Single Molecular Layer of Chitin Sub‐Nanometric Nanoribbons: One‐Pot Self‐Exfoliation and Crystalline Assembly into Robust, Sustainable, and Moldable Structural Materials

Abstract Sub‐nanometric materials (SNMs) represent a series of unprecedented size‐/morphology‐related properties applicable in theoretical research and diverse cutting‐edge applications. However, in‐depth investigation and wide utilization of organic SNMs are frequently hindered, owing to the complex synthesis procedures, insufficient colloidal stability, poor processability, and high cost. In this work, a low‐cost, energy‐efficient, convenient, effective, and scalable method is demonstrated for directly exfoliating chitin SNMs from their natural sources through a one‐pot “tandem molecular intercalation” process. The resultant solution‐like sample, which exhibits ribbon‐like feature and contains more than 85% of the single molecular layer (thickness <0.6 nm), is capable of being solution‐processed to different types of materials. Thanks to the sub‐nanometric size and rich surface functional groups, chitin SNMs reveal versatile intriguing properties that rarely observe in their nano‐counterparts (nanofibrils), e.g., crystallization‐like assembly in the colloidal state and alcoplasticity/self‐adhesiveness in the bulk aggregate state. The finding in this work not only opens a new avenue for the high value‐added utilization of chitin, but also provides a new platform for both the theoretical study and practical applications of organic SNMs.


Experimental Section
Material Eriocheir shells and squid pen were purchased from Gold-Shell Pharmaceutical Co. Ltd. (Zhejiang, China). Loligo pen, portunid and lobster shells were collected from local seafood market.
A standardized purification procedure was applied to these raw materials to remove minerals and proteins according to our previous work S1 . The DD of original α-chitin (eriocheir shell) and β-chitin (squid pen) is determined by potentiometric titration to be 3.4% and 3.3%, respectively. DMSO, acid anhydrides, KOH and other reagents were purchased from Sinopharm Chemical Reagent Co.
Ltd. (Beijing, China) and used without purification. Deionized water was utilized in the whole work.

Self-exfoliation of chitin nanoribbons
Pseudosolvent (DMSO saturated with 1 mg/mL KOH) was used to swell purified chitin under room temperature. The concentration of chitin was fixed at 4 mg/mL for β-chitin and 20 mg/mL for α-chitin, respectively. After agitating for designed duration (1~36 h for β-chitin and 24 h for αchitin), phthalic anhydride (PA/chitin structural unit = 0.1~5, mol/mol) was added and intensely stirred for another 15 min. The resultant viscous suspensions were diluted four times with pure DMSO (1 mg/mL for β-chitin and 5 mg/mL for α-chitin), followed by centrifugation at 9500 rpm for 20 min to remove the un-exfoliated samples. The supernatant transparent suspensions, which contained a high portion of monolayered ChNRs were collected and stored at ambient environment (~20 o C). In order to make the subsequent structural characterization (FT-IR, WAXS and solid-state 13C NMR) more accurate, ChNRs were precipitated and repeatedly centrifugally washed (CENCE H1850, Hunan, China) with ethanol. Then, dried and purified samples were obtained by solvent exchange to tertiary butanol and subsequent freeze-drying. The absence of FT-IR signals at 1850 Submitted to 4 and 1780 cm -1 in Figure S2B, Figure S3, Figure S9B and Figure S11B confirm that the purified ChNRs are free of the unreacted phthalic anhydride. The clean NMR spectrum shown in Figure   S11D further supports the above viewpoint.

Fabrication of ChNRs films, straws and their mechanics performance testing
ChNRs films were prepared by a successive process of filtering the suspension through 220 nm pore filter membranes and drying at 70 o C over 6 h. Oriented ChNRs films were obtained by stretching the native film in ethanol solution to a designed draw ratio (1~1.45). Rectangular strips were cut from dried ChNRs film and rolled up on a glass stick. The edges of the samples were brushed by trace DMSO and then sealed by pressing to adhere. After heating at 75 o C over 5 h, the joints were welded and the intact straws could be easily separated from the glass stick. Tensile stress-strain curves were recorded on an Electromechanical Universal Testing Machine (INSTRON 68TM-10) at an elongation rate of 2 mm/min. The Young's modulus and toughness were calculated from the initial slope and the lower area from tensile stress-strain curves.

