Organic Semiconductors Processed from Synthesis‐to‐Device in Water

Abstract Organic semiconductors (OSCs) promise to deliver next‐generation electronic and energy devices that are flexible, scalable and printable. Unfortunately, realizing this opportunity is hampered by increasing concerns about the use of volatile organic compounds (VOCs), particularly toxic halogenated solvents that are detrimental to the environment and human health. Here, a cradle‐to‐grave process is reported to achieve high performance p‐ and n‐type OSC devices based on indacenodithiophene and diketopyrrolopyrrole semiconducting polymers that utilizes aqueous‐processes, fewer steps, lower reaction temperatures, a significant reduction in VOCs (>99%) and avoids all halogenated solvents. The process involves an aqueous mini‐emulsion polymerization that generates a surfactant‐stabilized aqueous dispersion of OSC nanoparticles at sufficient concentration to permit direct aqueous processing into thin films for use in organic field‐effect transistors. Promisingly, the performance of these devices is comparable to those prepared using conventional synthesis and processing procedures optimized for large amounts of VOCs and halogenated solvents. Ultimately, the holistic approach reported addresses the environmental issues and enables a viable guideline for the delivery of future OSC devices using only aqueous media for synthesis, purification and thin‐film processing.

70 81.6 1.78 82 a Detailed synthetic procedures are described below in the method section. b For Mn and Đ values, analytical gel permeation chromatography measurements were performed @ 160 °C in trichlorobenzene of the polymers that were precipitated in methanol. Mn and Đ were calculated using PS as reference. Molecular weight plots for each polymers at different conditions are shown in Figure S2 below. c Low yield is due to low polymer conversion with mainly unreacted starting monomers.   Figure S3. Images of dialysis purification steps: a) The NP dispersions taken from the Schlenk tube after each reaction was injected into a Slide-A-Lyzer Dialysis Cassettes with a 2000 molecular weight cut-off (MWCO) permeation membrane; b) the cassette is completely immersed in deionised (DI) water and left to stir over a period of 72 hrs. DI water was replaced every 12 hours; c) Approximately 200 µL of the dispersionis taken out at every time interval to characterise the level of surfactant removed via TGA, and DLS measurements for stability of the NP dispersions. At the end of the dialysis process, the remaining NP dispersion was removed from the cassette, and filtered through a 1µm glass Acrodisc filter before processing into thin-films for OFETs.              a Volume of organic solvents are estimated for soxhlet extraction based on the same scale of reaction as described in the methods section in the main text b 100mL required to immerse the dialysis cassete containing 10 mL of the NP dispersion for purification from dialysis. This volume of water is replenished at least 7 times over a period of 72hrs in accordance to Figure S4.

Experimental Methods
Materials: All reagents were purchased from Sigma-Aldrich, Alfa Aesar, Acros and were used without further purification unless indicated. Solvents were purchased from Sigma-Aldrich or Alfa Aesar. Deionised (DI) water was used for dialysis. Aqueous solution used for the polymerization reactions were prepared using DI water that was bubbled with Argon gas for 72 hours prior to use. Monomer synthesis of M1 and M2 were achieved according to literature procedures. 32a 2,1,3-Benzothiadiazole-4,7-bis(boronic acid pinacol ester) (BT(Bpin)2) was purchased from Ossila Ltd and used without further purification.

Synthesis Characterization. 1 H and 13 C NMR spectroscopy experiments were recorded on
Bruker Avance 400 spectrometer. Chemical shifts are reported in parts per million (ppm) relative to the residual solvent (for instance CHCl3 in CDCl3 δ H = 7.26 ppm, δ C =77.36 ppm).
The multiplicity of the signals is reported using the following abbreviations: s=singlet, d=doublet, t=triplet, dd=doublet of doublet, ddd=doublet of doublet of doublets, ddt= doublet of doublet of triplets, m=multiplet. Gel permeation chromatography (GPC) was performed on Agilent 1260 Infinity II system at 160 °C using 1,3,5-trichlorobenzene as an eluent. The average molecular weight in number (Mn), in weight (Mw), and weight average dispersity Đ (Mw/Mn) was determined using narrow weight average dispersity (Đ < 1.10) polystyrene (PS) standard. GPC samples were prepared by precipitation of the crude product from the respective reactions in excess methanol. The solid was isolated by centrifugation and the supernatant decanted several times before drying under vacuum at 40 °C.

