Multi-Dimensional Machine Learning Analysis of Polyaniline Films Using Stitched Hyperspectral ToF-SIMS Data.

The self-organizing map with relational perspective mapping (SOM-RPM) is an unsupervised machine learning method that can be used to visualize and interpret high-dimensional hyperspectral data. We have previously used SOM-RPM for the analysis of time-of-flight secondary ion mass spectrometry (ToF-SIMS) hyperspectral images and three-dimensional (3D) depth profiles. This provides insightful visualization of features and trends of 3D depth profile data, using a slice-by-slice view, which can be useful for highlighting structural flaws including molecular characteristics and transport of contaminants to a buried interface and characterization of spectra. Here, we apply SOM-RPM to stitched ToF-SIMS data sets, whereby the stitched data are used to train the same model to provide a direct comparison in both 2D and 3D. We conduct an analysis of spin-coated polyaniline (PANI) films on indium tin oxide-coated glass slides that were subjected to heat treatment under atmospheric conditions to model PANI as a conformal aerospace industry coating. Replicates were shown to be precisely equivalent, both spatially and by composition, indicating a clear threshold for annealing of the film. Quantitative assessment was performed on the chemical breakdown trends accompanying annealing based on peak ratios, while spectral analysis alone shows only very subtle differences which are difficult to evaluate quantitatively. The SOM-RPM method considers data sets in their totality and highlights subtle differences between samples often simply differences in peak intensity ratios.

Interpretable Machine Learning Models for Phase Prediction in Polymerization-Induced Self-Assembly.

While polymerization-induced self-assembly (PISA) has become a preferred synthetic route toward amphiphilic block copolymer self-assemblies, predicting their phase behavior from experimental design is extremely challenging, requiring time and work-intensive creation of empirical phase diagrams whenever self-assemblies of novel monomer pairs are sought for specific applications. To alleviate this burden, we develop here the first framework for a data-driven methodology for the probabilistic modeling of PISA morphologies based on a selection and suitable adaption of statistical machine learning methods. As the complexity of PISA precludes generating large volumes of training data with in silico simulations, we focus on interpretable low variance methods that can be interrogated for conformity with chemical intuition and that promise to work well with only 592 training data points which we curated from the PISA literature. We found that among the evaluated linear models, generalized additive models, and rule and tree ensembles, all but the linear models show a decent interpolation performance with around 0.2 estimated error rate and 1 bit expected cross entropy loss (surprisal) when predicting the mixture of morphologies formed from monomer pairs already encountered in the training data. When considering extrapolation to new monomer combinations, the model performance is weaker but the best model (random forest) still achieves highly nontrivial prediction performance (0.27 error rate, 1.6 bit surprisal), which renders it a good candidate to support the creation of empirical phase diagrams for new monomers and conditions. Indeed, we find in three case studies that, when used to actively learn phase diagrams, the model is able to select a smart set of experiments that lead to satisfactory phase diagrams after observing only relatively few data points (5-16) for the targeted conditions. The data set as well as all model training and evaluation codes are publicly available through the GitHub repository of the last author.

Automation of liquid crystal phase analysis for SAXS, including the rapid production of novel phase diagrams for SDS-water-PIL systems.

Lyotropic liquid crystal phases (LCPs) are widely studied for diverse applications, including protein crystallization and drug delivery. The structure and properties of LCPs vary widely depending on the composition, concentration, temperature, pH, and pressure. High-throughput structural characterization approaches, such as small-angle x-ray scattering (SAXS), are important to cover meaningfully large compositional spaces. However, high-throughput LCP phase analysis for SAXS data is currently lacking, particularly for patterns of multiphase mixtures. In this paper, we develop semi-automated software for high throughput LCP phase identification from SAXS data. We validate the accuracy and time-savings of this software on a total of 668 SAXS patterns for the LCPs of the amphiphile hexadecyltrimethylammonium bromide (CTAB) in 53 acidic or basic ionic liquid derived solvents, within a temperature range of 25-75 °C. The solvents were derived from stoichiometric ethylammonium nitrate (EAN) or ethanolammonium nitrate (EtAN) by adding water to vary the ionicity, and adding precursor ions of ethylamine, ethanolamine, and nitric acid to vary the pH. The thermal stability ranges and lattice parameters for CTAB-based LCPs obtained from the semi-automated analysis showed equivalent accuracy to manual analysis, the results of which were previously published. A time comparison of 40 CTAB systems demonstrated that the automated phase identification procedure was more than 20 times faster than manual analysis. Moreover, the high throughput identification procedure was also applied to 300 unpublished scattering patterns of sodium dodecyl-sulfate in the same EAN and EtAN based solvents in this study, to construct phase diagrams that exhibit phase transitions from micellar, to hexagonal, cubic, and lamellar LCPs. The accuracy and significantly low analysis time of the high throughput identification procedure validates a new, rapid, unrestricted analytical method for the determination of LCPs.

