In recent years, the concept of a fully optical neural network (ONN) has attracted attention for its potential to reach significantly higher speed and lower energy consumption compared to semiconductor electronics-based artificial neural network (ANN). Although the technology promises significant improvements through the utilization of light instead of digital signals, an ONN also requires physical fabrication on a planar substrate that introduces new limitations and constraints to the network when compared to the conventional NN. Limitations on node connectivity, choices of possible nonlinear neuron activation functions, as well as constraints on parameters in the case of a passive-circuit ONN are among the few peculiar challenges that are faced by the ONNs. In this study, we aim to understand the effect of ONNs’ limitations and constraints on their fitting ability when compared to the unconstrained NN. We use digital NNs that simulate the properties of ONNs to investigate the potential and limitations of the ONNs and might be useful as a practical way to pre-calibrate the ONNs’ parameters before their physical production. The ONNs performance and potential are evaluated and compared with the conventional NN by comparing their fitting ability on different datasets. Furthermore, the use-case for ONNs’ application is built and demonstrated on the material and financial datasets.
Development of new functional materials for novel energy conversion and storage technologies is often assisted by ab initio modeling. Such modeling is usually done at the molecular level. Modeling of aggregate state effects is onerous, as packing may not be known or large simulation cells may be required for amorphous materials. Yet aggregate state effects are essential to estimate charge transport rates, and they may also have substantial effects on redox potentials (voltages) and optical properties. We summarize our recent studies of aggregation effects on electronic properties of organic materials used in optoelectronic devices and in organic batteries.
Absorption peak maxima of two organic dyes differing by the position of the methine unit differ by 61 nm in dioxane and by up to 139 nm in polar solvents. It was previously reported that the difference is not reproduced by time-dependent density functional theory (TDDFT) using ab initio or hybrid functionals. TDDFT errors are different between the molecules, leading to a qualitative failure of TDDFT to predict relative energetics of the dyes. We focus on the effect of polar solvents (acetonitrile, DMSO, methanol, and 2-propanol) on the absorption spectrum, specifically, on the different between the two molecules sign of the solvatochromic shift versus dioxane. Using the correction due to Peach et al., the absolute TDDFT errors can be brought within acceptable ranges of 0.2 to 0.3 eV, and the blue shift versus dioxane is reproduced, although both dyes are predicted to exhibit positive solvatochromism. The inclusion of explicit solvent molecules did not appreciably change either TDDFT energies or the correction term. These results show that in dye design by changing the conjugation order, computational errors are expected to be more important than in the case of an extension of the size of conjugation, especially when polar solvents are used.
KEYWORDS: Oxides, Photonics, Solar energy systems, Current controlled current source, Adsorption, Dye sensitized solar cells, Natural surfaces, Titanium dioxide, Oxidation
Adsorption geometry, nuclear vibrations, and molecular orientation of the dye with respect to the oxide surface affect significantly the performance of dye-sensitized solar cells. We compute the influence of these factors on injection and recombination conditions in organic amino-phenyl acid dyes differing by the donor group on the anatase
(101) surface of titania. Nuclear motions affect significantly and differently between the dyes the driving force to injection Δ G.
A temperature increase from 300 to 350 K does not have a noticeable effect on the distribution of injection rates in all studied system.
Molecular dynamics simulations predict configurations in which dyes tend to lay flat on the oxide surface. The resulting proximity of the oxidation equivalent hole to the oxide is expected to promote recombination. Temporal evolution of the driving force to injection is found to be independent of dye orientation and uncorrelated to the oscillations of the Odye Ti bonds through which the dye is attached to the surface. We conclude that the dynamics of Δ G(t) is explained by uncorrelated evolution of the energies of the dye excited state and of the conduction band minimum of the oxide due to their respective vibrations. This suggests that it must be possible to control independently conditions of recombination (e.g. by preventing the dye oxidation hole from approaching TiO2 by using co-adsorbates) and of injection (e.g. by designing dyes where non-equilibrium geometries strongly destabilize dye's LUMO to increase Δ G).
Absorption peak maxima of two organic dyes differing by the position of the methine unit differ by 61 nm in dioxane and 128 nm in acetonnitrile. The difference is not reproduced by TDDFT using ab initio or hybrid functionals. TDDFT errors are different between the molecules due to a different albeit small extent of charge transfer, leading to a qualitative failure of TDDFT to predict relative energetics of the dyes. The TDDFT errors in non-polar solvents (such as dioxane) could be corrected based on the approach of Peach et al. (J. Chem. Phy. 128, 044118 (2008)). Here, we focus on the effect on the absorption spectrum of a polar solvent, specifically of the different between the two molecules sign of the solvatochromic shift vs. dioxane. Using the corrrection due to Peach et al, the absolute TDDFT errors can be brought within accetable ranges of 0.2-0.3 eV with the PCM solvent model, and the blue shift vs.dioxane is reporoduced, although both dyes are predicted to exhibit positive solvatochromism. The inclusion of explicit solvent molecules forming hydrogen bonds with the dye did not appreciably change neither TDDFT energies nor the correction term. These results show the importance of a more careful assessment of computational errors in the strategy of computationaly dye design by changing the conjugation order, where they are expected to be more important than in the case of an extension of the size of conjugation, and more so when polar solvents are used.
We analyze the origin of the large (about 128 nm) difference in the maximum of the visible absorption spectrum of dyes
2-Cyano-3-[5'-(4''-(N,N-dimethylamino) phenyl) thiophen-2'-yl] acrylic acid and Cyano-[5-(4-(N,Ndimethyl-amino)
benzylidene)-5H-thiophen -2-ylidene]-acetic acid which differ by the position of the methine unit that was observed in
an acetonitrile solution. We perform an ab initio analysis of possible factors such as (non-)planarity of the molecule,
isomerization, and solvent effects as well as of the influence of computational parameters. Ground state calculations
failed to account for the difference in transition energies, but excited state optimization of deprotonated dyes in solution
resulted in values comparable to the experiment. We conclude that the most likely explanation for the difference is
different stabilization of the LUMO by the polar solvent.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.