Showing 51–90 of 90 results for author: Bhardwaj, R

More Identifiable yet Equally Performant Transformers for Text Classification

Authors: Rishabh Bhardwaj, Navonil Majumder, Soujanya Poria, Eduard Hovy

Abstract: Interpretability is an important aspect of the trustworthiness of a model's predictions. Transformer's predictions are widely explained by the attention weights, i.e., a probability distribution generated at its self-attention unit (head). Current empirical studies provide shreds of evidence that attention weights are not explanations by proving that they are not unique. A recent study showed theo… ▽ More Interpretability is an important aspect of the trustworthiness of a model's predictions. Transformer's predictions are widely explained by the attention weights, i.e., a probability distribution generated at its self-attention unit (head). Current empirical studies provide shreds of evidence that attention weights are not explanations by proving that they are not unique. A recent study showed theoretical justifications to this observation by proving the non-identifiability of attention weights. For a given input to a head and its output, if the attention weights generated in it are unique, we call the weights identifiable. In this work, we provide deeper theoretical analysis and empirical observations on the identifiability of attention weights. Ignored in the previous works, we find the attention weights are more identifiable than we currently perceive by uncovering the hidden role of the key vector. However, the weights are still prone to be non-unique attentions that make them unfit for interpretation. To tackle this issue, we provide a variant of the encoder layer that decouples the relationship between key and value vector and provides identifiable weights up to the desired length of the input. We prove the applicability of such variations by providing empirical justifications on varied text classification tasks. The implementations are available at https://github.com/declare-lab/identifiable-transformers. △ Less

Submitted 2 June, 2021; originally announced June 2021.

Comments: ACL 2021

Journal ref: ACL 2021

Large Magnetic Moment in Flexoelectronic Silicon at Room Temperature

Authors: Paul C. Lou, Anand Katailiha, Ravindra G. Bhardwaj, Ward P. Beyermann, Dominik M. Juraschek, Sandeep Kumar

Abstract: Time-dependent rotational electric polarizations have been proposed to generate temporally varying magnetic moments, for example, through a combination of ferroelectric polarization and optical phonons. This phenomenon has been called dynamical multiferroicity, but explicit experimental demonstrations have been elusive to date. Here, we report the detection of a temporal magnetic moment as high as… ▽ More Time-dependent rotational electric polarizations have been proposed to generate temporally varying magnetic moments, for example, through a combination of ferroelectric polarization and optical phonons. This phenomenon has been called dynamical multiferroicity, but explicit experimental demonstrations have been elusive to date. Here, we report the detection of a temporal magnetic moment as high as 1.2 mu_B/atom in charge-doped thin film of silicon under flexural strain. We demonstrate that the magnetic moment is generated by a combination of electric polarization arising from a flexoelectronic charge separation along the strain gradient and the deformation potential of phonons. The effect can be controlled by adjusting the external strain gradient, doping concentration and dopant, and can be regarded as a dynamical multiferroic effect involving flexoelectronics polarization instead of ferroelectricity. The discovery of a large magnetic moment in silicon may enable the use of non-magnetic and non-ferroelectric semiconductors in various multiferroic and spintronic applications. △ Less

Submitted 23 March, 2021; originally announced March 2021.

Journal ref: Nano Letters, 2021

Numerical Simulation of Vortex-Induced Vibration With Bistable Springs : Consistency with the Equilibrium Constraint

Authors: Rameez Badhurshah, Rajneesh Bhardwaj, Amitabh Bhattacharya

Abstract: We present results from 2-D numerical simulations based on Immersed Boundary Method of a cylinder in uniform fluid flow attached to bistable springs undergoing Vortex-Induced Vibrations (VIV). The elastic spring potential for the bistable springs, consisting of 2 potential wells, is defined by the spacing between the potential minima and the depth of the potential wells. We perform simulations of… ▽ More We present results from 2-D numerical simulations based on Immersed Boundary Method of a cylinder in uniform fluid flow attached to bistable springs undergoing Vortex-Induced Vibrations (VIV). The elastic spring potential for the bistable springs, consisting of 2 potential wells, is defined by the spacing between the potential minima and the depth of the potential wells. We perform simulations of VIV with linear spring, as well as bistable springs with two different inter-well separations, over a wide range of reduced velocity. As expected, large oscillation amplitudes correspond to lock-in of the lift force with the natural frequency of the spring-mass system. The range of reduced velocity over which lock-in occurs is significantly higher for VIV with bistable springs compared to VIV with linear springs, although the maximum possible amplitude appears to be independent of the spring type. For VIV with bistable springs, the cylinder undergoes double-well oscillations in the lock-in regime. Range of reduced velocity over which lock-in occurs increases when the inter-well distance is reduced. The vortex shedding patterns and amplitude trends look similar at the same equivalent reduced velocity for the different springs. The results here are consistent with our prior theory, in which we propose a new "Equilibrium-Constraint (EC)" based on average kinetic energy budget of the structure. For a given spring potential, the intersection of natural frequency curves with the EC curve yields the possible range of reduced velocities over which lock-in should occur. Our numerical simulations show a collapse of the amplitude-versus-structure frequency data for all the simulations onto roughly the same curve, thus supporting the existence of the EC, and providing an explanation for the trends in the VIV oscillations. △ Less