Antimicrobial, antibiofilm potential, and anti-quorum sensing activity of silver nanoparticles synthesized from Cyanobacteria Oscillatoria princeps.

Cyanobacteria are among the beneficial and environmentally friendly natural candidates used in the biosynthesis of nanoparticles, with their ability to accumulate heavy metals from their environment, thanks to their biologically active compounds. In the current study, an aqueous extract of Oscillatoria princeps fresh biomass was used for the green synthesis of AgNPs. UV-vis spectrum, Fourier transforms infrared, scanning electron microscopy, and energy-dispersive spectroscopy were used to validate and characterize biosynthesized of OSC-AgNPs. The biosynthesis of AgNPs was visually verified in terms of the change in the color of the AgNO3 solution from yellowish brown to brown colors from 72 h onwards. An absorption peak of approximately 420 nm was detected in the UV-vis spectrum, corresponding to the plasmon resonance of AgNPs. FT-IR analysis showed the presence of free amino groups in addition to sulfur-containing amino acid derivatives that act as stabilizing agents. SEM images detected the roughly spherical shape of OSC-AgNPs with an average size of 38 nm. The pathogens tested were all susceptible to OSC-AgNPs showing varying antimicrobial effects on pathogenic microorganisms. E. coli and C. albicans displayed the maximum susceptibility, with zones of inhibition of 14.6 and 13.8 mm at 3-mM concentration, respectively, while B. cereus had the lowest zone of inhibition (10.6 mm) at 3-mM OSC-AgN03 concentration. In conclusion, AgNPs synthesized from Oscillatoria princeps inhibit biofilm formation, suggesting that AgNPs may be a promising candidate for the prevention and treatment of biofilm-associated infections caused by bacteria and yeasts.

Machine learning investigation of viscosity and ionic conductivity of protic ionic liquids in water mixtures.

Ionic liquids (ILs) are well classified as designer solvents based on the ease of tailoring their properties through modifying the chemical structure of the cation and anion. However, while many structure-property relationships have been developed, these generally only identify the most dominant trends. Here, we have used machine learning on existing experimental data to construct robust models to produce meaningful predictions across a broad range of cation and anion chemical structures. Specifically, we used previously collated experimental data for the viscosity and conductivity of protic ILs [T. L. Greaves and C. J. Drummond, Chem. Rev. 115, 11379-11448 (2015)] as the inputs for multiple linear regression and neural network models. These were then used to predict the properties of all 1827 possible cation-anion combinations (excluding the input combinations). These models included the effect of water content of up to 5 wt. %. A selection of ten new protic ILs was then prepared, which validated the usefulness of the models. Overall, this work shows that relatively sparse data can be used productively to predict physicochemical properties of vast arrays of ILs.

Electrochemical Stability of Zinc and Copper Surfaces in Protic Ionic Liquids.

