A Competitive n‐Type OECT Material via Copolymerization of Electron Deficient Building Blocks

The classical acceptor motifs diketopyrrolopyrrole (DPP) and thienopyrrolodione (TPD) are copolymerized to yield the acceptor–acceptor polymer “Poly(DPP‐TPD).” The fundamental design idea is to maximize the electron affinity (EA), thus increasing the ambient stability of the reduced state against oxygen and water while ensuring high ion compatibility through the incorporation of hydrophilic oligoethylene glycol N‐substituents. Additionally, a highly planarized polymer structure is anticipated, due to the extended noncovalent interactions (conformational locking) between the carbonyl oxygen and the thiophene protons. Cyclic voltammetry, spectroelectrochemistry, ultraviolet photoelectron spectroscopy, ultraviolet‐visible absorption spectroscopy, and organic field effect transistor (OFET) characterization demonstrate the suitability for n‐type organic electrochemical transistor (OECT) devices. High EA, ionization potential, and good electron mobility (µe(OFET)) are shown, in addition to the electrochemical reduction of polymer films in aqueous electrolyte. In n‐type OECTs, poly(DPP‐TPD) demonstrates a moderate threshold voltage of Vth = 0.58 V and an outstanding µC* value of 7.62 F cm−1 V−1 s−1. Cycling studies consisting of pulsed on‐ and off‐switching of the device at gate voltages between Vg = 0.6–0.8 V in the saturation regime reveal high stability for more than 2700 cycles with rapid switching kinetics.

The classical acceptor motifs diketopyrrolopyrrole (DPP) and thienopyrrolodione (TPD) are copolymerized to yield the acceptor-acceptor polymer "Poly(DPP-TPD)." The fundamental design idea is to maximize the electron affinity (EA), thus increasing the ambient stability of the reduced state against oxygen and water while ensuring high ion compatibility through the incorporation of hydrophilic oligoethylene glycol N-substituents. Additionally, a highly planarized polymer structure is anticipated, due to the extended noncovalent interactions (conformational locking) between the carbonyl oxygen and the thiophene protons. Cyclic voltammetry, spectroelectrochemistry, ultraviolet photoelectron spectroscopy, ultraviolet-visible absorption spectroscopy, and organic field effect transistor (OFET) characterization demonstrate the suitability for n-type organic electrochemical transistor (OECT) devices. High EA, ionization potential, and good electron mobility (µ e(OFET) ) are shown, in addition to the electrochemical reduction of polymer films in aqueous electrolyte. In n-type OECTs, poly(DPP-TPD) demonstrates a moderate threshold voltage of V th = 0.58 V and an outstanding µC* value of 7.62 F cm −1 V −1 s −1 . Cycling studies consisting of pulsed on-and off-switching of the device at gate voltages between V g = 0.6-0.8 V in the saturation regime reveal high stability for more than 2700 cycles with rapid switching kinetics.
Recent reports point out the importance of further material parameters, such as threshold voltage (V th ) response time, subthreshold swing, and cycling stability. [9,12,13] In the palette of OMIEC materials, an abundance of highly stable accumulation mode p-type materials with decent performance, as per µC* quantification, is known. [14] The best-performing n-type materials, in contrast, are all derived either from naphthalene diimide (NDI) derivatives or polymers with extended fused-ring systems and exhibit relatively low stability and transconductance. [15,16] Based on these materials, fundamental questions, such as the impact of side chain length, comonomer compositions, and side chain hydrophilicity, were elucidated and understood. [17] However, high-performing n-type polymers for OECT based on further backbone motifs are sparse. Fused-ring systems are in a much earlier stage of development and may therefore help advance the field further. Due to the stiff nature of the fused-ring polymer backbone, the low solubility of those materials is an intrinsic disadvantage of this material class. We thus aimed for a new n-type OMIEC which is neither based on an NDI-based nor fused ring motif. As described previously, the stability issues of n-type polymers originate from insufficiently high electron affinities (or insufficiently low lowest unoccupied molecular orbital (LUMO) levels) of the materials. [18,19] The low µC* of most n-type polymers further was concluded to be caused by low charge carrier mobility. [11] Based on these factors, we designed an acceptor-acceptor copolymer to lower the LUMO, while aiming for a planarized molecular structure. To mimic the bulk properties of ladder-type polymers, we chose two acceptor monomers diketopyrrolopyrrole (DPP) and thienopyrrolodione (TPD), that are known to form planarized structures in the solid state, caused by noncovalent interactions between the thiophene protons and carbonyl oxygen atoms. [6][7][8][9][20][21][22][23] By alternation of hydrogen bond donor and acceptor, a quasi-ladder type polymer in the solid state is anticipated. To ensure ion compatibility as well as sufficient solubility both monomers are equipped with triethylene glycol substituents. [17,24]

