-
The influence of silicon on the formation and transformation of corrosion products
Authors:
Fabio E. Furcas,
Shishir Mundra,
Barbara Lothenbach,
Camelia N. Borca,
Thomas Huthwelker,
Ueli M. Angst
Abstract:
Accurate model predictions of corrosion-driven damage in reinforced concrete structures necessitate a comprehensive understanding of the rate of corrosion product formation. Here, we investigate the influence of dissolved Si characteristic of cementitious systems on the rate of corrosion product transformation at alkaline pH. Compared to systems aged in the absence of Si, small amounts of Si retar…
▽ More
Accurate model predictions of corrosion-driven damage in reinforced concrete structures necessitate a comprehensive understanding of the rate of corrosion product formation. Here, we investigate the influence of dissolved Si characteristic of cementitious systems on the rate of corrosion product transformation at alkaline pH. Compared to systems aged in the absence of Si, small amounts of Si retard the formation rate of the thermodynamically stable corrosion product goethite by a factor of 10. The estimated first order rate constant of transformation k decreases exponentially as a function of the dissolved Si concentration and follows the progression log10k = log10k_0 - 14.65[Si]^0.28. Findings further suggest that the observed retardation is primarily due to the formation of a mobile aqueous Fe-Si complex. The concentration of Si in cementitious systems has a crucial influence, and additional research is required to fully incorporate this factor into reactive transport models, ultimately essential for accurate service life predictions.
△ Less
Submitted 20 February, 2024;
originally announced February 2024.
-
Speciation controls the kinetics of iron hydroxide precipitation and transformation
Authors:
Fabio Enrico Furcas,
Shishir Mundra,
Barbara Lothenbach,
Ueli Angst
Abstract:
The formation of energetically favourable and metastable mineral phases within the Fe-H2O system controls the long-term mobility of iron complexes as well as other aqueous phase constituents in natural aquifers, soils and other environmentally and industrially relevant systems. The fundamental mechanism controlling the formation of these solid phases has remained enigmatic. Here, we develop a gene…
▽ More
The formation of energetically favourable and metastable mineral phases within the Fe-H2O system controls the long-term mobility of iron complexes as well as other aqueous phase constituents in natural aquifers, soils and other environmentally and industrially relevant systems. The fundamental mechanism controlling the formation of these solid phases has remained enigmatic. Here, we develop a general state-of-the-art partial equilibrium model and succeed in predicting the rate of amorphous 2-line ferrihydrite precipitation, dissolution and overall transformation to crystalline goethite at alkaline pH. All mechanistic steps constituting the transformation mechanism accurately predict the experimentally measured solid and aqueous phase composition over time, involving only a single kinetic rate constant each. It is found that the precipitation of goethite (i) occurs from solution and (ii) is limited by the comparatively slow dissolution of the first forming amorphous phase 2-line ferrihydrite. A generalised transformation mechanism applicable to near-neutral and mildly acidic pH further illustrates that differences in the kinetics of Fe(III) precipitation are controlled by the coordination environment of the predominant Fe(III) hydrolysis product. Findings provide a framework for the modelling of other iron(bearing) phases across a broad range of aqueous phase compositions.
△ Less
Submitted 21 November, 2023;
originally announced November 2023.
-
Finding AI-Generated Faces in the Wild
Authors:
Gonzalo J. Aniano Porcile,
Jack Gindi,
Shivansh Mundra,
James R. Verbus,
Hany Farid
Abstract:
AI-based image generation has continued to rapidly improve, producing increasingly more realistic images with fewer obvious visual flaws. AI-generated images are being used to create fake online profiles which in turn are being used for spam, fraud, and disinformation campaigns. As the general problem of detecting any type of manipulated or synthesized content is receiving increasing attention, he…
▽ More
AI-based image generation has continued to rapidly improve, producing increasingly more realistic images with fewer obvious visual flaws. AI-generated images are being used to create fake online profiles which in turn are being used for spam, fraud, and disinformation campaigns. As the general problem of detecting any type of manipulated or synthesized content is receiving increasing attention, here we focus on a more narrow task of distinguishing a real face from an AI-generated face. This is particularly applicable when tackling inauthentic online accounts with a fake user profile photo. We show that by focusing on only faces, a more resilient and general-purpose artifact can be detected that allows for the detection of AI-generated faces from a variety of GAN- and diffusion-based synthesis engines, and across image resolutions (as low as 128 x 128 pixels) and qualities.
△ Less
Submitted 5 April, 2024; v1 submitted 14 November, 2023;
originally announced November 2023.