Ionic liquids are versatile solvents that can be tailored through modification of the cation and anion species. Relatively little is known about the corrosive properties of protic ionic liquids. In this study, we have explored the corrosion of both zinc and copper within a series of protic ionic liquids consisting of alkylammonium or alkanolammonium cations paired with nitrate or carboxylate anions along with three aprotic imidazolium ionic liquids for comparison. Electrochemical studies revealed that the presence of either carboxylate anions or alkanolammonium cations tend to induce a cathodic shift in the corrosion potential. The effect in copper was similar in magnitude for both cations and anions, while the anion effect was slightly more pronounced than that of the cation in the case of zinc. For copper, the presence of carboxylate anions or alkanolammonium cations led to a notable decrease in corrosion current, whereas an increase was typically observed for zinc. The ionic liquid-metal surface interactions were further explored for select protic ionic liquids on copper using X-ray photoelectron spectroscopy (XPS) and scanning electron microscopy (SEM) to characterize the interface. From these studies, the oxide species formed on the surface were identified, and copper speciation at the surface linked to ionic liquid and potential dependent surface passivation. Density functional theory and ab initio molecular dynamics simulations revealed that the ethanolammonium cation was more strongly bound to the copper surface than the ethylammonium counterpart. In addition, the nitrate anion was more tightly bound than the formate anion. These likely lead to competing effects on the process of corrosion: the tightly bound cations act as a source of passivation, whereas the tightly bound anions facilitate the electrodissolution of the copper.

Lyotropic liquid crystal phase behavior of a cationic amphiphile in aqueous and non-stoichiometric protic ionic liquid mixtures.

Protic ionic liquids (PILs) are the largest and most tailorable known class of non-aqueous solvents which possess the ability to support amphiphile self-assembly. However, little is known about the effect of solvent additives on this ability. In this study, the lyotropic liquid crystal phase (LLCP) behavior of the cationic surfactant cetyltrimethylammonium bromide (CTAB) was investigated in the model PILs of ethylammonium nitrate (EAN) and ethanolammonium nitrate (EtAN), and derived multi-component solvent systems containing them to determine phase formation and diversity with changing solvent composition. The solvent systems were composed of water, nitric acid and ethylamine (or ethanolamine), with 26 unique compositions for each PIL covering the apparent pH and ionicity ranges of 0-13.5 and 0-11 M, respectively. The LLCPs were studied using cross polarized optical microscopy (CPOM) and small and wide-angle X-ray scattering (SAXS/WAXS). Partial phase diagrams were constructed for CTAB concentrations of 50 wt% and 70 wt% in the temperature range of 25 °C to 75 °C to characterise the effect of surfactant concentration and temperature on the LLCPs in each solvent environment. Normal micellar (L1), hexagonal (H1) and bicontinuous cubic (V1) phases were identified at both surfactant concentrations, and from temperatures as low as 35 °C, with large variations dependent on the solvent composition. The thermal stability and diversity of phases were greater and broader in solvent compositions with excess precursor amines present compared to those in the neat PILs. In acid-rich solvent combinations, the same phase diversity was found, though with reduced onset temperatures of phase formation; however, some structural changes were observed which were attributed to oxidation/decomposition of CTAB in a nitric acid environment. This study showed that the ability of PIL solutions to support amphiphile self-assembly can readily be tuned, and that the ability of PILs to promote amphiphile self-assembly is robust, even with other solvent species present.

Solvation properties of protic ionic liquid-molecular solvent mixtures.

Ionic liquids (ILs) are increasingly receiving interest for a wide range of applications. However, for many applications their cost and/or viscosity can be too high. This can be addressed by using protic ionic liquids as cheaper alternatives, and through mixing with molecular solvents. However, mixing ILs with a molecular solvent adds another dimension to the compositional space, as well as increasing the complexity of solvent-solute interactions. In this study, we have investigated the solvation properties of binary mixtures of PILs with molecular solvents. The selected binary solvent systems are the PILs ethylammonium nitrate (EAN) and propylammonium nitrate (PAN) combined with either water, methanol, acetonitrile or DMSO. In addition, water is combined with the other molecular solvents for comparison. The mole fractions of the secondary solvents were 0, 0.25, 0.5, 0.75, 0.9 and 1 for all combinations, which resulted in a total of 66 solvent mixtures. The solvation properties in each of these mixtures were determined from spectroscopic measurements of 4 well-known solvatochromic probe molecules as solutes. The solvation properties were comparatively investigated, and interpreted, in terms of the specific and non-specific interactions between PIL-solvent, PIL-solute and solvent-solute. All 66 solvent mixtures were also analysed using FTIR with no probe molecules present. In addition, through molecular dynamics simulations, the dye-solvent interactions were simulated for two of the dye molecules in water-EAN binary systems, and the radial distribution functions for the key interactions were obtained. The results showed that the solvation parameters of the binary mixtures deviated considerably from the ideal solvation behaviour. In many cases, bulk compositions and the estimated excess compositions in the solvation shells of the probes were different, suggesting preferential solvation, the extent of which is solute dependent. Our results clearly show that using PILs in a mixture with molecular solvents can strongly enhance the solvation capability.