Results and Discussion
The monomers DPP and TPD were chosen based on their role as acceptor components in traditional donor-acceptor (D-A) copolymers, which were in turn reported to exhibit high planarity and charge carrier mobility. [23,[25][26][27][28] As previously stated, both monomers were reported to exhibit strong noncovalent interactions, leading to backbone planarization of the respective polymers in the solid state. Such conformational locks were reported to lead to electronic properties, similar to ladder typepolymers. [29] As these interactions are noncovalent and dynamic in nature, the drawback of ladder-type polymers of drastically reduced solubility can be avoided or at least weakened. For DPP, the aromatic unit, attached directly to the dilactam unit, can promote, or disrupt planarization, and additionally influence the optoelectronic properties decisively. [23,27] We equipped the DPP monomer with thiophene flanking units, which are known to exhibit low dihedral angles to the lactam unit. [22] The molecular geometry of such poly(DPP-TPD) copolymers was previously simulated by the group of LeClerc and was shown to exhibit relatively low dihedral angles. [30] This hydrophilic A-A type copolymer based on a DPP-thiophene-TPD backbone is synthesized via a streamlined direct arylation procedure and subsequently characterized with respect to its basic thermal and optoelectronic properties, film morphology, as well as OECT device performance and stability.

Synthesis
The DPP chromophore was synthesized according to Müller et al. with the side chain substitution and bromination procedures from Krauss et al. [23,31] TPD synthesis was adapted and optimized, starting from the procedure of Reynolds et al. [32] For polymerization, direct arylation polymerization (DArP) was employed, following general Ozawa-derived conditions. Previous reports, e.g., by LeClerc were used as starting point for optimization but had to be adapted in respect of reaction solvent, time, temperature, heat source, reactant concentration, and ratio. [30] High conversions and molecular mass polymers were obtained by reactions in aromatic hydrophobic solvents such as toluene or xylenes, while polymerization attempts in polar solvents such as N-methyl-2-pyrrolidon or dimethylacetamide resulted in the formation of oligomers, despite the higher solubility of the monomers. The use of o-xylene with a high reaction molarity of 2.5 mol L −1 enabled microwave-assisted synthesis of the hydrophilic polymer, minimizing polymerization time, and avoiding the use of superheated solvents. Catalyst and oligomer removal was conducted via soxhlet extraction and precipitation, resulting in the polymer poly(DPP-TPD), as evidenced by the spectroelectrochemistry (SEC) trace in Figure 1 and the broad 1 H-NMR signals in Figure S1 in the Supporting Information, characteristic of polymers. The exact synthetic procedures are described in the Experimental Section.

