Two-Step Machine Learning Approach for Charge-Transfer Coupling with Structurally Diverse Data.

Electronic coupling is important in determining charge-transfer rates and dynamics. Coupling strength is sensitive to both intermolecular, e.g., orientation or distance, and intramolecular degrees of freedom. Hence, it is challenging to build an accurate machine learning model to predict electronic coupling of molecular pairs, especially for those derived from the amorphous phase, for which intermolecular configurations are much more diverse than those derived from crystals. In this work, we devise a new prediction algorithm that employs two consecutive KRR models. The first model predicts molecular orbitals (MOs) from structural variation for each fragment, and coupling is further predicted by using the overlap integral included in a second model. With our two-step procedure, we achieved mean absolute errors of 0.27 meV for an ethylene dimer and 1.99 meV for a naphthalene pair, much improved accuracy amounting to 14-fold and 3-fold error reductions, respectively. In addition, MOs from the first model can also be the starting point to obtain other quantum chemical properties from atomistic structures. This approach is also compatible with a MO predictor with sufficient accuracy.

Formation of a super-dense hydrogen monolayer on mesoporous silica.

Adsorption on various adsorbents of hydrogen and helium at temperatures close to their boiling points shows, in some cases, unusually high monolayer capacities. The microscopic nature of these adsorbate phases at low temperatures has, however, remained challenging to characterize. Here, using high-resolution cryo-adsorption studies together with characterization by inelastic neutron scattering vibration spectroscopy, we show that, near its boiling point (~20 K), H2 adsorbed on a well-ordered mesoporous silica forms a two-dimensional monolayer with a density more than twice that of bulk-solid H2, rather than a bilayer. Theoretical studies, based on thorough first-principles calculations, rationalize the formation of such a super-dense phase. The strong compression of the hydrogen surface layer is due to the excess of surface-hydrogen attraction over intermolecular hydrogen repulsion. Use of this super-dense hydrogen monolayer on an adsorbent might be a feasible option for the storage of hydrogen near its boiling point, compared with adsorption at 77 K.

High-Mobility Semiconducting Two-Dimensional Conjugated Covalent Organic Frameworks with p-Type Doping.

Two-dimensional conjugated covalent organic frameworks (2D c-COFs) are emerging as a unique class of semiconducting 2D conjugated polymers for (opto)electronics and energy storage. Doping is one of the common, reliable strategies to control the charge carrier transport properties, but the precise mechanism underlying COF doping has remained largely unexplored. Here we demonstrate molecular iodine doping of a metal-phthalocyanine-based pyrazine-linked 2D c-COF. The resultant 2D c-COF ZnPc-pz-I2 maintains its structural integrity and displays enhanced conductivity by 3 orders of magnitude, which is the result of elevated carrier concentrations. Remarkably, Hall effect measurements reveal enhanced carrier mobility reaching ∼22 cm2 V-1 s-1 for ZnPc-pz-I2, which represents a record value for 2D c-COFs in both the direct-current and alternating-current limits. This unique transport phenomenon with largely increased mobility upon doping can be traced to increased scattering time for free charge carriers, indicating that scattering mechanisms limiting the mobility are mitigated by doping. Our work provides a guideline on how to assess doping effects in COFs and highlights the potential of 2D c-COFs to display high conductivities and mobilities toward novel (opto)electronic devices.

Fragment-Based Restricted Active Space Configuration Interaction with Second-Order Corrections Embedded in Periodic Hartree-Fock Wave Function.

We present a computational scheme for restricted-active-space configuration interaction (RASCI) calculations combined with second-order perturbation theory (RASCI-PT2) on a fragment of a periodic system embedded in the periodic Hartree-Fock (HF) wave function. This method allows one to calculate the electronic structure of localized strongly correlated features in crystals and surfaces. The scheme was implemented via an interface between the Cryscor and Q-Chem codes. To evaluate the performance of the embedding method, we explored dissociation of a fluorine atom from a lithium fluoride surface and partially fluorinated graphane layer. The results show that RASCI and RASCI-PT2 embedded in periodic HF are able to produce well-behaved potential energy surfaces and accurate dissociation energies.

Two-Dimensional Boronate Ester Covalent Organic Framework Thin Films with Large Single Crystalline Domains for a Neuromorphic Memory Device.