NP Dispersion Characterization. Dynamic Light Scattering (DLS) measurements using a
Malvern Zetasizer Nano ZS instrument were used to determine the particle size of the NPs dispersed in water. An aliquot (5 µL) of the NP dispersion was diluted into a 1 ml volume of DI water before measuring each sample at 25°C. The DLS results are quoted from an average of three measurements. Thermogravometric analysis (TGA) measurements were performed on a Discovery SDT650 TGA from TA instruments. The samples were measured at 10 °C/min under a N2 atmosphere in an alumina pan from 30°C to 800 °C. Samples were prepared by sequential addition of 20 µL x 10 (Total = 200 µL) of the respective NP dispersions into the alumina pans left at 80°C to allow the solvent (water) to evaporate to obtain a minimum mass of approximately 2 mg. The concentration of each respective NP (aq) dispersion was estimated from the measured wt% loss of the polymer with respect to the amount of water. For PIDTBT (aq) and PDPPTBT (aq), 1.8 mg (90 %) and and 1.9 mg (95 %) was measured to give an estimated NP (aq) concentration of 0.9 wt% and 1.0 wt%, respectively.
General procedure for mini-emulsion polymerization synthesis. Firstly, SDS (75 mg) and the relevant base (3 molar equivalents) were dissolved in 10 ml of DI water in an argon purged Schlenk tube. The aqueous solution was stirred and bubbled with argon gas for a further 30 minutes to ensure complete dissolution of the reagents. The organic phase was prepared in a separate argon purged Schlenk tube by first adding the relevant catalyst, ligand, particle stabilizer (hexadecane), and monomer M1 or M2 and BT(Bpin)2 at the corresponding stoichiometric amounts for each polymer reaction. The reagents were then dissolved in 1 mL of anhydrous toluene and subsequently bubbled with argon gas for 5 minutes. Next, the organic solution was injected into the Schlenk tube containing the aqueous solution and this was transferred into an ice bath. The sonicator tip was then submerged into the reaction mixture under an argon purge, and sonicated for a total of 4 minutes (2 x 2 minutes) using a Sonics VCX-750 Vibra Cell ultrasonicator at 22% amplitude and a 6.4 mm sonicator tip. The Schlenk tube was then immediately sealed and placed into a heated oil bath at the corresponding reaction temperature and stirred for 24 h. The reaction was cooled to 40 °C, left unsealed under argon flow for 2-3 h to evaporate the toluene solvent to give the NP dispersion. This dispersion was then injected into a Thermo Scientific Slide-A-Lyzer dialysis cassette (2K MWCO), immersed in DI water and left to stir over a period of 72 hours. The DI water was replaced every 12 hours.
After dialysis, the volume of the solution had the tendency to increase slightly to approximately 12mL. The NP dispersions were left to stir in a Schlenk tube under an argon flow to evaporate the excess water until the total volume returned to 10mL. The resulting NP dispersion was used for further characterization and processing into thin-films. PIDTBT (aq). The synthesis was performed according to the general procedure described above. K2CO3 (31.4 mg, 0.22 mmnol) and SDS (75 mg) were dissolved in 10 mL of DI water to form the aqueous phase. M1 (100 mg, 75.7 µmol), BT(Bpin)2 (29.3 mg, 75.7 µmol) , Pd2(dba)3 (1.4 mg, 1.5 µmol), tri(otolyl)phosphine (1.2mg, 3.8 µmol), and 78 µL of hexadecane were dissolved in 1 mL of toluene to form the organic phase. Emulsification was achieved by ultrasonication and the reaction was subsequently heated at the respective temperature. At the end of the reaction and after purification via dialysis, a dark blue NP dispersion of PIDTBT (aq) was obtained. Refer to Figure S15 and S16 for representative NMR spectrum of the isolated PIDTBT (aq). DPPTBT (aq). The synthesis was performed according to the general procedure described above. NaOH (12mg, 303 µmol) and SDS (75 mg) were dissolved in 10 mL of DI to form the aqueous phase.