Thermal Properties and Morphology
It is commonly assumed that for OECT application, a balance between crystalline and amorphous domains plays a decisive role, as electronic transport takes place preferentially in the former, ion transport, and storage in the latter. While this is a simplified description of the fundamental processes, it sufficiently demonstrates that crystallinity and morphology play a role in mixed conduction.
To investigate crystallinity as well as melting and crystallization of poly(DPP-TPD), flash differential scanning calorimetry (flash DSC) was employed with heating and cooling rates between 50 and 2000 K s −1 since no transition was observed in conventional DSC. While flash DSC technique is not suitable to determine the real melting point and degree of crystallinity of a material, based on one measurement, its outstanding sensitivity makes it suitable for the detection of weakly pronounced melting transitions. The high scanning speed or short measurement times, enable investigation of thermal transitions above the materials degradation temperature at T 5% = 305 °C (cf. thermogravimetric analysis (TGA), Figure S2a, Supporting Information). As shown in Figure S2b in the Supporting Information, however, no melting or crystallization can be detected, www.advelectronicmat.de which could lead to the impression that the material is either amorphous or exhibits a melting transition at temperatures outside the measurement range. As DSC does not detect crystallinity per se, but rather the associated thermal transitions, a more direct method, namely grazing incidence wide-angle X-ray scattering (GIWAXS), was employed to investigate poly(DPP-TPD) films ( Figure 1b). As-cast polymer films show weakly pronounced 100 and 010 peaks as well as a relatively intensive anisotropic amorphous halo, which qualitatively demonstrates the weakly ordered nature of the polymer in the pristine state. The horizontal and vertical cake cuts depicted in Figure 1c with an integration width of 20° indicate a weak anisotropy of the 010 peak, which corresponds to a modest preferential face-on chain orientation. Upon annealing for 90 min at 200 °C, both 100 and 010 peaks increase significantly in intensity, revealing additional higher order 200 and 300 peaks as well as a pronounced faceon orientation of the polymer crystallites. This is indicative of an increase in order during the annealing procedure. Further, a decreased peak width of all signals is indicative of an increase in coherence length, which in turn is either caused by an increase in crystallite size and/or a decrease in the degree of paracrystalline disorder. The full width at half maximum (FWHM) of the signals as well as the respective coherence lengths determined via the Scherrer equation are listed in Table S1 in the Supporting Information. Additionally, an underlying halo of isotropic amorphous domains can still be observed. Also listed are the 100 and 010 distances that decrease upon annealing, attrib-uted to a thermally enabled contraction of the π-π and lamellar distance, which in turn leads to a decreasing unit cell size. Complementary to GIWAXS, the topology of poly(DPP-TPD) film was investigated based on scanning electron microscopy and atomic force microscopy (AFM) imaging and the results are given in Figure S3 in the Supporting Information. Upon high magnification, only holes in the film, caused by the fastevaporating solvent, are observable. In the AFM phase image of the annealed film, weakly ordered assemblies are visible. However, it is not possible to identify any characteristic lamellar or nanofibril structures usually expected from highly ordered aggregates of conjugated polymers.

Optoelectronic Properties
For alternating D-A conjugated copolymers, the highest occupied molecular orbital (HOMO) level is predominantly determined by the more electron-rich repeating unit, while the LUMO is mainly decided by the more electron-deficient moiety. [21,33] This leads to the typical low optical gap of D-A type polymers and the resulting low energy ground state absorption. Here, however, both units are electron deficient (A-type), possibly rendering the push-pull effect less pronounced than in standard D-A copolymers. To determine the optical properties of the copolymer, UV-vis spectroscopy in solution and solid state was recorded.

