Machine learning for accuracy in density functional approximations.

Machine learning techniques have found their way into computational chemistry as indispensable tools to accelerate atomistic simulations and materials design. In addition, machine learning approaches hold the potential to boost the predictive power of computationally efficient electronic structure methods, such as density functional theory, to chemical accuracy and to correct for fundamental errors in density functional approaches. Here, recent progress in applying machine learning to improve the accuracy of density functional and related approximations is reviewed. Promises and challenges in devising machine learning models transferable between different chemistries and materials classes are discussed with the help of examples applying promising models to systems far outside their training sets.

Number of sites-based solver for determining coverages from steady-state mean-field micro-kinetic models.

Kinetic models parameterized by ab-initio calculations have led to significant improvements in understanding chemical reactions in heterogeneous catalysis. These studies have been facilitated by implementations which determine steady-state coverages and rates of mean-field micro-kinetic models. As implemented in the open-source kinetic modeling program, CatMAP, the conventional solution strategy is to use a root-finding algorithm to determine the coverage of all intermediates through the steady-state expressions, constraining all coverages to be non-negative and to properly sum to unity. Though intuitive, this root-finding strategy causes issues with convergence to solution due to these imposed constraints. In this work, we avoid explicitly imposing these constraints, solving the mean-field steady-state micro-kinetic model in the space of number of sites instead of solving it in the space of coverages. We transform the constrained root-finding problem to an unconstrained least-squares minimization problem, leading to significantly improved convergence in solving micro-kinetic models and thus enabling the efficient study of more complex catalytic reactions.

Interpretable Machine Learning Models for Practical Antimonate Electrocatalyst Performance.

Computationally predicting the performance of catalysts under reaction conditions is a challenging task due to the complexity of catalytic surfaces and their evolution in situ, different reaction paths, and the presence of solid-liquid interfaces in the case of electrochemistry. We demonstrate here how relatively simple machine learning models can be found that enable prediction of experimentally observed onset potentials. Inputs to our model are comprised of data from the oxygen reduction reaction on non-precious transition-metal antimony oxide nanoparticulate catalysts with a combination of experimental conditions and computationally affordable bulk atomic and electronic structural descriptors from density functional theory simulations. From human-interpretable genetic programming models, we identify key experimental descriptors and key supplemental bulk electronic and atomic structural descriptors that govern trends in onset potentials for these oxides and deduce how these descriptors should be tuned to increase onset potentials. We finally validate these machine learning predictions by experimentally confirming that scandium as a dopant in nickel antimony oxide leads to a desired onset potential increase. Macroscopic experimental factors are found to be crucially important descriptors to be considered for models of catalytic performance, highlighting the important role machine learning can play here even in the presence of small datasets.

Effect of home-based online training and activity feedback on oxygen uptake in patients after surgical cancer therapy: a randomized controlled trial.

BACKGROUND: Exercise training is beneficial in enhancing physical function and quality of life in cancer patients. Its comprehensive implementation remains challenging, and underlying cardiopulmonary adaptations are poorly investigated. This randomized controlled trial examines the implementation and effects of home-based online training on cardiopulmonary variables and physical activity. METHODS: Of screened post-surgical patients with breast, prostate, or colorectal cancer, 148 were randomly assigned (1:1) to an intervention (2 × 30 min/week of strength-endurance training using video presentations) and a control group. All patients received activity feedback during the 6-month intervention period. Primary endpoint was change in oxygen uptake after 6 months. Secondary endpoints included changes in cardiac output, rate pressure product, quality of life (EORTC QoL-C30), C-reactive protein, and activity behavior. RESULTS: One hundred twenty-two patients (62 intervention and 60 control group) completed the study period. Change in oxygen uptake between intervention and control patients was 1.8 vs. 0.66 ml/kg/min (estimated difference after 6 months: 1.24; 95% CI 0.23 to 2.55; p = 0.017). Rate pressure product was reduced in IG (estimated difference after 6 months: - 1079; 95% CI - 2157 to - 1; p = 0.05). Physical activity per week was not different in IG and CG. There were no significant interaction effects in body composition, cardiac output, C-reactive protein, or quality of life. CONCLUSIONS: Home-based online training among post-surgery cancer patients revealed an increase of oxygen uptake and a decrease of myocardial workload during exercise. The implementation of area-wide home-based training and activity feedback as an integral component in cancer care and studies investigating long-term effects are needed. TRIAL REGISTRATION: DRKS-ID: DRKS00020499 ; Registered 17 March 2020.

Data-driven and constrained optimization of semi-local exchange and nonlocal correlation functionals for materials and surface chemistry.