Structural investigations of molecular solutes within nanostructured ionic liquids.

Ionic liquids (ILs) containing sufficiently long alkyl chains form amphiphilic nanostructures with well-defined polar and non-polar domains. Here we have explored the robustness of these amphiphilic nanostructures to added solutes and gained insight into how the nature of the solute and IL ions affect the partitioning of these solutes within the nanostructured domains of ILs. To achieve this, small angle X-ray scattering (SAXS) investigations were performed and discussed for mixtures of 9 different molecular compounds with 6 different ILs containing imidazolium cations. The amphiphilic nanostructure of ILs persisted to high solute concentrations, over 50 mol% of added solute for most 1-butyl-3-methylimidazolium ILs and above 80 mol% for most 1-decyl-3-methylimidazolium ILs. Solute partitioning within these domains was found to be controlled by the inherent polarity and size of the solute, as well as specific interactions between the solute and IL ions, with SAXS results corroborated with IR spectroscopy and molecular dynamics simulations. Molecular dynamics simulations also revealed the ability to induce π+-π+ stacking between imidazolium cations through the use of these added molecular compounds. Collectively, these results provide scope for the selection of IL ions to rationally influence and control the partitioning behaviour of given solutes within the amphiphilic nanostructure of ILs.

Evaluation of phytotherapeutic activities and phytochemical content of Phormidium autumnale Gomont from natural freshwater sources.

The information available on microalgae-sourced compounds, especially antibiotics and other bioactive compounds, and their potential commercial applications is still insufficient. In this study, antibacterial activity, metabolites, and molecular characterization of Phormidium autumnale, which was isolated from samples collected from different natural freshwater sources in Ankara, Turkey, were investigated. Sequencing results of 16s rDNA confirmed the molecular identification of P. autumnale by 99%. It was determined that the peak values of some phenolic compounds and cyclic peptides were consistent with the 1653-1389 cm-1 band regions in the FTIR spectra of the species. The antibacterial activities of P. autumnale cyanobacteria (CBA) extracts that were obtained by using different solvents were tested on Escherichia coli, Staphylococcus epidermidis, methicillin-resistant (MR) Staphylococcus aureus, Streptococcus agalactiae, and Enterococcus faecalis by using a disc diffusion method. Also, the minimum inhibition concentration (MIC) and antimicrobial indexes of all extracts were determined. It was found that P. autumnale methanol extracts showed antibacterial activity on all test bacteria, whereas acetone extracts showed effects only on E. coli. For the inhibition of MR S. aureus, the control methanol extract was found to give very similar results to those exhibited by the control antibiotics, and the antimicrobial index results were determined to be 58.7-67.5%. According to the results of the analysis of methanol extract, gentisic acid, vanillic acid, 4-hydroxybenzoic acid, p-coumaric acid, and catechin (especially phenolic compounds) were determined to be the active compounds. It can be concluded that P. autumnale is an alternative to current commercial applications as an antibacterial agent in phytotherapy.

Solvation properties of protic ionic liquids and molecular solvents.

