Solution-processable and functionalizable ultra-high molecular weight polymers via topochemical synthesis

Topochemical polymerization reactions hold the promise of producing ultra-high molecular weight crystalline polymers. However, the totality of topochemical polymerization reactions has failed to produce ultra-high molecular weight polymers that are both soluble and display variable functionality, which are restrained by the crystal-packing and reactivity requirements on their respective monomers in the solid state. Herein, we demonstrate the topochemical polymerization reaction of a family of para-azaquinodimethane compounds that undergo facile visible light and thermally initiated polymerization in the solid state, allowing for the first determination of a topochemical polymer crystal structure resolved via the cryoelectron microscopy technique of microcrystal electron diffraction. The topochemical polymerization reaction also displays excellent functional group tolerance, accommodating both solubilizing side chains and reactive groups that allow for post-polymerization functionalization. The thus-produced soluble ultra-high molecular weight polymers display superior capacitive energy storage properties. This study overcomes several synthetic and characterization challenges amongst topochemical polymerization reactions, representing a critical step toward their broader application.


General
All reactions were carried out in oven-dried glassware sealed with rubber septa under an atmosphere of nitrogen unless otherwise noted and were stirred using Teflon-coated magnetic stir bars. Large volumes of volatile solvents were removed using rotary evaporation, and small volumes of volatile solvents were removed using nitrogen gas flow. All commercially available chemicals were purchased from Alfa Aesar, Spectrum Chemicals, Acros Organics, TCI America, or Sigma-Aldrich, and were used without further purification. Deuterated solvents were purchased from Cambridge Isotope Laboratories and used as received. Solid-state 13 C NMR spectra were carried out at 500.12 MHz (11.7 T) on a Bruker Avance spectrometer with a Bruker 4 mm narrow bore H/C/N magic angle spinning probe. Solid state 13 C NMR spectra were in general acquired by cross-polarization from 1 H with a contact time of 5 ms at a spinning rate of 10 kHz. All solution NMR spectra were recorded at 298 K on a Bruker 500 MHz Avance instrument unless otherwise specified. All chemical shifts are quoted using the δ scale, and all coupling constants (J) are expressed in Hertz (Hz were solved using SHELXT 4 and were refined by least-square refinement against F 2 by SHELXL. 5

Electron Cryomicroscopy (CryoEM)
Microcrystals of polymer P2 were deposited on TEM grids by preparing 10 mg mL -1 solutions of monomer 2 in chloroform, tetrahydrofuran, and dichloromethane. TEM grids were dipped into the solutions, removed, and allowed to air dry for a few seconds (Supplementary Figure 1). Afterwards, the TEM grids were placed in clean vials, and the vials were placed on a windowsill that receives direct sunlight. After three days on the windowsill, the polymerization was complete.
Coated TEM grids were loaded in a Gatan 626 single-tilt cryo holder and cooled with liquid nitrogen. Screening and data collection was performed using a Thermo-Fisher F200C transmission electron microscope operating at 200 kV and equipped with a Thermo-Fisher Ceta-D detector.
Screening was done in low-dose mode and diffraction was identified through condensing of the electron beam. After selection of a crystal on the grid, the crystal was centered, the eucentric height adjusted by tilting the crystal through the desired rotation range, and the selected area aperture and beam stop were inserted. 6 Images were collected in a movie format as crystals were continuously rotated in the electron beam. Typical data collection was performed using a constant tilt rate of 0.3°/s between the minimum and maximum tilt ranges of -72° to +72°, respectively. During continuous rotation, the camera integrated frames continuously at a rate of 3 seconds per frame. The dose rate was calibrated to <0.03 e-/Å 2 s. Crystals selected for data collection were isolated by a selected area aperture to reduce the background noise contributions and calibrated to eucentric height to stay in the aperture over the entire tilt range. All diffraction data were processed using the XDS suite of programs. 7 The resulting crystal structure was solved ab initio using direct methods in SHELXT 4 and refined with SHELXL 8 using ShelXle. 9 Structure refinement was performed using electron scattering factors reported by Peng. 10 Thermal parameters were refined anisotropically for all non-hydrogen atoms.
Supplementary Figure 1. A schematic representation of the preparation of polymer P2 microcrystals on TEM grids for cryoEM.

Dielectric Measurements
Indium tin oxide (ITO) coated glass substrates (2-3 Ω sq -1 , Thin Film Devices Inc., USA) were pre-cleaned using soapy water, deionized (DI) water, acetone, and isopropanol, sequentially. The substrates were then heated at 100 ºC for at least four hours, followed by ultraviolet/ozone (UV/O3) treatment for 20 min prior to use.

