Unveiling the topology of partially disordered micro-crystalline nitro-perylenediimide with X-aggregate stacking: an integrated approach

Profound knowledge of the molecular structure and supramolecular organization of organic molecules is essential to understand their structure–property relationships. Herein we demonstrate the packing arrangement of partially disordered nitro-perylenediimide (NO2-PDI), revealing that the perylenediimide units exhibit an X-shaped packing pattern. The packing of NO2-PDI is derived using a complementary approach that utilises solid-state NMR (ssNMR) and 3D electron diffraction (3D ED) techniques. Perylenediimide (PDI) molecules are captivating due to their high luminescence efficiency and optoelectronic properties, which are related to supramolecular self-assembly. Increasing the alkyl chain length on the imide substituent poses a more significant challenge in crystallizing the resulting molecule. In addition to the alkyl tails, other functional groups, like the nitro group attached as a bay substituent, can also cause disorder. Such heterogeneity could lead to diffuse scattering, which then complicates the interpretation of diffraction experiment data, where perfect periodicity is expected. As a result, there is an unmet need to develop a methodology for solving the structures of difficult-to-crystallize materials. A synergistic approach is utilised in this manuscript to understand the packing arrangement of the disordered material NO2-PDI by making use of 3D ED, ssNMR and density functional theory calculations (DFT). The combination of these experimental and theoretical approaches provides great promise in enabling the structural investigation of novel materials with customized properties across various applications, which are, due to the internal disorder, very difficult to study by diffraction techniques. By effectively addressing these challenges, our methodology opens up new avenues for material characterization, thereby driving exciting advancements in the field.


Materials:
The samples were synthesized as in the literature, and characterization was done with solution-state NMR, elemental analysis, and IR spectroscopy. 1 3D electron diffraction: 3D Electron diffraction (3D ED) experiments were performed on an FEI Tecnai G 2 20 microscope (200 kV, = 0.0251 Å) with a LaB 6 cathode equipped with a Cheetah ASI direct detection camera (16 bit).The temperature of measurement was 100 K (sample holder tip temperature).Data were measured by 3D electron diffraction method using continuous rotation with integration semi-angle of 0.15°.The powder was directly deposited on the Cu TEM grid.Data were processed with PETS2. 2 Optical distortions were compensated using calibrated values. 3

Solution-State NMR:
The solution-state NMR chemical shift of the monomer is assigned with the help of 1 H, 1 H-1 H COSY (correlation spectroscopy), 1 H-13 C HSQC (heteronuclear single quantum correlation), and 13 C NMR (Figure S2-4).The 1 H-1 H COSY is used to get the proton correlation, whereas the 1 H-13 C HSQC helped to get information on proton-attached carbon.
With the above-mentioned information, all the carbon atoms are assigned in the 13 C solutionstate NMR spectra.To understand the solid-state packing of NO 2 -PDI, solid-state NMR is carried out at room temperature.The 13 C solid-state NMR chemical shift (Figure S4) is assigned with solution-state NMR results.Due to the intricate overlapping of the 13 C signals in the aromatic region, the assignment of quaternary carbon atoms is challenging.We must remember that the molecule's supramolecular packing could somewhat influence the chemical shift of quaternary carbons on the core region.The CH 3 and CH 2 groups in the alkyl tail have a solidstate NMR chemical shift of 11 ppm and 24 ppm, respectively, with a small difference of 1 ppm compared to the solution-state NMR.In the alkyl tail, the CH group has a unique chemical shift rotor.The 1 H-13 C ramp-cross-polarization (rampCP) 4 experiments were performed with a 13 C nutation frequency of 60 kHz, and the 1 H nutation frequency optimized at the +1 Hartman-Hahn condition 5 ~72 kHz.The repetition delay was 2-3 s. 1 H{ 13 C} CP-HETCOR experiments were performed with a contact time of 50 μs and 1.0 ms, and 32-128 t 1 increments consisting of 256-512 scans were collected in each experiment.SPINAL64 decoupling 6 ( 1 H B 1 field of ~83 kHz) was applied during acquisition.DUMBO 7 homonuclear decoupling schemes were employed during the t 1 evolution at ~100 kHz.The 1 H scaling factor for DUMBO HETCOR was corrected using alanine as the standard, and 13 C shift scale were calibrated with adamantane as an external standard.
The 2D data sets were zero-filled to 256 × 2048 points and apodized by an exponential 50 Hz fwhm Lorentzian broadening was applied both dimensions.
Experiments at 100 kHz: -For all experiments in the 0.7 mm MAS probe the 1 H /2 and  pulse lengths were 0.7 and 1.4 s in duration, corresponding to a 360 kHz RF field.T1: The 1 H spin lattice relaxation times were measured using a saturation recovery experiment.
Four scans were collected with a relaxation delay of 5s with τ delays varying from 0.5 ms to 100 s in a pseudo 2D mode.The data was analyzed with the help of Bruker T 1 /T 2 analytical routine.