Reliable predictions of surface chemical reaction energetics require an accurate description of both chemisorption and physisorption. Here, we present an empirical approach to simultaneously optimize semi-local exchange and nonlocal correlation of a density functional approximation to improve these energetics. A combination of reference data for solid bulk, surface, and gas-phase chemistry and physical exchange-correlation model constraints leads to the VCML-rVV10 exchange-correlation functional. Owing to the variety of training data, the applicability of VCML-rVV10 extends beyond surface chemistry simulations. It provides optimized gas phase reaction energetics and an accurate description of bulk lattice constants and elastic properties.

Atom-Specific Probing of Electron Dynamics in an Atomic Adsorbate by Time-Resolved X-Ray Spectroscopy.

The electronic excitation occurring on adsorbates at ultrafast timescales from optical lasers that initiate surface chemical reactions is still an open question. Here, we report the ultrafast temporal evolution of x-ray absorption spectroscopy (XAS) and x-ray emission spectroscopy (XES) of a simple well-known adsorbate prototype system, namely carbon (C) atoms adsorbed on a nickel [Ni(100)] surface, following intense laser optical pumping at 400 nm. We observe ultrafast (∼100 fs) changes in both XAS and XES showing clear signatures of the formation of a hot electron-hole pair distribution on the adsorbate. This is followed by slower changes on a few picoseconds timescale, shown to be consistent with thermalization of the complete C/Ni system. Density functional theory spectrum simulations support this interpretation.

Acoustic Propulsion of Nano- and Microcones: Dependence on the Viscosity of the Surrounding Fluid.

This article investigates how the acoustic propulsion of cone-shaped colloidal particles that are exposed to a traveling ultrasound wave depends on the viscosity of the fluid surrounding the particles. Using acoustofluidic computer simulations, we found that the propulsion of such nano- and microcones decreases strongly and even changes sign for increasing shear viscosity. In contrast, we found only a weak dependence of the propulsion on the bulk viscosity. The obtained results are in line with the findings of previous theoretical and experimental studies.

Symmetry-resolved CO desorption and oxidation dynamics on O/Ru(0001) probed at the C K-edge by ultrafast x-ray spectroscopy.

We report on carbon monoxide desorption and oxidation induced by 400 nm femtosecond laser excitation on the O/Ru(0001) surface probed by time-resolved x-ray absorption spectroscopy (TR-XAS) at the carbon K-edge. The experiments were performed under constant background pressures of CO (6 × 10-8 Torr) and O2 (3 × 10-8 Torr). Under these conditions, we detect two transient CO species with narrow 2π* peaks, suggesting little 2π* interaction with the surface. Based on polarization measurements, we find that these two species have opposing orientations: (1) CO favoring a more perpendicular orientation and (2) CO favoring a more parallel orientation with respect to the surface. We also directly detect gas-phase CO2 using a mass spectrometer and observe weak signatures of bent adsorbed CO2 at slightly higher x-ray energies than the 2π* region. These results are compared to previously reported TR-XAS results at the O K-edge, where the CO background pressure was three times lower (2 × 10-8 Torr) while maintaining the same O2 pressure. At the lower CO pressure, in the CO 2π* region, we observed adsorbed CO and a distribution of OC-O bond lengths close to the CO oxidation transition state, with little indication of gas-like CO. The shift toward "gas-like" CO species may be explained by the higher CO exposure, which blocks O adsorption, decreasing O coverage and increasing CO coverage. These effects decrease the CO desorption barrier through dipole-dipole interaction while simultaneously increasing the CO oxidation barrier.

MCML: Combining physical constraints with experimental data for a multi-purpose meta-generalized gradient approximation.

The predictive power of density functional theory for materials properties can be improved without increasing the overall computational complexity by extending the generalized gradient approximation (GGA) for electronic exchange and correlation to density functionals depending on the electronic kinetic energy density in addition to the charge density and its gradient, resulting in a meta-GGA. Here, we propose an empirical meta-GGA model that is based both on physical constraints and on experimental and quantum chemistry reference data. The resulting optimized meta-GGA MCML yields improved surface and gas phase reaction energetics without sacrificing the accuracy of bulk property predictions of existing meta-GGA approaches.

Ultrafast Adsorbate Excitation Probed with Subpicosecond-Resolution X-Ray Absorption Spectroscopy.