Submitted 22 March, 2021; originally announced March 2021.

Comments: 37 pages. The present study provides fundamental insights into VIV characteristics of bistable springs, which may be useful for designing broadband energy harvesters

On an Inflated Unit-Lindley Distribution

Authors: Sudeep R. Bapat, Rohit Bhardwaj

Abstract: Modeling fractional data in various real life scenarios is a challenging task. This paper consider situations where fractional data is observed on the interval [0,1]. The unit-Lindley distribution has been discussed in the literature where its support lies between 0 and 1. In this paper, we focus on an inflated variant of the unit-Lindley distribution, where the inflation occurs at both 0 and 1. V… ▽ More Modeling fractional data in various real life scenarios is a challenging task. This paper consider situations where fractional data is observed on the interval [0,1]. The unit-Lindley distribution has been discussed in the literature where its support lies between 0 and 1. In this paper, we focus on an inflated variant of the unit-Lindley distribution, where the inflation occurs at both 0 and 1. Various properties of the inflated unit-Lindley distribution are discussed and examined, including point estimation based on the maximum likelihood method and interval estimation. Finally, extensive Monte Carlo simulation and real-data analyses are carried out to compare the fit of our proposed distribution along with some of the existing ones such as the inflated beta and the inflated Kumaraswamy distributions. △ Less

Submitted 9 February, 2021; originally announced February 2021.

Recognizing Emotion Cause in Conversations

Authors: Soujanya Poria, Navonil Majumder, Devamanyu Hazarika, Deepanway Ghosal, Rishabh Bhardwaj, Samson Yu Bai Jian, Pengfei Hong, Romila Ghosh, Abhinaba Roy, Niyati Chhaya, Alexander Gelbukh, Rada Mihalcea

Abstract: We address the problem of recognizing emotion cause in conversations, define two novel sub-tasks of this problem, and provide a corresponding dialogue-level dataset, along with strong Transformer-based baselines. The dataset is available at https://github.com/declare-lab/RECCON. Introduction: Recognizing the cause behind emotions in text is a fundamental yet under-explored area of research in NL… ▽ More We address the problem of recognizing emotion cause in conversations, define two novel sub-tasks of this problem, and provide a corresponding dialogue-level dataset, along with strong Transformer-based baselines. The dataset is available at https://github.com/declare-lab/RECCON. Introduction: Recognizing the cause behind emotions in text is a fundamental yet under-explored area of research in NLP. Advances in this area hold the potential to improve interpretability and performance in affect-based models. Identifying emotion causes at the utterance level in conversations is particularly challenging due to the intermingling dynamics among the interlocutors. Method: We introduce the task of Recognizing Emotion Cause in CONversations with an accompanying dataset named RECCON, containing over 1,000 dialogues and 10,000 utterance cause-effect pairs. Furthermore, we define different cause types based on the source of the causes, and establish strong Transformer-based baselines to address two different sub-tasks on this dataset: causal span extraction and causal emotion entailment. Result: Our Transformer-based baselines, which leverage contextual pre-trained embeddings, such as RoBERTa, outperform the state-of-the-art emotion cause extraction approaches Conclusion: We introduce a new task highly relevant for (explainable) emotion-aware artificial intelligence: recognizing emotion cause in conversations, provide a new highly challenging publicly available dialogue-level dataset for this task, and give strong baseline results on this dataset. △ Less

Submitted 28 July, 2021; v1 submitted 21 December, 2020; originally announced December 2020.