Ionic liquids (ILs) are highly tailorable solvents with many potential applications. Knowledge about their solvation properties is highly beneficial in the utilization of ILs for specific tasks, though for many ILs this is currently unknown. In this study, we have investigated the solvation properties of 12 protic ionic liquids (PILs) and 9 molecular solvents based on the Kamlet-Abboud-Taft' (KAT) multi-parameter solvation scales. The KAT parameters, which are dipolarity/polarizability (π*), HBD acidity (α), HBA basicity (ß), and the electronic transition energy (ET) were first obtained for the molecular solvents with an extensive set of 11 solvatochromic probe dye molecules. Based on these results the dyes which exhibited the highest sensitivities to polarity changes, and had the greatest chemical stability, were used to determine the KAT parameters of 12 PILs which contained alkyl-, dialkyl-, alkanol-, or dialkanolammonium cations paired with nitrate, formate or acetate anions. Solvation parameters were also obtained for the PILs using the three fluorescent probes pyrene, Coumarin 153 and Nile red for comparison. The PILs containing nitrate anions showed the greatest polarity, polarizability and HBD acidity followed by those containing formates and acetates. Almost all the PILs were found to have solvation properties comparable to water and single short chain alcohols like methanol and ethanol. The relative order of the IL polarities was similar for the solvatochromic and fluorescent probes. Through this study, in addition to the well-known distinct solvent properties of alkylammonium cation PILs, the high solvation capability of these PILs has been explicitly shown, which makes this class of ILs desirable for solvent-sensitive applications which require high polarity and H bonding ability.

Machine Learning Approaches for Further Developing the Understanding of the Property Trends Observed in Protic Ionic Liquid Containing Solvents.

Ionic liquid containing solvent systems are candidates for very large compositional space exploration due to the immensity of the possible combination of ions and molecular species. The prediction of key properties of such multicomponent solvent systems plays a vital role in the design and optimization of their structures for specific applications. In this study, we have explored two machine learning algorithms for predicting the surface tension and liquid nanostructure of solvents containing a protic ionic liquid (PIL) with water and excess acid or base present. Machine learning algorithms of multiple linear regression (MLR) and Bayesian regularized artificial neural networks (ANNs) were used to develop semiempirical structure-property models for the data set, which was comprised of 207 surface tension and 80 liquid nanostructure data elements which we previously reported ( Phys. Chem. Chem. Phys. 2019, 21, 6810-6827). On the basis of the models, the significance levels for the impact of the alkyl chain length and the presence of hydroxyl groups on cation, type of anion, nonstoichiometry, and presence of water were elucidated. Both models are statistically applicable for designing new PIL containing solvent systems. Furthermore, the generated models were used to create response-surface plots, for both surface tension and liquid nanostructure, interpolated across the compositional space. An additional surface tension data set with 18 new data points within the same compositional space was used to test the prediction ability of models, and the results showed all of the models were successful for prediction. These machine learning approaches are highly suited to the development of structure-property relationships for ionic liquids and particularly for the increasing use of ionic liquid-molecular solvent mixtures.

High throughput approach to investigating ternary solvents of aqueous non-stoichiometric protic ionic liquids.

The use of ionic liquids (ILs) is limited for many applications due to their cost and/or viscosity. An efficient solution is to make mixtures of ILs with molecular solvents. However, it is well known that there are a large number of possible cation and anion combinations resulting in ILs, and this becomes a vast number when these are then combined with a molecular solvent. Therefore, we need structure-property relationships to design new IL-molecular solvent systems. In this work we have applied high throughput methods to investigate IL containing solutions to provide systematic data of a broad compositional space. We have principally focused on the surface tension, apparent pH and liquid nanostructure to identify potential self-assembly and protein stabilizing ability of solvent systems. Non-stoichiometric and aqueous IL-solvents were prepared in a high-throughput manner based on a deliberate experimental design approach such that 26 samples were prepared for each cation-anion-water combination. A selection of 8 protic ionic liquids (PILs) were used as starting materials, comprising ethanol-, ethyl-, butyl-, and pentylammonium cations combined with formate, acetate and nitrate anions. This resulted in a total of 208 different solvent systems. The measured solvent properties showed different trends in base-rich and acid-rich solvent combinations. Surface tensions of base-rich samples exhibited a relatively linear relationship with increasing excess amine, while acid-rich samples were more dominantly affected by the change in water content. Liquid nanostructure of acid-rich samples was retained upon water dilution, whereas a significant SAXS peak shift towards lower scattering angles was observed in the presence of excess amines, indicating larger nanosized aggregates were forming. The design of experiment approach used here is considered to be applicable to any multi-component solvent compositional space due to its suitability in using small data sets to cover large compositional spaces, and hence can be employed to decrease the time and sample quantities required.