We use a pump-probe scheme to measure the time evolution of the C K-edge x-ray absorption spectrum from CO/Ru(0001) after excitation by an ultrashort high-intensity optical laser pulse. Because of the short duration of the x-ray probe pulse and precise control of the pulse delay, the excitation-induced dynamics during the first picosecond after the pump can be resolved with unprecedented time resolution. By comparing with density functional theory spectrum calculations, we find high excitation of the internal stretch and frustrated rotation modes occurring within 200 fs of laser excitation, as well as thermalization of the system in the picosecond regime. The ∼100 fs initial excitation of these CO vibrational modes is not readily rationalized by traditional theories of nonadiabatic coupling of adsorbates to metal surfaces, e.g., electronic frictions based on first order electron-phonon coupling or transient population of adsorbate resonances. We suggest that coupling of the adsorbate to nonthermalized electron-hole pairs is responsible for the ultrafast initial excitation of the modes.

Specific Reaction Parameter Density Functional Based on the Meta-Generalized Gradient Approximation: Application to H2 + Cu(111) and H2 + Ag(111).

Specific reaction parameter density functionals (SRP-DFs), which can describe dissociative chemisorption reactions on metals to within chemical accuracy, have so far been based on exchange functionals within the generalized gradient approximation (GGA) and on GGA correlation functionals or van der Waals correlation functionals. These functionals are capable of describing the molecule-metal surface interaction accurately, but they suffer from the general GGA problem that this can be done only at the cost of a rather poor description of the metal. Here, we show that it is possible also to construct SRP-DFs for H2 dissociation on Cu(111) based on meta-GGA functionals, introducing three new functionals based on the "made-simple" (MS) concept. The exchange parts of the three functionals (MS-PBEl, MS-B86bl, and MS-RPBEl) are based on the expressions for the PBE, B86b, and RPBE exchange functionals. Quasi-classical trajectory (QCT) calculations performed with potential energy surfaces (PESs) obtained with the three MS functionals reproduce molecular beam experiments on H2, D2 + Cu(111) with chemical accuracy. Therefore, these three non-empirical functionals themselves are also capable of describing H2 dissociation on Cu(111) with chemical accuracy. Similarly, QCT calculations performed on the MS-PBEl and MS-B86bl PESs reproduced molecular beam and associative desorption experiments on D2, H2 + Ag(111) more accurately than was possible with the SRP48 density functional for H2 + Cu(111). Also, the three new MS functionals describe the Cu, Ag, Au, and Pt metals more accurately than the all-purpose Perdew-Burke-Ernzerhof (PBE) functional. The only disadvantage we noted of the new MS functionals is that, as found for the example of H2 + Cu(111), the reaction barrier height obtained by taking weighted averages of the MS-PBEl and MS-RPBEl functionals is tunable over a smaller range (9 kJ/mol) than possible with the standard GGA PBE and RPBE functionals (33 kJ/mol). As a result of this restricted tunability, it is not possible to construct an SRP-DF for H2 + Ag(111) on the basis of the three examined MS meta-GGA functionals.

Frequency and profile of Parkinson's disease prodromi in patients with malignant melanoma.

OBJECTIVE: The results of register studies suggest an association between Parkinson's disease (PD) and melanoma. We studied the frequency and profile of early markers of PD in patients with malignant melanoma. METHODS: 100 participants were enrolled in a prospective observational study, of whom 65 had a history of high-risk cutaneous (n=53) or uveal (n=12) melanoma (31 women; age, 61.2±14.9 years) and another 35 served as control participants (19 women; 54.6±20.5 years). Participants underwent assessments of motor function (Unified PD Rating Scale; keyboard tapping test), olfactory function, colour vision, depressive symptoms, the Non-Motor Symptoms Questionnaire, and transcranial brain sonography. Raters were blinded to the diagnosis and clinical data of study participants. RESULTS: Patients with melanoma showed increased frequency of substantia nigra hyperechogenicity and prodromal motor and non-motor features of PD, especially asymmetric motor slowing and apathy. Hyposmia and colour vision disturbance were, however, infrequent. Larger echogenicity of substantia nigra correlated with lower serum iron in patients with melanoma, similar to previously reported findings in PD, and independently from the earlier findings, with lighter skin pigmentation. Substantia nigra hyperechogenicity, combined with motor asymmetry or hyposmia, was present at baseline in all participants with mild or definite parkinsonism diagnosed after 1 year. Parkinsonism was specifically related to melanoma location at the sun-exposed skin of the head or neck. CONCLUSIONS: History of melanoma was associated with increased prevalence of prodromal markers of PD. Their predictive value needs to be established in long-term investigations. The similarity of serum iron characteristics found in patients with melanoma and PD deserves further research.

Density Functional Tight-Binding Models for Band Structures of Transition-Metal Alloys and Surfaces across the d-Block.