Despite the recent progress in the synthesis of crystalline boronate ester covalent organic frameworks (BECOFs) in powder and thin-film through solvothermal method and on-solid-surface synthesis, respectively, their applications in electronics, remain less explored due to the challenges in thin-film processability and device integration associated with the control of film thickness, layer orientation, stability and crystallinity. Moreover, although the crystalline domain sizes of the powder samples can reach micrometer scale (up to ≈1.5 µm), the reported thin-film samples have so far rather small crystalline domains up to 100 nm. Here we demonstrate a general and efficient synthesis of crystalline two-dimensional (2D) BECOF films composed of porphyrin macrocycles and phenyl or naphthyl linkers (named as 2D BECOF-PP or 2D BECOF-PN) by employing a surfactant-monolayer-assisted interfacial synthesis (SMAIS) on the water surface. The achieved 2D BECOF-PP is featured as free-standing thin film with large single-crystalline domains up to ≈60 µm2 and tunable thickness from 6 to 16 nm. A hybrid memory device composed of 2D BECOF-PP film on silicon nanowire-based field-effect transistor is demonstrated as a bio-inspired system to mimic neuronal synapses, displaying a learning-erasing-forgetting memory process.

Unveiling Electronic Properties in Metal-Phthalocyanine-Based Pyrazine-Linked Conjugated Two-Dimensional Covalent Organic Frameworks.

π-Conjugated two-dimensional covalent organic frameworks (2D COFs) are emerging as a novel class of electroactive materials for (opto)electronic and chemiresistive sensing applications. However, understanding the intricate interplay between chemistry, structure, and conductivity in π-conjugated 2D COFs remains elusive. Here, we report a detailed characterization for the electronic properties of two novel samples consisting of Zn- and Cu-phthalocyanine-based pyrazine-linked 2D COFs. These 2D COFs are synthesized by condensation of metal-phthalocyanine (M = Zn and Cu) and pyrene derivatives. The obtained polycrystalline-layered COFs are p-type semiconductors both with a band gap of ∼1.2 eV. A record device-relevant mobility up to ∼5 cm2/(V s) is resolved in the dc limit, which represents a lower threshold induced by charge carrier localization at crystalline grain boundaries. Hall effect measurements (dc limit) and terahertz (THz) spectroscopy (ac limit) in combination with density functional theory (DFT) calculations demonstrate that varying metal center from Cu to Zn in the phthalocyanine moiety has a negligible effect in the conductivity (∼5 × 10-7 S/cm), charge carrier density (∼1012 cm-3), charge carrier scattering rate (∼3 × 1013 s-1), and effective mass (∼2.3m0) of majority carriers (holes). Notably, charge carrier transport is found to be anisotropic, with hole mobilities being practically null in-plane and finite out-of-plane for these 2D COFs.

First Principle Prediction of Intramolecular Singlet Fission and Triplet Triplet Annihilation Rates.

Intramolecular singlet fission and triplet-triplet annihilation (TTA) has been experimentally observed and reported. However, problems remain in theoretically accounting for the corresponding intramolecular electronic couplings and their rates. We used the fragment excitation difference (FED) scheme to calculate the coupling with states from restricted active-space spin-flip configuration interaction. We investigated three covalently linked pentacene dimers via a phenyl group in an ortho-, meta-, and para-arrangement. The singlet fission and TTA couplings were enhanced when two chromophores were covalently linked. With the Fermi golden rule, both the estimated singlet fission and TTA rates were in line with the experimental results. For systems with significant singlet-fission coupling, charge-transfer components were observed in the excited states involved, and charge-transfer states were also seen within 1 eV above the singlet excited states. Our approach allows for an analysis of through-bond versus through-space singlet fission in the full electronic wave functions. The FED scheme is useful for calculating intramolecular singlet-fission and TTA couplings.

Using lithium carbonate-based electron injection structures in high-performance inverted organic light-emitting diodes.

A lithium carbonate-based bi-layered electron injection layer was introduced into inverted organic light-emitting diodes (OLEDs) to reduce operation voltages and achieve carrier balance. Ultraviolet photoemission spectroscopy was used to confirm the existence of an interfacial dipole between the organic and lithium carbonate layers, which is a dominating factor related to the device performance. The respective maximum efficiencies of 15.9%, 16.9%, and 8.4% were achieved for blue, green, and red phosphorescent inverted OLEDs with identical architectures, indicating that carrier balance was easily obtained. Moreover, adoption of this sophisticated electron injection layer design resulted in respective turn on voltages of only 3.4 V, 3.2 V, and 3.2 V. Furthermore, the inverted OLEDs equipped with silicon dioxide nanoparticle based light-extraction films achieved an approximately 1.3 fold efficiency improvement over pristine devices due to the low refractive index of the silicon dioxide nanoparticles along with an effective scattering function. The blue, green, and red inverted OLEDs with the nanocomposite layer achieved respective peak efficiencies of 20.9%, 21.3%, and 10.1%.