-
Mechanism of cathodic protection of iron and steel in porous media
Authors:
Federico Martinelli-Orlando,
Shishir Mundra,
Ueli M. Angst
Abstract:
Cathodic protection (CP) was introduced two centuries ago and since has found widespread application in protecting structures such as pipelines, offshore installations, and bridges from corrosion. Despite its extensive use, the fundamental working mechanism of CP remains debated, particularly for metals in porous media such as soil. Here, we offer resolution to the long-standing debate by employin…
▽ More
Cathodic protection (CP) was introduced two centuries ago and since has found widespread application in protecting structures such as pipelines, offshore installations, and bridges from corrosion. Despite its extensive use, the fundamental working mechanism of CP remains debated, particularly for metals in porous media such as soil. Here, we offer resolution to the long-standing debate by employing in-situ and ex-situ characterisation techniques coupled with electrochemical measurements to characterise the spatio-temporal changes occurring at the steel-electrolyte interface. We show that upon CP, the interfacial electrolyte undergoes alkalinisation and deoxygenation, and that depending on polarisation conditions, an iron oxide film can simultaneously form on the steel surface. We further demonstrate that these changes in interfacial electrolyte chemistry and steel surface state result in altered anodic and cathodic reactions and their kinetics. We propose a mechanism of CP that integrates the long debated theories, based on both concentration and activation polarisation, complimentarily. Implications of this coherent scientific understanding for enhancing corrosion protection technologies and the safe, economic, and environmental-friendly operation of critical steel-based infrastructures are discussed.
△ Less
Submitted 30 August, 2023;
originally announced August 2023.
-
Transformation of 2-line ferrihydrite to goethite at alkaline pH
Authors:
Fabio E. Furcas,
Barbara Lothenbach,
Shishir Mundra,
Camelia Borca,
Cristhiana C. Albert,
O. Burkan Isgor,
Thomas Huthwelker,
Ueli M. Angst
Abstract:
The transformation of 2-line ferrihydrite to goethite from supersaturated solutions at alkaline pH >= 13.0 was studied using a combination of benchtop and advanced synchrotron techniques such as X-ray diffraction, thermogravimetric analysis and X-ray absorption spectroscopy. In comparison to the transformation rates at acidic to mildly alkaline environments, the half-life,t_1/2, of 2-line ferrihyd…
▽ More
The transformation of 2-line ferrihydrite to goethite from supersaturated solutions at alkaline pH >= 13.0 was studied using a combination of benchtop and advanced synchrotron techniques such as X-ray diffraction, thermogravimetric analysis and X-ray absorption spectroscopy. In comparison to the transformation rates at acidic to mildly alkaline environments, the half-life,t_1/2, of 2-line ferrihydrite reduces from several months at pH = 2.0, and approximately 15 days at pH = 10.0, to just under 5 hours at pH = 14.0. Calculated first order rate constants of transformation, k, increase exponentially with respect to the pH and follow the progression log_10 k = log_10 k_0 + a*pH^E3. Simultaneous monitoring of the aqueous Fe(III) concentration via inductively coupled plasma optical emission spectroscopy demonstrates that (i) goethite likely precipitates from solution and (ii) its formation is rate-limited by the comparatively slow re-dissolution of 2-line ferrihydrite. The analysis presented can be used to estimate the transformation rate of naturally occurring 2-line ferrihydrite in aqueous electrolytes characteristic to mine and radioactive waste tailings as well as the formation of corrosion products in cementitious pore solutions.
△ Less
Submitted 6 July, 2023;
originally announced July 2023.
-
Visually Aware Skip-Gram for Image Based Recommendations
Authors:
Parth Tiwari,
Yash Jain,
Shivansh Mundra,
Jenny Harding,
Manoj Kumar Tiwari
Abstract:
The visual appearance of a product significantly influences purchase decisions on e-commerce websites. We propose a novel framework VASG (Visually Aware Skip-Gram) for learning user and product representations in a common latent space using product image features. Our model is an amalgamation of the Skip-Gram architecture and a deep neural network based Decoder. Here the Skip-Gram attempts to capt…
▽ More
The visual appearance of a product significantly influences purchase decisions on e-commerce websites. We propose a novel framework VASG (Visually Aware Skip-Gram) for learning user and product representations in a common latent space using product image features. Our model is an amalgamation of the Skip-Gram architecture and a deep neural network based Decoder. Here the Skip-Gram attempts to capture user preference by optimizing user-product co-occurrence in a Heterogeneous Information Network while the Decoder simultaneously learns a map** to transform product image features to the Skip-Gram embedding space. This architecture is jointly optimized in an end-to-end, multitask fashion. The proposed framework enables us to make personalized recommendations for cold-start products which have no purchase history. Experiments conducted on large real-world datasets show that the learned embeddings can generate effective recommendations using nearest neighbour searches.
△ Less
Submitted 16 August, 2020;
originally announced August 2020.
-
Neural DrugNet
Authors:
Nishant Nikhil,
Shivansh Mundra
Abstract:
In this paper, we describe the system submitted for the shared task on Social Media Mining for Health Applications by the team Light. Previous works demonstrate that LSTMs have achieved remarkable performance in natural language processing tasks. We deploy an ensemble of two LSTM models. The first one is a pretrained language model appended with a classifier and takes words as input, while the sec…
▽ More
In this paper, we describe the system submitted for the shared task on Social Media Mining for Health Applications by the team Light. Previous works demonstrate that LSTMs have achieved remarkable performance in natural language processing tasks. We deploy an ensemble of two LSTM models. The first one is a pretrained language model appended with a classifier and takes words as input, while the second one is a LSTM model with an attention unit over it which takes character tri-gram as input. We call the ensemble of these two models: Neural-DrugNet. Our system ranks 2nd in the second shared task: Automatic classification of posts describing medication intake.
△ Less
Submitted 31 August, 2018;
originally announced September 2018.