DFT calculation:
The FORCITE module in the Materials studio was used for the initial relaxation of the structures with the Universal force field.Geometry optimizations were performed using the Smart algorithm with a convergence tolerance energy of 0.0001 kcal mol −1 and force of 0.005 kcal Å −1 with a maximum number of iterations of 500.Plane-wave DFT calculations were done with the gauge-including projected augmented wave (GIPAW) 10,11 approach as implemented in the CASTEP 12 version 2017.Geometry optimization and NMR properties were calculated using the generalized gradient approximation (GGA) with the exchange-correlation PBESOL functional, 13 with On-the-Fly ultra-Pseudopotential. 14And the Tkatchenko and Scheffler method was employed for dispersion corrections. 15An energy cutoff of 630 eV with a Monkhorst−Pack grid 16 with a k-point spacing of 0.07 Å −1 was chosen to maximize the calculation efficiency.
Molecular electrostatic surface potential (ESP) maps were calculated from the cube files generated by Gaussian 13 energy calculations of NO 2 -PDI using GaussView 5.0.8 software. 17 Symmetry Adapted Perturbation Theory (SAPT) Analysis: Symmetry Adapted Perturbation Theory (SAPT): SAPT(0) analysis was employed to determine the non-covalent interaction energies of dimer molecules.The SAPT module of the psi4 code was employed, with aug-cc-pVDZ basis set.SAPT(0) calculations provide the contributing components of interaction energy.The results obtained from SAPT(0) analysis is a second order perturbation expansion constituting first order electrostatic ( ) and exchange energy ( ) parts, and second order dispersion ( ), induction ( ) and their exchange counterparts as the PXRD: Powder X-ray diffraction pattern was collected using Rigaku Smartlab X-ray diffractometer with Cu Kα radiation.

Density Measurement:
The experiments were performed with a Micrometrics Accupyc II 1340 Gas Displacement Pycnometer System with temperature control and a 100 cm 3 nominal cell volume containing 10 cm 3 cups.For each analysis 0.152 g were weighed directly into the cup.The cup with the sample was placed inside the cell, which thereafter was placed inside the sample compartment and sealed.Nitrogen was passed into the sample compartment until reaching an equilibrium rate of 0.0050 psig/min.After pressure stabilization, gas was allowed to expand into the reference compartment.The pressure before and after expansion was measured automatically and used to compute the sample volume.The ratio of known sample mass and volume yields the density.Each analysis took between 45 mins to 1 hour with pressure equilibration being the most time-consuming step.
Conformer Distribution Search: Given the significant conformational flexibility of NO 2 -PDI, the set of low energy conformers of this molecule were determined by performing a conformer distribution search using Spartan'18.The conformer distribution search finds the equilibrium geometry of each conformer by considering rotations around single bonds and puckering of the rings.The calculation was performed using the MMFF molecular mechanics model.During the search a maximum of 500 conformers were requested and only conformers within a relative energy of 40 kJ mol -1 were accepted.The search led to a total of 286 conformers, which were overlayed (Figure S20) to provide a qualitative assessment of the origins of conformational disorder in this molecule.

FigureFigure S6 :Section
Figure S2: 1 H-NMR spectra of the sample collected at room temperature on 500 MHz instrument……………………………………………………………………….SI 17 Figure S3: 1 H-1 H COSY spectrum of the sample collected at room temperature on 500 MHz instrument…………………………………………………………….SI 18 of 58.2 ppm due to the presence of the imide group.The 15C, 15'C, 15aC, 15'aC have an equivalent chemical shift in solid-state and solution-state NMR similarly for 14C, 14'C 14aC, and 14'aC.Like the other carbon atoms in the alkyl tail, 13C and 13aC carbon atoms have identical chemical shifts.The chemical shift of aromatic CH carbons is assigned with the help of one-dimensional cross-polarization experiments (Figure S2-S4).The 5C, 4C, 10aC, and 9aC have a chemical shift of 135, 132, 123, and 135 ppm, respectively in accordance with the corresponding chemical shift observed in the solution-state NMR.Quaternary carbons 7C, 2C, 7aC, 2aC, 3C, 1aC, 1C, and 3aC have a chemical shift from 125 ppm to 135 ppm.The 11C, 12C, 11aC, and 12aC are quaternary carbons with carbonyl group and have a chemical shift in the range of 161.8 to 162.7 ppm, which is in consonance with 163.3 ppm in the solution-state NMR.Quaternary carbon 4a, with a nitro group, has a unique chemical shift of 146 ppm in the solid-state NMR and 147.75 ppm in the solution-state NMR.Further experiments with high spinning speeds are needed to resolve the overlapping peaks of quaternary carbon atoms.The broad peaks observed at 11 and 24 ppm corresponding to alkyl tails indicate disorder.Solid-State NMR: All solid-state NMR spectra were measured with a Bruker Avance HD 600 WB NMR spectrometer (ν 0 ( 1 H) = 600.1 MHz).Finely powdered samples were filled in 0.7 mm, 1.3 mm, and 4.0 mm and spun at MAS rates of 100 kHz, 60 kHz, or 12 kHz, respectively. 1H and 13 C chemical shifts are quoted relative to neat tetramethylsilane (TMS) using adamantane as a secondary chemical shift standard.Experiments at 12 kHz: All experiments performed at 12 kHz used the 4.0 mm double resonance probe head.The fine powder of the samples was packed in a 4.0 mm outer diameter (o.d.) ZrO 2 Photoluminescence Measurements: Photophysical measurements of the derivatives were carried out in a quartz cover slip.The powder sample was packed inside a quartz cover slip and used for the measurements.Emission and excitation spectra were recorded on Horiba Jobin Yvon Fluorolog spectrometer.