www.advelectronicmat.de
As shown in Figure 2a, the polymer absorbs light in the rednear infrared (NIR) region, with a pronounced ground state peak at λ = 670 nm (solution) and λ = 750 nm (film), accompanied by a minor absorption at λ = 380 or 400 nm. This shows that despite the lack of donor units, the acceptor-acceptor copolymers exhibit a sufficiently strong push-pull effect to result in a relatively small optical gap of ≈1.33 eV. Further, both in the solid state and solution, pronounced aggregation features can be seen.
As electron transport is facilitated by the occupation of the LUMO with a generated or injected electron, low electron affinity (EA) is predominantly decisive for such a process of electron injection or acceptance and electron transport. For EA determination, cyclic voltammetry (CV) using a three-electrode setup was employed. To gain representative data for the intended application in OECT devices, the polymers were characterized as films in contact with an aqueous 0.1 m KCl electrolyte. As shown in Figure 2b, reduction and reoxidation were detectable at −0.75 and −0.49 eV versus Ag/AgCl, respectively. From the halfway potential, as well as the reference couple Fc/Fc + , an LUMO level of −4.0 eV was determined. Differential pulse voltammetry (DPV) in acetonitrile containing tetrabutylammonium tetrafluoroborate (TBABF) (0.1 m) indicates an LUMO level of −4.15 eV (cf. Figure S4a, Supporting Information). The small discrepancy between the DPV and CV data can be explained due to various degrees of stabilization of the radical anion in the two different solvent media As evident in Figure 2b, no oxidation was detectable during CV studies of poly(DPP-TPD). To still determine the solid-state ionization potential (IP) of the polymer, ultraviolet photoelectron spectroscopy (UPS) was employed. While this method is not fully comparable with swollen OECT operation conditions, it is a more precise method than CV for the determination of IP in general, as it directly measures the energy levels of occupied valence band states with respect to the vacuum level in the solid state. In Figure 2c, the secondary electron cutoff (SECO) and valence band maximum (VBM) edge of poly(DPP-TPD) with their onset energies are depicted. The spectrometer work function is calibrated such that it is located at 0 eV (see the Supporting Information), hence the work function of poly(DPP-TPD) is calculated by subtracting the absolute position of the SECO from the He I light source energy (21.22 eV). Accordingly, the work function of poly(DPP-TPD) was determined to be (3.9 ± 0.1) eV. Adding the VBM position to the work function then yields the IP. From the VBM and SECO edge onset, the IP was determined to be (5.2 ± 0.1) eV. From CV, DPV, and UPS experiments, a bandgap of ≈1.2 eV can be derived, similar to the optical bandgap of ≈1.33 eV (Figure 2a). The low EA and IP confirm the acceptor-acceptor-like character, significantly lowering both frontier orbital energies compared to donoracceptor copolymers. [34] Electrochemical reduction of poly(DPP-TPD) is accompanied by a change of UV-vis absorption which is monitored via spectroelectrochemical (SEC) analysis. For this, films of poly(DPP-TPD) are deposited on conductive indium tin oxide (ITO) substrates and stepwise reduced in a potential range between V = 0 and V = −0.8 V, followed by reoxidation by increasing the potential to V = 0 V again. At an onset of V = −0.3 V, the ground state absorption (λ = 740 nm) decreases, accompanied by the appearance of a polaron absorption (λ = 1050 nm). In Figure 3a,b, the UV-vis spectra of the electrochemical reduction and reoxidation of poly(DPP-TPD) in aqueous (aq.) 0.1 m KCl are depicted. The difference in absorption (ΔOD) obtained by subtracting the absorption at definite wavelengths of pristine film from reduced/oxidized samples is included in Figure S5 in the Supporting Information. The absolute values of optical density of the pristine film (in Figure 3a) and the reoxidized film (in Figure 3b) at 740 nm are not identical, indicating that the reduced species may undergo some changes leading to spectral changes during the redox process. The polaron absorption, in contrast, vanishes completely. This behavior is further visualized in Figure 3c, where the ΔOD at 740 nm (pristine absorption) and 1050 nm (polaron absorption) are depicted together. While the polaron absorption shows little hysteresis, a pronounced offset is apparent in the plateau values of the ground state absorption maximum at 740 nm. Considering the long measurement time of more than 1 h for one set of SEC data in an aqueous medium under ambient conditions, this can be expected if the reduced species undergo further reactions under these conditions.
To evaluate the suitability of poly(DPP-TPD) as an n-type semiconductor in general, organic field effect transistor (OFET) devices were fabricated. Doped Si bottom gate, gold bottom contact devices with thermally grown SiO x layer, modified with an hexamethyldisilazane (HMDS) monolayer were  Figure S7 in the Supporting Information. Unipolar electron conduction as well as solvent processability opens up further applications, like as electron transport layer (ETL) in organic or perovskite solar cells. As fitting orthogonal solvents exist (dimethylformamide (DMF) for perovskite active layers, toluene for organic active layers), poly(DPP-TPD) can be incorporated in both conventional and inverted stacks. Another possibility is to use such polymers in their n-doped state in thermoelectric devices. [35]