Parylene-C Films
A ~1.5 μm thick layer of Parylene-C was deposited on ITO-coated glass substrates using chemical vapor deposition (PDS 2010 Labcoter, SCS) as follows. Parylene-C Dimer (2 g) was placed in the sublimation chamber, which was heated to 175 ºC under high vacuum. The dimer vapor was cleaved to monomer vapor at 690 ºC and flowed into the unheated deposition chamber, which contained the ITO coated glass substrates. The thickness was checked using a Dektak 3030ST profilometer to be between 1.4 μm and 1.6 μm. A gold electrode (4 mm 2 area with 30 nm thickness) was deposited on the top surface of the polymer film using a thermal evaporator (MBRAUN, Germany) for dielectric characterization.

Polymer P3 Films
Polymer P3 (20 mg) was dissolved in 3 mL tetrahydrofuran (THF) and magnetically stirred for 24 h to yield a transparent solution. Afterwards the solution was drop-cast on ITO substrates and they were placed in a covered petri dish to slow down the solvent evaporation process. The coated substrates were then thermally treated at 105 ºC for 24 h in a vacuum oven to remove solvent residual.
The thickness was checked using a Dektak 150 profilometer to be between 2.6 μm and 4.4 μm. A gold electrode (4 mm 2 area with 30 nm thickness) was deposited on the top surface of the polymer film using a thermal evaporator (MBRAUN, Germany) for dielectric characterization.

Crystallizations of Monomers 1-4
Nearly all crystallization conditions attempted on monomers 1-4 presented herein formed similar long aspect-ratio needle or hair-like crystals ( Figure 1, Supplementary Figures 3 and 4). This consistency presents a trade-off, in that it both allows for flexibility in the chosen method of crystallization, while also restricting the morphology of the obtained crystal to a single type. A typical slow evaporation crystallization is detailed as follows. A small amount of the monomer compound was dissolved in a solvent (most commonly CH2Cl2, CHCl3, tetrahydrofuran, or toluene) or combination thereof and filtered through a syringe filter into an aluminum foil-wrapped vial. The vial cap was slightly unscrewed and the solvent was allowed to evaporate over the course of days to weeks in the dark and protected from excessive vibrations. In general, number-averaged molecular weights and crystallinity in the polymer crystals produced from these monomer crystals both increased with higher boiling-point solvents. After generating crystals of 1 via the method detailed above, the vial of crystals was sparged with N2, recapped, the aluminum foil wrapping was removed, and the vial was placed on a sunny windowsill or under a controlled light source. After roughly three days, the crystals had converted from yellow to transparent-white, indicating the conversion to P1. Solid-state (CP-MAS) 13

Polymer P2
After generating crystals of 2 via the method detailed above, the vial of crystals was sparged with N2, recapped, the aluminum foil wrapping was removed, and the vial was placed on a sunny windowsill or under a controlled light source. After the formation of crystals, the vial was sealed, the aluminum foil was removed, and the vial was placed on a sunny windowsill. After roughly three days, the crystals had converted from yellow to transparent-white, indicating the conversion to P2. Solid-state (CP-MAS) 13

Polymer P3
After generating crystals of 3 via the method detailed above, the vial of crystals was sparged with N2, recappd, the aluminum foil wrapping was removed, and the vial was placed on a sunny windowsill or under a controlled light source. After roughly three days, the crystals had converted from yellow to transparent-white, indicating the conversion to P3. The highest molecular weight polymer crystals of  16 evaporated and the polymeric product was resuspended in acetone and precipitated with water. The precipitated solution was filtered, and washed with water, methanol, hexanes, diethyl ether, and acetone before being washed through with dichloromethane (78% recovery). 1  (Theoretical). Intensity axis is scaled for each trace so as to allow for comparison of the data. Note: The peak shifts between P2 experimental and P2 theoretical are ascribed to a temperature effect, since P2 experimental was acquired at 25 ºC while the theoretical trace was based on data acquired at -196 ºC. This temperature effect was supported by variable temperature PXRD results shown in Supplementary Figures 18 and 19.   (8)   The variation rate (Rv) of discharged energy density or charge-discharge efficiency is described as where Ud1 is the discharged energy density and η1 is the charge-discharge efficiency of the 1 st cycle, Udi is the discharged energy density and ηi is the charge-discharge efficiency of the i th cycle, i refers to the number of charge-discharge cycles. The variation rate in discharged energy density is < 1.2% while that in charge-discharge efficiency is < 1.6%, corresponding to outstanding device cyclability under a high electric field.