[4-({6-[(3-ethyloxetan-3-yl)methoxy]hexyl}oxy)cyclohexyl]-N-(4-methoxycyclohexyl)cyclohexan-1-amine (2)
3. 4-({6-[(3-ethyloxetan-3-yl)methoxy]hexyl}oxy)-N,N-bis(4-substituted phenyl)aniline; 4-({6-[(3-ethyloxetan-3-yl)methoxy]hexyl}oxy)-N-[4-({6-[(3-ethyloxetan-3-yl)methoxy]hexyl}oxy)phenyl]-N-(4-methoxyphenyl)aniline (3)

Compound 3 possesses two oxetane functionalities and, therefore, needs no comonomer for polymerization.
 
Spectroscopy
UV-Vis absorption and photoluminescence (PL) spectra were recorded in solution (dichloromethane). The absorption maxima for all three TPAs were at 305 nm. Interestingly, the photoluminescence peak of TPA 2 with a value of 386 nm is blue-shifted compared to the other two TPAs (1 and 3 both at 402 nm).

Organic Light-Emitting Diodes
The TPA materials in this work were synthesized as hole-transport materials (HTMs) in organic light-emitting diodes (OLEDs). To put the following results into perspective, a short introduction into OLED construction and working principles will be given here.
For a simple OLED setup, a glass substrate with transparent anode layer is used on which an electroluminescent active layer has been deposited. Via vapour deposition a metal cathode is deposited on top. By applying an external voltage exceeding the work function difference of the electrode materials, charges (electrons and holes) are injected into the organic semiconductor. Both charges migrate through the transporting layers and recombine forming so-called excitons. Ultimately, excitons decay radiatively by emitting light. Overall, the process can be unravelled into the following basic steps: charge injection, charge transport, recombination, light emission and outcoupling of light. As it would go beyond the scope of this abstract, the underlying physical principles of the individual steps will not be discussed here.
A highly efficient OLED needs to fulfil several requirements like a high photoluminescence quantum yield, the HOMO and LUMO energy levels of the used organic material need to match the ones of the electrodes, and the organic material must exhibit high hole and electron mobilities. Due to these requirements a multilayer OLED is often used. By utilization of hole- (HTL) and electron- (ETL) transport layers a more efficient and balanced charge injection into the emissive layer is obtained, due to lowered injection barriers at the respective electrodes.
In this work, the OLED stack consists of a glass substrate covered with indium tin oxide (ITO; 140 nm) as anode, followed by a HIL consisting of PEDOT:PSS (35 nm) and the synthesized TPAs as HTL (approximately 50 nm). On top, the emitter material poly[9,9-dioctylfluorenyl-2,7-diyl)-alt-(benzo[2,1,3]thiadiazol-4,7-diyl)] (F8BT; 160 nm) is applied, followed by a cathode consisting of CsF/Al (2/60 nm).
Solution-processed fabrication of OLEDs has several advantages in view of fast processing and cost-efficiency. Here, PEDOT:PSS is spincoated onto the ITO layer from an aqueous suspension. Due to the solvent resistance of PEDOT:PSS regarding organic solvents a second layer (like TPA) can be deposited on top from THF solution by spincoating. A common method for crosslinking organic semiconductors was developed in our group. The functionalization of small molecules or polymers with at least two oxetane groups enables the possibility of a polymerization via cationic ring opening polymerization (CROP), which is photochemically initiated in presence of small amounts of a photo acid generator (PAG) like (4-(octyloxy)phenyl)(phenyl)iodonium hexafluoroantimonate (OPPI).3 The resulting 3D-networks lead to insoluble layers. Here, TPAs with only one oxetane functionality were copolymerized with resins containing several oxetane and/or epoxy groups in this step. The emissive layer (F8BT) was also spincoated, and finally the cathode was evaporated on top of the organic layers in a high-vacuum chamber by use of a shadow mask.
To compare the OLED devices with a reference, a device with crosslinked N4,N4′-bis(4-(6-((3-ethyloxetan-3-yl)methoxy)hexyloxy)phenyl)-N4,N4′-bis(4-methoxyphenyl)biphenyl-4,4′-diamine (QUPD) (24 nm) and N4,N4′-bis(4-(6-((3-ethyloxetan-3-yl)methoxy)hexyl)phenyl)-N4,N4′-diphenylbiphenyl-4,4′-diamine (OTPD) (29 nm) as HTLs were used. In all other stacks TPA 2 crosslinked with either resin SU8 or resin 165 was used as HTL. Both are commercially available solid multifunctional cresol novolac epoxy resins containing eight (resin SU8) or 5 to 6 (resin 165) epoxide groups per molecule. By analysing the three parameters current density [mA/cm2], brightness [Cd/m2] and efficiency [Cd/A] of four different OLED devices, each containing 5 wt% or 25 wt% of the respective resin and comparing them to a reference device with QUPD/OTPD several conclusions can be drawn:
First, a lower content of co-monomer/resin enhances the performance of the device (ergo: devices with 5 wt% resin perform better than devices with 25 wt% resin).
Second, devices containing resin SU8 perform better than devices with resin 165. This could be because resin 165 possesses five to six reactive epoxide groups whereas resin SU8 possesses eight and therefore will copolymerize to a denser network together with TPA.
Third, a reduced performance of all stacks with TPA as HTL can be observed compared to the reference stack. A possible reason for that could be the use of two energetic steps (QUPD and OTPD) in the reference HTL. Further experiments featuring multilayer stacks out of new TPAs have to be conducted to test this assumption. 
DFT calculations
DFT and time-dependent DFT (TD-DFT) calculations were performed by considering the range-separated functional B97X-D and the triple-split Pople basis set with diffuse functions 6-311G*. Solvent effects were included implicitly via the conductor polarizable continuum model (CPCM, solvent: CH2Cl2). Mono-cation species (+1) were computed at the spin-polarized unrestricted UDFT level, while the dication (+2) at the restricted one. For the dication species, a further wavefunction stability analysis was performed in order to find possible instabilities which would lead to broken symmetry (BS) solutions. All dication species were found to be stable in the singlet state.
The mono- and dication excited state transitions of two compounds 4-methoxy-N,N-bis(4-methylphenyl)aniline [short: (Me)2(MeO)TPA] and 4-methoxy-N,N-bis(4-methoxyphenyl)aniline [short: (MeO)3TPA] have been computed at the TD-DFT level. By comparison of the spectroelectrochemical data of 1 and 2 with the calculated absorption spectra several observations could be made: (Me)2(MeO)TPA is isoelectronic with 2 as well as (MeO)3TPA with 1. TD-DFT spectra are slightly blue-shifted compared to experimental data, as expected considering the level of theory adopted; overall a good qualitative agreement is achieved. The most intense band for both mono-cation species of 1 (730 nm) and 2 (720 nm) can be assigned to a transition between frontier Singly Occupied/Unoccupied Molecular Orbitals, namely SOMO  SUMO transition. The other absorption bands at higher energies are described at the TD-DFT level as multiple transitions between singly occupied/unoccupied orbitals, possibly leading to a multi-reference character of the excited states.