OECT Characterization
Poly(DPP-TPD) fulfills the fundamental conditions necessary for OECT application: OFET and CV studies reveal that unidirectional n-type transport is favored. Further, SEC demonstrated reduction at a relatively low onset voltage of ≈300 mV in aqueous medium. Together, these results suggest the high suitability of poly(DPP-TPD) as n-type material for bioelectronic applications. Hence, the real applicability of poly(DPP-TPD) was put to the test by the fabrication of OECTs with the polymer as active channel material. The architecture of the used devices is depicted schematically in Figure 4a and consisted of interdigitated gold source and drain electrodes, gated with a Ag/AgCl electrode in an aqueous 0.1 m KCl solution.
Drain sweeps in Figure 4b show well-defined n-type output behavior with a pinch-off point at ≈200 mV at V g = 0.8 V and a steady current of I max = 80 µA at drain voltages above 200 mV. For quantification of the aforementioned benchmark values µC* and V th , transfer characteristics were recorded. Linear extrapolation of the square root of the drain current to I d 0.5 = 0 A 0.5 , the X-axis intercept, reveals a threshold voltage of V th = 570 mV, as depicted in Figure S8a in the Supporting Information. By forming the first derivative, dI/dV, the transconductance was calculated, as plotted in Figure 4c. The maximum transconductance of g m = 822 µS presents an extraordinarily high value for n-type materials. For comparative purposes, this value was also normalized by film thickness and the electrode dimensions to obtain a transconductance g m ' = 1.31 S cm −1 . The sub-threshold swing was calculated to be dV = 68 mV dec −1 (cf. Figure S8b, Supporting Information). To determine µC* of poly(DPP-TPD), transfer curves were processed by plotting transconductance against device geometry factor, Wd/L (W and L being width and length of the channel and d being the thickness of the semiconductor layer) yielding the plots in Figure 4d where the slope of the linear regime corresponds to µC* of the material. As indicated in Figure 4, an average µC* of (7.62 ± 0.32) F cm −1 V −1 s −1 and champion value of 8.05 F cm −1 V −1 s −1 was reached. Further, the potential-dependent volumetric capacitance was measured by electrochemical impedance spectroscopy ( Figure S9, Supporting Information). Taking electrochemical impedance spectroscopy (EIS) measurements for C* determination into account, both parameters µ OECT and C* can be deconvoluted, and the electron mobility in the swollen state, µ OECT , determined, as listed in Table 1. Due to the direct inaccessibility of µ OECT , however, this value must only be considered as an approximation. Overall, the low errors of the values, averaged over six devices also demonstrate the excellent reproducibility of the OECT characteristics, as indicated in Figure S10 and Table S2 in the Supporting Information. The details of all OECT devices are given in Table S2 in the Supporting Information.
To put the rest of our findings in perspective, for an n-type OECT, the determined threshold voltage of V th = 0.58 V is relatively low, while the average µC* is outstandingly high. Most state-of-the-art n-type systems are either based on Poly(benz imidazobenzophenanthroline) or poly(naphthalene diimide) (PNDI)-derived D-A copolymers. Despite the high µC* values of the former material of up to µC* = 26 F cm −1 V −1 s −1 , the low processability of the material makes it nonoptimal for applications.. [15] PNDI-based systems, on the other hand, are readily soluble in common organic solvents, and thus became the working horse in the field of n-type OECT research. By variations of the polymer structure, µC* values of 1.29 F cm −1 V −1 s −1 were reached. [36]

OECT Cycling Stability and Response Time
For long-term usage, the operation stability of the active layer material is a vital prerequisite. Prolonged device operation was simulated by repeatedly switching on and off a freshly prepared OECT device with poly(DPP-TPD) as active n-channel material while monitoring the drain current. Devices were cycled in the saturation regime at a drain voltage of V d = 0.4 V (cf. Figure 4).