Figure S1 :
Figure S1: The structure of NO 2 -PDI and related molecules.

Figure
Figure S2: 1 H-NMR spectra of the sample collected at room temperature on 500 MHz instrument with CDCl 3 solvent.

Figure
Figure S3: 1 H-1 H COSY spectrum of the sample collected at room temperature on 500 MHz instrument with CDCl 3 solvent..

Figure S4 :Figure S5 :
Figure S4: The 1 H-13 C HSQC spectra of the sample showing all the CH carbon atoms collected at room temperature with CDCl 3 solvent.

Figure S6 :
Figure S6: Difference potential at 2 sigma level after constrained core refinement with anisotropic ADPs with inserted MolView optimized molecule to best represent the difference potential features in the peripheral parts of the molecule (A).Difference potential at 2.5 sigma level after the whole molecule constrained refinement (B).The distance between the oxygen atoms of the neighboring nitro groups (dashed line) is only about 2.0 Å.The absolute value of the isosurface level for both parts of the figure is 0.40 e.Å -1 .The unit cells are viewed along [100] direction.

Figure S7 :
Figure S7: The T 1 measurement with monoexponential fit is shown here.The T 1 of aliphatic tails are short indicating the mobility or disorder of the species at room temperature.

Figure
FigureS8: 1 H-1 H DQ-SQ NMR spectrum of NO 2 -PDI collected at 60 kHz spinning speed at room temperature, using the BABA pulse sequence with a recoupling time of one rotor periods corresponding to 16 s.

Figure
FigureS9: 1 H-1 H DQ-SQ NMR spectrum of NO 2 -PDI collected at 100 kHz spinning speed at room temperature, using the BABA pulse sequence with a short recoupling time of one rotor periods.

Figure S10 :
Figure S10:The 1 H-1 H DQ-SQ quantum correlation spectra collected at 100 kHz with a recoupling time of 20 s (two rotor periods).The correlation among aromatic protons is challenging to discern due to the broad peak widths.

Figure S11 :
Figure S11: The two dimensional 1 H{ 13 C} CP-HETCOR NMR spectra of the NO 2 -PDI sample obtained at a contact time of 0.128 ms with a spinning frequency of 12 kHz.The spectra at short mixing time indicates all the carbons attached to proton.

Figure S12 :
Figure S12: The 1 H{ 13 C} CP-HETCOR spectra collected at 60 kHz spinning at room temperature with short contact time of 0.25 ms A) aromatic and B) aliphatic region.

Figure S13 :
Figure S13: The one-dimensional CP-MAS (cross polarization magic angle spinning) spectra of NO 2 -PDI obtained at 0.128 ms and 4 ms.The CP-MAS spectra at long contact time gives indication about the carbon that are far away from the proton.Set of new correlation peaks start observing at long contact time could be either inter or intra molecular correlations.

Figure S14 :
Figure S14: Overlay of all 286 predicted conformers of NO 2 -PDI from the molecular mechanics search indicating that any conformational disorder in this molecule will likely arise from different orientations of the alkyl tail given the significant conformational variability in the orientation of the alky tails.

Figure S15 :
Figure S15: Four cases used in the solid-state NMR studies.Case 1 represents the structure as obtained from 3D electron diffraction and cases 2 to 4 simulate disorder.The molecular conformation is the same in all four cases.

Figure S16 :
Figure S16: The schematic representation of X-aggregate stacking of NO 2 -PDI showing the angle between the prependicualr plane of the molecules as 45.72 degrees.

Figure S17 :
Figure S17: Energy comparison of non X-shaped packing and X-shaped packing.

Figure S18 :
Figure S18: Excitation spectra of NO 2 -PDI in solid-state (powder-black line) and in monomer concentration (solution-red line).Emission spectra of NO2-PDI in solid-state (powder-black dotted line) and in monomer concentration (solution-red dotted line).

Table S5 :
The 1 H-NMR chemical shifts of NO

Table S10 :
Solution and solid-state

Table S2 :
Packing of related molecule from the literature.Summary of the information obtained from various techniques.

Table S3 :
Density of the related compounds from the literature.

Table S4 :
Experimental and calculated 13 C -NMR chemical shift of the NO 2 -PDI molecule.

Table S6 :
The experimental chemical shift of 1 H and calculated chemical shift of four different cases

Table S7 :
Solution and solid-state 13 C-NMR chemical shift assignments of the NO 2 -PDI

Table S10 :
Excitation and emission properties of NO 2 -PDI in solution-state (monomer) and solid-state (aggregate).