Spectroelectrochemistry
For these measurements, TPA 1 and 2 were spincoated from solution on ITO substrates and crosslinked by CROP together with the resin SU8. The resulting thin films on ITO were measured in deaerated and water free dichloromethane solution with 0.25M tetrabutylammonium hexafluorophosphate (TBA PF6) as supporting electrolyte against the ferrocene/ferrocenium (Fc/Fc+) redox couple with a scan rate of 2 mV/s, and an absorption spectra was recorded simultaneously.
Compound 1 showed two characteristic absorption bands at 375 and 730 nm for its oxidized form and compound 2 showed three absorption bands at 370, 585 and 720 nm. After reduction these characteristic bands vanished again, indicating a reversible redox reaction. Therefore, the spectroelectrochemical measurements proved the reversible formation of the TPA monocations (1 and 2, respectively) with characteristic absorption bands, in agreement with the calculated TD-DFT data.

Cyclic voltammetry (CV)
Cyclic voltammograms of thin films on ITO were measured as described above. The HOMO energies of the HTMs were determined by measuring the oxidation potentials of the TPAs via CV, which is especially important regarding the OLED layer construction. To obtain the HOMO energy levels, the obtained oxidation levels have to be correlated to the vacuum level using the following equation:
EHOMO = -5.1 eV – EOx [1]9
where -5.1 eV corresponds to the HOMO energy level of the Fc/Fc+ reference.
This leads to the following HOMO energy levels for the three TPA compounds: EHOMO (1) = -5.224 eV; EHOMO (2) = -5.364 eV and EHOMO (3) = -5.246 eV.
The novel TPAs show one reversible redox reaction, whereas
commonly in our workgroup used molecules like 1-[Bis4-[N,N-di(4-tolyl)amino]phenyl]-cyclohexanes (TAPCs) and triphenylamine-dimers (TPDs) show two reversible redox reactions in the investigated potential range.