First-principles electronic structure simulations are an invaluable tool for understanding chemical bonding and reactions. While machine-learning models such as interatomic potentials significantly accelerate the exploration of potential energy surfaces, electronic structure information is generally lost. Particularly in the field of heterogeneous catalysis, simulated electron band structures provide fundamental insights into catalytic reactivity. This ab initio knowledge is preserved in semiempirical methods such as density functional tight binding (DFTB), which extend the accessible computational length and time scales beyond first-principles approaches. In this paper we present Shell-Optimized Atomic Confinement (SOAC) DFTB electronic-part-only parametrizations for bulk and surface band structures of all d-block transition metals that enable efficient predictions of electronic descriptors for large structures or high-throughput studies on complex systems outside the computational reach of density functional theory.

[eHealth applications for promotion of physical activity after visceral surgery : A systematic review]. / eHealth-Anwendungen zur Förderung körperlicher Aktivität nach viszeralen Operationen : Eine systematische Übersichtsarbeit.

BACKGROUND: eHealth applications can support early mobilization and physical activity (PA) after surgery. This systematic review provides an overview of eHealth services to enhance or record PA after visceral surgery interventions. METHODS: Two electronic databases (MEDLINE PubMed and Web of Science) were systematically searched (November 2023). Articles were considered eligible if they were controlled trials and described digital devices used to promote PA after visceral surgery. The Cochrane risk of bias (RoB-2) tool was used to determine the methodological quality of studies. RESULTS: A total of nine randomized controlled studies (RCT) were included in this systematic review. The studies differed with respect to the interventions, surgical indications and evaluation variables. The risk of bias of the individual studies was moderate. The six studies using activity trackers (AT) predominantly showed insignificant improvements in the postoperative step count. The more complex fitness applications could partially reveal significant advantages compared to the control groups and the home-based online training also showed a significant increase in functional capacity. CONCLUSION: Activity tracking alone has so far failed to show clinically relevant effects. In contrast, the more complex eHealth applications revealed advantages compared to usual postoperative care. More high-quality studies are needed for evidence-based recommendations for eHealth services in conjunction with visceral surgery.

Adherence to long-term telemonitoring-supported physical activity in patients with chronic heart failure.

Chronic heart failure (CHF) is one of the most common diseases with a prevalence of 1-2% in adults, disproportionately affecting the elderly. Despite consistent drug therapy, physical activity (PA) is an integral part of current guidelines. Yet adherence to regular PA and exercise interventions is poor and potential predictors and barriers to PA remain elusive. We examined the effects of a telemonitoring-based exercise intervention in 699 CHF patients in a prospective, randomized-controlled (1:1), multicenter trial. The study was registered in the German Clinical Trials Register under DRKS00019022 on 28.05.2020. For both, the exercise and control group, self-reported PA (MET*h/week) increased and sedentary behavior declined during the 12-month intervention period. In the exercise group, daily step count as analyzed via activity trackers remained stable (pre: 6459 [4016] steps/day, post: 6532 [3858] steps/day; p = 0.621). The average number of completed exercise instruction videos provided via an online application was 1.50 [1.44] videos/week at the beginning and gradually decreased to 1.00 [1.50] videos/week; p < 0.001). Multivariate regression model revealed that exercise-related PA (MET*h/week) and exercise capacity (Wmax) at baseline, CHF severity, atrial fibrillation and age predicted changes in self-reported exercise-related PA (R2 = 0.396). Furthermore, the BMI and the average number of completed videos per week at baseline were associated with the change in completed videos over the course of the study (R2 = 0.251). Our results show the influence of certain baseline characteristics as barriers and predictors of PA progression. Therefore, exercise programs should pay attention to patients' individual conditions to set achievable goals, and eventually affect the adherence and sustainability of exercise-focused interventions.

Thermionic current densities from first principles.

We present a density functional theory-based method for calculating thermionic emission currents from a cathode into vacuum using a non-equilibrium Green's function approach. It does not require semi-classical approximations or crude simplifications of the electronic structure used in previous methods and thus provides quantitative predictions of thermionic emission for adsorbate-coated surfaces. The obtained results match well with experimental measurements of temperature-dependent current densities. Our approach can thus enable computational design of composite electrode materials.

Dependence of the acoustic propulsion of nano- and microcones on their orientation and aspect ratio.