www.advelectronicmat.de
The gate potential on the same device was increased stepwise from V g = 0.6 V to V g = 0.7 V and finally, V g = 0.8 V. Each cycle consisted of one second without applied drain bias ("off-state"), followed by one second under applied bias ("on-state"). Consequently, 30 min of device sampling corresponds to 900 onoff cycles. After this first 30 min cycling, the gate voltage was increased and the measurement cycles were repeated resulting in a total cycling time of 90 min. Thus, all sampling studies were conducted consecutively on the same device, adding up to a total stress of 2700 on-off cycles (in 90 min). As depicted in Figure 5, a slight decrease in drain current is observable under all sampling conditions. However, as expected, the current retention depends on the gate voltage applied. The higher the gate voltage applied, the lower the current retention, hinting at either side reactions of the reduced species in water and ambient conditions, which agrees qualitatively with the limited reversibility of the reduction process, observed in SEC studies, or kinetic effects limiting the current. While high  [mV]   Table 2. The respective off-current values (I off ) for gate potential, V g of 0.6, 0.7, and 0.8 V are 600, 480, and 150 nA, respectively. Thus on changing the gate potential from 0.6 to 0.7 and 0.8 V, a decreasing trend in I off can be observed. (cf. Figure 5a-c). This behavior can be explained if the gate leakage current during cycling, as depicted in Figure S11 in the Supporting Information, is considered. During the off-state of the transistor, the gate leakage current is relatively high, as injected charges are not efficiently harvested via the drain electrode. Instead it leads to potential-driven current flow to the gate electrode. Upon increase of the gate potential, despite a higher degree of chemical reduction (doping) of the active material, the drain off-current slightly decreases, because the increased gate potential leads to a higher extent of charge carrier leakage toward the gate electrode, instead of the drain electrode.
Under extreme stress conditions of the high number of on-off cycles, the observed current retention values are outstanding in the context of n-type OECT materials. This showcases the viability of acceptor-acceptor materials with low-lying LUMO for OECT applications.
Next, we investigated the response time of a single on-and off-switching event with increased time resolution. To determine the switching speed of poly(DPP-TPD), rectangular potentials were applied simultaneously via the gate and drain electrodes as depicted in Figure S12 in the Supporting Information for V g = V d = 0.7 V and Figure S13 in the Supporting Information for V g = V d = 0.8 V. Neither on-nor off-switching did represent ideal exponential behavior, hinting the interplay of different electrochemical processes such as ion diffusion and polymer reduction. Therefore, the velocity of on-and off-switching were determined at times where 90% of the maximum current was reached during on-switching (τ 90-on ) and 90% decrease of the maximum current for off-switching (τ 90-off ). For a gate and drain potential of 0.7 V, switching times of τ 90-on = 17.2 and τ 90-off = 4.0 ms were observed, which decreased to τ 90-on = 12.5 ms and τ 90-off = 0.6 ms when the increased gate and drain potentials of 0.8 V were  Table 2. Current retention after switching an OECT on and off for 900 consecutive times at each gate potential. Each on-and off period lasted 1 s, resulting in 30 min of stress per step. All measurements were performed successively on the identical device. applied. Rapid switching kinetics as demonstrated for this material here is, e.g., required for circuitry applications or precise time-resolved bioelectronic monitoring.

Conclusion
By the rational combination of the two electron-deficient building blocks DPP and TPD, a new hydrophilic acceptor-acceptor copolymer poly(DPP-TPD) was designed and synthesized using direct arylation polymerization in a microwave reactor. The design was focused on lowering the LUMO level as well as the planarization of the polymer chain by reducing the torsional angles. Monomer synthesis was streamlined starting from literature-known procedures and polymerized subsequently via direct arylation polymerization. By employing microwave-assisted synthesis, low reaction times and temperatures were achieved. The purified polymer was first evaluated concerning its suitability for OECT application based on its optoelectronic properties. Cyclic voltammetry in H 2 O/KCl and differential pulse voltammetry in MeCN/Bu 4 N + BF 4 − showed low LUMO values below 4 eV, suitable for stable operation under ambient conditions. UV-vis spectroelectrochemistry further demonstrated reduction in aqueous media, complemented by moderate OFET electron mobilities of 4.4 × 10 −4 cm 2 V −1 s −1 .
N-type accumulation mode OECT characterization revealed an outstanding average µC* of (7.62 ± 0.32) F cm −1 V −1 s −1 , moderate threshold voltage (575 ± 9) mV, and low sub-threshold swing of (68.2 ± 1.3) mV dec −1 . Further, rapid response kinetics with switching speeds of up to τ 90-on = 12.5 and τ 90-off = 0.6 ms were demonstrated. Dynamic stress studies demonstrate outstanding cycling stability after a total of 2700 cycles and a current retention of 65.0% after 900 cycles at a gate voltage of 800 mV in the saturation regime in aqueous medium under ambient conditions. The high benchmark values, in combination with the streamlined, relatively nontoxic synthesis makes this system optimal for further studies to optimize the stability and performance of n-type mixed conductors.