Organic Batteries
In organic batteries organic pi-conjugated materials are used as active material for the cathode, which comes along with a great variety of available redox-active organic compounds. A main difference compared to inorganic batteries is the simple redox reaction instead of complex intercalation mechanisms, leading to high rate performances and long life cycles.10
The charge and discharge performance of most organic electrodes is initially tested vs. a lithium metal anode due to the simple one-electron reaction and low redox potential (-3.04 V vs. SHE) of lithium as well as the fast reaction, which provides a bottleneck-free measurement of the kinetic properties of the organic cathode. Therefore, the electrodes are placed in an ion-conducting electrolyte and separated by a porous ion-permeable membrane.
TPA 3 was tested as cathode material in a coin cell setup. The cathode consisted of 60% carbon black, 30% 3 and 10% poly(vinylidene difluoride) (PVDF) binder, and lithium metal was used as anode material. 1M LiPF6 in a 1:1 mixture of ethylene carbonate (EC) and dimethyl carbonate (DMC) was used as electrolyte.
The theoretical specific capacity of 38 mAh/g was calculated by dividing the Faraday constant through the molar mass of 3. The kinetic properties of the batteries were evaluated by charging and discharging the batteries at different current and therefore different C-rates. The C-rate is the inverse of the time needed for the charging or discharging of the battery, for example a current, that charges the battery in 15 min (0.25 h) corresponds to a C-rate of 4.

 
Summary and Outlook
TPAs are a promising and versatile material class for use as HTM in OLEDs. It is shown that the crosslinking of TPAs containing one oxetane functional group is possible with various co-monomers containing two or more oxetane or epoxy functions, while the different substituents of the TPAs in para-position can tune the energetic levels of the molecule.
Spectroelectrochemical data show the reversible formation of radical cations, and the spectra are in agreement with the theoretical DFT calculations even though not exactly the same side chains have been considered.
Implementation of the novel materials in an OLED stack and a first implementation as cathode material in an organic battery was successfully demonstrated.
 
Literature
1. Zacharias, P., Gather, M. C., Rojahn, M., Nuyken, O. & Meerholz, K. New crosslinkable hole conductors for blue-phosphorescent organic light-emitting diodes. Angew. Chemie – Int. Ed. 46, 4388–4392 (2007).
2. Liaptsis, G. & Meerholz, K. Crosslinkable TAPC-based hole-transport materials for solution-processed organic light-emitting diodes with reduced efficiency roll-off. Adv. Funct. Mater. 23, 359–365 (2013).
3. Rudati, P. S., Mueller, D. C. & Meerholz, K. Preparation of Insoluble Hole-Injection Layers by Cationic Ring-Opening Polymerisation of Oxetane-Derivatized TriPhenylamineDimer for Organic Electronics Devices. Procedia Chem. 4, 216–223 (2012).
4. Hartwig, J. F. Recent advances in palladium- and nickel- catalyzed chemistry provide new ways to construct C À N and C À O bonds. Angew. Chem. Int. Ed. 37, 2046–2067 (1998).
5. Yamamoto, T., Nishiyama, M. & Koie, Y. Palladium-catalyzed synthesis of triarylamines from aryl halides and diarylamines. Tetrahedron Lett. 39, 2367–2370 (1998).
6. Surry, D. S. & Buchwald, S. L. Biarylphosphanliganden in der palladiumkatalysierten Aminierung. Angew. Chemie 120, 6438–6461 (2008).
7. Surry, D. S. & Buchwald, S. L. Dialkylbiaryl phosphines in Pd-catalyzed amination: A user’s guide. Chem. Sci. 2, 27–50 (2011).
8. Bacher, E. et al. Synthesis and Characterization of Photo-Cross-Linkable Hole-Conducting Polymers. 1640–1647 (2005).
9. Cardona, C. M., Li, W., Kaifer, A. E., Stockdale, D. & Bazan, G. C. Electrochemical considerations for determining absolute frontier orbital energy levels of conjugated polymers for solar cell applications. Adv. Mater. 23, 2367–2371 (2011).
10. Muench, S. et al. Polymer-Based Organic Batteries. Chem. Rev. 116, 9438–9484 (2016).