Comments: https://github.com/declare-lab/RECCON, Accepted at Cognitive Computation

Ekya: Continuous Learning of Video Analytics Models on Edge Compute Servers

Authors: Romil Bhardwaj, Zhengxu Xia, Ganesh Ananthanarayanan, Junchen Jiang, Nikolaos Karianakis, Yuanchao Shu, Kevin Hsieh, Victor Bahl, Ion Stoica

Abstract: Video analytics applications use edge compute servers for the analytics of the videos (for bandwidth and privacy). Compressed models that are deployed on the edge servers for inference suffer from data drift, where the live video data diverges from the training data. Continuous learning handles data drift by periodically retraining the models on new data. Our work addresses the challenge of jointl… ▽ More Video analytics applications use edge compute servers for the analytics of the videos (for bandwidth and privacy). Compressed models that are deployed on the edge servers for inference suffer from data drift, where the live video data diverges from the training data. Continuous learning handles data drift by periodically retraining the models on new data. Our work addresses the challenge of jointly supporting inference and retraining tasks on edge servers, which requires navigating the fundamental tradeoff between the retrained model's accuracy and the inference accuracy. Our solution Ekya balances this tradeoff across multiple models and uses a micro-profiler to identify the models that will benefit the most by retraining. Ekya's accuracy gain compared to a baseline scheduler is 29% higher, and the baseline requires 4x more GPU resources to achieve the same accuracy as Ekya. △ Less

Submitted 18 December, 2020; originally announced December 2020.

Signatures of magnetic field effects in non-sequential double ionization manifesting as back-scattering for molecules versus forward-scattering for atoms

Authors: Georgios Petros Katsoulis, Matthew Benjamin Peters, André Staudte, Ravi Bhardwaj, Agapi Emmanouilidou

Abstract: For two-electron diatomic molecules, we investigate magnetic field effects in non-sequential double ionization where recollisions prevail. We do so by formulating a three-dimensional semi-classical model that fully accounts for the Coulomb singularities and for magnetic field effects during time propagation. Using this model, we identify a prominent signature of non-dipole effects. Namely, we demo… ▽ More For two-electron diatomic molecules, we investigate magnetic field effects in non-sequential double ionization where recollisions prevail. We do so by formulating a three-dimensional semi-classical model that fully accounts for the Coulomb singularities and for magnetic field effects during time propagation. Using this model, we identify a prominent signature of non-dipole effects. Namely, we demonstrate that the recolliding electron back-scatters along the direction of light propagation. Hence, this electron escapes opposite to the direction of change in momentum due to the magnetic field. This is in striking contrast to strongly-driven atoms where the recolliding electron forward-scatters along the direction of light propagation. We attribute these distinct signatures to the different gate that the magnetic field creates jointly with a soft recollision in molecules compared to a hard recollision in atoms. These two different gates give rise, shortly before recollision, to different momenta and positions of the recolliding electron along the direction of light propagation. As a result, we show that the Coulomb forces from the nuclei act to back-scatter the recolliding electron in molecules and forward-scatter it in atoms along the direction of light propagation. △ Less

Submitted 11 December, 2020; originally announced December 2020.

Comments: 14 pages, 8 figures

Journal ref: Phys. Rev. A 103, 033115 (2021)

Investigating Gender Bias in BERT

Authors: Rishabh Bhardwaj, Navonil Majumder, Soujanya Poria

Abstract: Contextual language models (CLMs) have pushed the NLP benchmarks to a new height. It has become a new norm to utilize CLM provided word embeddings in downstream tasks such as text classification. However, unless addressed, CLMs are prone to learn intrinsic gender-bias in the dataset. As a result, predictions of downstream NLP models can vary noticeably by varying gender words, such as replacing "h… ▽ More Contextual language models (CLMs) have pushed the NLP benchmarks to a new height. It has become a new norm to utilize CLM provided word embeddings in downstream tasks such as text classification. However, unless addressed, CLMs are prone to learn intrinsic gender-bias in the dataset. As a result, predictions of downstream NLP models can vary noticeably by varying gender words, such as replacing "he" to "she", or even gender-neutral words. In this paper, we focus our analysis on a popular CLM, i.e., BERT. We analyse the gender-bias it induces in five downstream tasks related to emotion and sentiment intensity prediction. For each task, we train a simple regressor utilizing BERT's word embeddings. We then evaluate the gender-bias in regressors using an equity evaluation corpus. Ideally and from the specific design, the models should discard gender informative features from the input. However, the results show a significant dependence of the system's predictions on gender-particular words and phrases. We claim that such biases can be reduced by removing genderspecific features from word embedding. Hence, for each layer in BERT, we identify directions that primarily encode gender information. The space formed by such directions is referred to as the gender subspace in the semantic space of word embeddings. We propose an algorithm that finds fine-grained gender directions, i.e., one primary direction for each BERT layer. This obviates the need of realizing gender subspace in multiple dimensions and prevents other crucial information from being omitted. Experiments show that removing embedding components in such directions achieves great success in reducing BERT-induced bias in the downstream tasks. △ Less

Submitted 10 September, 2020; originally announced September 2020.