Synthesis: P(DPP[T] 2 {TEG}-TPD{TEG})/poly(DPP-TPD):
The monomers TPD{TEG} (0.04 g, 0.15 mmol, 1 eq.) and DPP[T] 2 {TEG}-Br 2 (0.113 g, 0.15 mmol, 1 eq.), ligands pivalic acid (4.6 mg, 0.045 mmol, 0.3 eq.) and tris(2-methoxyphenyl)phosphine (4.2 mg, 0.012 mmol, 0.08 eq.), the base Cs 2 CO 3 (0.114 g, 0.35 mmol, 2.32 eq.) and catalyst Pd-Hermann (2.8 mg, 3.0 µmol, 0.02 eq.) were weighed in this order in a microwave vial under ambient conditions. A stir bar was added and the vial was sealed with a septum. The septum is punctured with a canula and placed in a Schlenk flask. After the application of high vacuum on the solids for 15 min, the vessels were backfilled with argon, followed by the removal of the canula with the microwave vial still in the Schlenk flask. 0.6 mL (0.25 m, relative to monomer TPD{TEG}) anhydrous o-xylene is added to the vial and purged for 20 min under rapid stirring. Polymerization is conducted for 2 h at 160 °C under microwave irradiation. Afterward, the polymerization mixture is diluted with chloroform, precipitated in methanol, and filtered through a cellulose Soxhlet thimble. Oligomer, catalyst, and additive residues are removed by extraction with methanol, followed by 2-butanone. The remaining polymer is redissolved in hexafluoro isopropanol and precipitated in water, to ensure quantitative carbonate removal. 0.082 g (62%) polymer is obtained as a black powder.  Figure S1, Supporting Information).

2,5-Bis(2-(2-(2-methoxyethoxy)ethoxy)ethyl)-3,6-di(5-bromothiophen-2-yl)pyrrolo[3,4-c]pyrrole1,4(2H,5H)-dione/DPP[T] 2 {TEG}-Br 2 : A dried
Schlenk flask was charged with DPP[T] 2 {TEG} and kept under high vacuum (HV) for 15 min. After backfilling with argon, anhydrous CHCl 3 was added. The solution was cooled to 0 °C under stirring, the flask was wrapped in aluminum foil and N-bromosuccinimide was added under Ar counter flow. The ice bath was removed and the reaction was allowed to progress at room temperature (RT) for 4 d. After workup with aq. HNaCO 3 (sat.), the solution was concentrated and poured into a large excess of methanol. The solution was stored overnight at −25 °C to encourage crystallization. The crystalline product was filtered and washed with cold methanol. After column chromatography (silica, gradient dichloromethane (DCM) → ethyl acetate, product in second fraction), and subsequent solvent removal, 2.7 g (3.6 mmol, 53%) pure product is obtained as a metallic red powder. 3 mmol, 1 eq.), freshly dried (HV, 500 °C) K 2 CO 3 (7.37 g, 53.3 mmol, 4 eq.) and 18-crown-6 (0.176 g, 0.67 mmol, 0.05 eq.) and kept under HV for 15 min. After carefully backfilling the flask with argon, anhydrous DMF (82.5 mL, 0.16 m, relative to DPP[T] 2 ) was added. The dispersion was heated to 110 °C for 1 h and subsequently cooled to 0 °C, where TEG-Br (9.99 g, 44.0 mmol, 3.3 eq.) was added slowly. The reaction mixture was allowed to heat to room temperature and heated to 130 °C for 24 h. After the reaction mixture reached RT, it was washed with brine three times. The combined aqueous phases were extracted with chloroform. The combined organic phases were dried over MgSO 4 , followed by filtration and complete solvent removal in vacuo. The dry crude product was redissolved in a small amount of chloroform, poured into a large excess of MeOH, and stored at −25 °C overnight. 2.0 g (3.4 mmol, 25%) pure product is obtained as red crystals.  2 : In a three-necked 500 mL round-bottom flask sodium-2-methylbutan-2-oleate (19 g/172 mmol) was dissolved in dry 2-methylbutan-2-ol (70 mL) at 100 °C. Thiophene-2-carbonitrile (12.5 g/115 mmol) was added in one part and diethyl succinate (10 g/57 mmol) was added slowly (3 mL h −1 ) using a syringe-pump and the resulting dark red mixture was stirred overnight at 100 °C. Thereafter, it was cooled to 65 °C, and methanol (200 mL) was added carefully. Acetic acid (11 mL) was added to neutralize the mixture and precipitate the crude product which was filtered off, while still warm. The crude solid was washed with methanol and water, dried at 40 °C in vacuo, and used as obtained. Yield: 8.8 g (29 mmol/51%) of a dark red solid.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.