Characterization
Transmission electron microscopy (TEM) images of ChNRs were recorded on a Hitachi TEM (H-7650) instrument operating at an accelerating voltage of 80 kV. Atomic force microscope (AFM) measurements were used to accurately evaluate the thickness of monolayered ChNRs, using an Agilent 5400 in an intermittent mode at a scan rate of 1 Hz. TEM and AFM samples were prepared by dripping a drop of diluted ChNRs suspensions (dispersed in the pseudosolvent of DMSO/KOH) on the surface of a copper grid and a clean mica sheet, respectively, followed by vacuum-drying under 60 o C. Scanning electron microscopy (SEM) measurements were performed on a Field Emission Scanning Electron Microscope (Hitachi S4800, Japan) operating at a voltage of 3 kV. The samples were treated with sputtering platinum before observation. Before SEM test, the ChNRs suspensions were firstly concentrated into an organogel by vacuum-filtration using a microporous membrane with a pore size of ~220 nm. The SEM samples were then obtained by solvent exchange to tertiary butanol and subsequent freeze-drying. Moreover, the successful and efficient interception of ChNRs further indicate that the modified chitin is in colloidal state rather than molecular state (dissolved in the pseudosolvent).
2D WAXS profiles of all the samples were recorded on a Small Angle X-ray scatterometer (Xeuss 2.0) using a 2D-sensor and Genix 3D X beamline with wavelength (λ) of 1.54 Å. The sample-to-detector distance was fixed at 151.7 mm. The corresponding 1D WAXS curves were obtained by circularly averaging the intensity by the build-in software. In order to prevent the selfassembly of ChNRs, the samples were directly precipitated by ethanol, followed by tertiary butanolexchange and freeze drying. In the case of ethanol treated ChNRs film (alcoplastic film), the sample was totally exchanged to t-BuOH before undergoing a freezing-dry process with liquid nitrogen and a freezer dryer. FT-IR measurements were performed on a Nicolet iS50 Fourier transform infrared spectrometer. Zeta potential of ChNRs aqueous suspensions (0.1 mg/mL, pH = 10.5) was recorded on a Zatasizer Nano-ZS90 (Malven Instruments, UK) at 25 °C. The carboxylate contents of the ChNRs were tested by the electrical conductivity titration method. Before the zeta-potential and titration measurements, the solvent of DMSO was exchanged to aqueous solution via prolonged dialysis. Light transmission of ChNRs suspensions and films was evaluated on a Shimadzu UV3600 UV-Vis spectrophotometer using a quartz cuvette with an optical path of 1 cm.
Rheology measurements were performed on a controlled stress/strain rheometer (TA discovery HR2) with a cone-plate geometry. Dynamic frequency sweep test (0.1 to 100 rad/s) was carried out to compare the rheology behavior of β-chitin/DMSO/KOH suspension (4 mg/mL) before and after phthalation. For each measurement, the strain amplitude was set as 10% and the temperature was fixed at 25 o C. Water contact angle was tested with a water droplet of fixed 2 L on a drop shape Submitted to 6 analysis system (JC2000D1). The reported values were calculated by averaging values measured at five different surface locations. Solid-state 13 C NMR spectra of ChNRs were recorded on a JNM-ECZ600R/S3 spectrometer operated at a 13 C frequency of 100 MHz. The experimental parameters were set as the following: the spinning speed of 12 kHz, the contact time of 5 ms, the acquisition time of 50 ms and the recycle delay of 2 s. The carboxylate content of ChNRs were determined by titration. The Herman's orientation factor (f) was calculated from the azimuthal profile according to the equation of 〈 〉 . The average cosine 〈 〉 was obtained from the following . Table S1. Calculation of energy-consumption and time-consumption for different chitin exfoliation procedures during the pre-treatment process. Table S2. Calculation of energy-consumption and time-consumption for different chitin exfoliation procedures during the disintegration process.     The weak absorbances at 1603 and 1490 cm -1 seen in Figure S3A are attributed to the stretching vibration of aromatic ring skeleton (C=C). Figure