M2
(103.0 mg, 101 µmol), BT(Bpin)2 ( (39.2 mg, 101 µmol), tetrakis(triphenylphosphine)palladium (2.2 mg, 1.0 µmol), and 78 µL of hexadecane were dissolved in 1 mL of toluene to form the organic phase. Emulsification was achieved by ultrasonication and the reaction was subsequently heated at the respective temperature. At the end of the reaction and after purification via dialysis, a dark blue NP dispersion of PDPPTBT(aq) was obtained. Refer to Figure S17 and S18 for representative NMR spectrum for the isolated PIDTBT (aq).
General procedure for conventional polymerization synthesis. Three equivalents of the respective base and a drop of Aliquat 336 were dissolved in 1 mL of DI water in an argon purged Schlenk tube. In a separate tube, the respective monomer M1 or M2 and BT(Bpin)2, the relevant catalyst and ligand were added at the corresponding stoichiometric amounts for each polymer reaction. The reagents were then dissolved in 4 mL of anhydrous toluene and subsequently bubbled with argon gas for 5 minutes. Next, the organic solution was injected into the Schlenk tube containing the aqueous solution and this was transferred into a heated oil bath at the specified temperature. The reaction was left stirring for 24 hrs under an argon atmosphere.
After which the reaction was cooled to room temperature and excess methanol was added into Cr/Au bottom electrodes were thermally evaporated under high vacuum (10 -7 mbar) using a metal shadow mask with a patterned channel width = 1000 µm and length = 60 µm. The substrates were then treated with oxygen plasma for 3 minutes, to generate a hydrophilic surface to ensure sufficient wetting of the aqueous NP dispersions. NP dispersions were deposited on the substrate and left for 60s before spin coating at 500 rpm for 60 s and 6000 rpm for 10s to form the thin-film. The thin-films were annealed at 150°C for 30 minutes and cooled down to room temperature for approximately 30 minutes. The thickness of the resulting polymer semiconductor thin-films were 50-60 nm as measured by contact profilometer (Dektak XT).
For the surfactant washing step, Ethanol was dropped over the surface of the thin-film and left for 60s before spin-washing at 6000 rpm for 30s, and the substrates were annealed at 100°C for 5 minutes and left in a vacuum chamber for 15 minutes to allow the removal of any remaining solvent. For DCB-processed devices, PIDTBT and PDPPTBT were solubilized in anhydrous DCB at 5 mg/mL and stirred overnight at 60 °C, filtered through a 1 µm filter before spin coating at 1000 rpm in a N2 purged glovebox. The dielectric layer was then spin-coated over the semiconducting thin-films from a solution of poly(methyl methacrylate) (PMMA, Mw = 120,000 Da) in butyl acetate (80 mg/mL), and annealed at 80 °C for 30 minutes. Finally, an aluminum gate electrode (60 nm) was thermally evaporated on the semiconductor/PMMA films using a metal shadow mask. The electrical properties of the OFETs were measured in the dark under ambient condition using an Agilent B1500 semiconductor parameter analyzer. The fieldeffect mobility (μ) was extracted using equation (1) in the saturation regime from the drain current versus gate voltage sweep: where ID is the drain current in the saturated regime, μsat is the field-effect mobility, Ci is the capacitance per unit area of the gate dielectric layer, VG and VT are gate voltage and threshold voltage and W and L are channel width and length, respectively. The estimated capacitance Ci = 5.78 nF cm -2 of PMMA was measured at 100kHz using an Agilent 4284A precision LCR meter. See Figure S9 for a schematic description of the processing procedure of the NP dispersions.

Statistical analysis.
The μ values were estimated from transfer curves as representatively shown in Figure S10-S13. The curves were obtained by measurement of a total of 12 transistors fabricated in three independent batches for each processing conditon to obtain individual mean values and the standard deviation. Values for μ are presented in Table S2, and as a box and whisker plot in Figure 3e. The rectangular box is determined by the 25th and 75th percentiles and the whiskers are determined by the 5th and 95th percentiles. The square box is mean and the line is the median.     36, 134.29, 130.40, 123.99, 76.30, 75.99, 75.67, 45.33, 36.74, 30.91, 30.87, 30.14, 29.15, 28.96, 28.68, 28.63, 28.36, 25.18, 22.41, 21.67, 13.10.