Recent research revealed the orientation-dependent propulsion of a cone-shaped colloidal particle that is exposed to a planar traveling ultrasound wave. Here, we extend the previous research by considering nano- and microcones with different aspect ratios and studying how the propulsion of a particle depends on its orientation and aspect ratio. We also study how the orientation-averaged propulsion of a cone-shaped particle, which corresponds to an isotropic ultrasound field, depends on its aspect ratio and identify an aspect ratio of 1/2 where the orientation-averaged propulsion is particularly strong. To make our simulation results easier reusable for follow-up research, we provide a corresponding simple analytic representation.

Impact of home-based training and nutritional behavior on body composition and metabolic markers in cancer patients: data from the CRBP-TS study.

Introduction: Obesity and physical inactivity are known to affect cancer's development and prognosis. In this context, physical aerobic and resistance training as well as a Mediterranean nutrition have been proven to have many positive health effects. The aim of this study was therefore to investigate the effect of home-based training on body composition and certain metabolic laboratory parameters. Methods: Patients with breast, colorectal and prostate cancer who underwent curative surgery at stages T1N0M0-T3N3M0 were eligible for this trial and randomized to an intervention and control group. In the intervention group the patients carried out online-based strength-endurance home training during the 6-month study period. Body composition was assessed via bioelectrical impedance analysis (baseline, 3 months and 6 months). Metabolic blood parameters were also analyzed and nutrition behavior determined using the Mediterranean Diet Adherence Screener (MEDAS). Results: The intervention group's fat mass decreased while their lean body mass increased (time effect p = 0.001 and p = 0.001, respectively). We found no interaction effect in body weight (p = 0.19), fat mass [p = 0.06, 6-months estimates -0.9 (95% CI -1.8 to -0.1)] and lean body mass (p = 0.92). Blood samples also failed to show a statistically significant interaction effect between time × group for HbA1c% (p = 0.64), Insulin (p = 0.33), Adiponectin (p = 0.87), Leptin (p = 0.52) and Triglycerides (p = 0.43). Only Adiponectin revealed significance in the time effect (p < 0.001) and Leptin in the group effect (p = 0.03). Dietary behavior during the study period was similar in patients in the intervention and control groups (interaction p = 0.81; group p = 0.09 and time p = 0.03). Discussion: Individualized online-based home training in postoperative cancer patients revealed only minor changes, with no group differences in body composition or metabolic laboratory parameters, which were predominantly in the reference range at baseline. More studies investigating effects of online-based home training on body composition and nutrition behavior are needed. Trial registration: https://drks.de/search/en/trial/DRKS00020499, DRKS-ID: DRKS00020499.

Light-Induced Agglomeration of Single-Atom Platinum in Photocatalysis.

With recent advances in the field of single-atoms (SAs) used in photocatalysis, an unprecedented performance of atomically dispersed co-catalysts has been achieved. However, the stability and agglomeration of SA co-catalysts on the semiconductor surface may represent a critical issue in potential applications. Here, the photoinduced destabilization of Pt SAs on the benchmark photocatalyst, TiO2 , is described. In aqueous solutions within illumination timescales ranging from few minutes to several hours, light-induced agglomeration of Pt SAs to ensembles (dimers, multimers) and finally nanoparticles takes place. The kinetics critically depends on the presence of sacrificial hole scavengers and the used light intensity. Density-functional theory calculations attribute the light induced destabilization of the SA Pt species to binding of surface-coordinated Pt with solution-hydrogen (adsorbed H atoms), which consequently weakens the Pt SA bonding to the TiO2 surface. Despite the gradual aggregation of Pt SAs into surface clusters and their overall reduction to metallic state, which involves >90% of Pt SAs, the overall photocatalytic H2 evolution remains virtually unaffected.

Orientation-Dependent Propulsion of Triangular Nano- and Microparticles by a Traveling Ultrasound Wave.

Previous studies on ultrasound-propelled nano- and microparticles have considered only systems in which the particle orientation is perpendicular to the direction of propagation of the ultrasound. However, in future applications of these particles, they will typically be able to attain other orientations. Therefore, using direct acoustofluidic simulations, here we study how the propulsion of triangular nano- and microparticles, which are known to have a particularly efficient acoustic propulsion and are therefore promising candidates for future applications, depends on their orientation relative to the propagation direction of a traveling ultrasound wave. Our results reveal that the propulsion of the particles depends strongly on their orientation relative to the direction of wave propagation and that the particles tend to orient perpendicularly to the wave direction. We also address the orientation-averaged translational and angular velocities of the particles, which correspond to the particles' effective propulsion for an isotropic exposure to ultrasound. Our results allow assessment of how free ultrasound-propelled colloidal particles move in three spatial dimensions and thus constitute an important step toward the realization of envisaged future applications of such particles.