A finite-element model for computing fluid flow inside a sessile evaporating droplet on a solid surface

Authors: Manish Kumar, Rajneesh Bhardwaj

Abstract: A finite element model was developed to compute the fluid flow inside a sessile evaporating droplet on hydrophilic substrate in ambient conditions. The evaporation is assumed as quasi-steady and the flow is considered as axisymmetric with a pinned contact line. The Navier-Stokes equations in cylindrical coordinates were solved inside the droplet. Galerkin weight residual approach and velocity pres… ▽ More A finite element model was developed to compute the fluid flow inside a sessile evaporating droplet on hydrophilic substrate in ambient conditions. The evaporation is assumed as quasi-steady and the flow is considered as axisymmetric with a pinned contact line. The Navier-Stokes equations in cylindrical coordinates were solved inside the droplet. Galerkin weight residual approach and velocity pressure formulation was used to discretise the governing equations. Six node triangular mesh and quadratic shape functions were used to obtain higher accuracy solutions. Radial velocity profiles in axial directions calculated by the FEM solver were compared with a existing analytical model and were found in excellent agreement. The contours of velocity magnitude and streamlines show the characteristic flow i.e. radially outward inside the evaporating droplet. △ Less

Submitted 7 September, 2020; originally announced September 2020.

Terahertz Nonlinear Optical Response of Water Vapor

Authors: Payman Rasekh, Akbar Safari, Murat Yildirim, Ravi Bhardwaj, Jean-Michel Ménard, Ksenia Dolgaleva, Robert W. Boyd

Abstract: We report on the nonlinear spectroscopy of water vapor at THz frequencies. Atmospheric water vapor has a rich spectrum with several strong resonances at frequencies below 3 THz, falling within the range of operation of most existing THz sources. We observe an extremely large nonlinear response to THz radiation at the positions of these resonances. Using the optical Kerr model for the nonlinear res… ▽ More We report on the nonlinear spectroscopy of water vapor at THz frequencies. Atmospheric water vapor has a rich spectrum with several strong resonances at frequencies below 3 THz, falling within the range of operation of most existing THz sources. We observe an extremely large nonlinear response to THz radiation at the positions of these resonances. Using the optical Kerr model for the nonlinear response, we estimate a minimum nonlinear refractive index of the order of 10^2 m^2/W. Our results provide insight into the energy levels of the water molecule and give a more accurate picture of its response to electromagnetic radiation, paving the way to more accurate THz spectroscopy, imaging and sensing systems, and thereby facilitating future emerging THz technologies. △ Less

Submitted 8 June, 2020; originally announced June 2020.

Likelihood of survival of coronavirus in a respiratory droplet deposited on a solid surface

Authors: Rajneesh Bhardwaj, Amit Agrawal

Abstract: We predict and analyze the drying time of respiratory droplets from a COVID-19 infected subject, which is a crucial time to infect another subject. The drying of the droplet is predicted by diffusion-limited evaporation model for a sessile droplet placed on a partially-wetted surface with a pinned contact line. The variation of droplet volume, contact angle, ambient temperature, and humidity are c… ▽ More We predict and analyze the drying time of respiratory droplets from a COVID-19 infected subject, which is a crucial time to infect another subject. The drying of the droplet is predicted by diffusion-limited evaporation model for a sessile droplet placed on a partially-wetted surface with a pinned contact line. The variation of droplet volume, contact angle, ambient temperature, and humidity are considered. We analyze the chances of the survival of the viruses present in the droplet, based on the lifetime of the droplets in several conditions, and find that the chances of survival of the virus are strongly affected by each of these parameters. The magnitude of shear stress inside the droplet computed using the model is not large enough to obliterate the virus. We also explore the relationship between the drying time of a droplet and the growth rate of the spread of COVID-19 for five different cities, and find that they are weakly correlated. △ Less

Submitted 9 May, 2020; originally announced May 2020.

Improving Aspect-Level Sentiment Analysis with Aspect Extraction

Authors: Navonil Majumder, Rishabh Bhardwaj, Soujanya Poria, Amir Zadeh, Alexander Gelbukh, Amir Hussain, Louis-Philippe Morency

Abstract: Aspect-based sentiment analysis (ABSA), a popular research area in NLP has two distinct parts -- aspect extraction (AE) and labeling the aspects with sentiment polarity (ALSA). Although distinct, these two tasks are highly correlated. The work primarily hypothesize that transferring knowledge from a pre-trained AE model can benefit the performance of ALSA models. Based on this hypothesis, word emb… ▽ More Aspect-based sentiment analysis (ABSA), a popular research area in NLP has two distinct parts -- aspect extraction (AE) and labeling the aspects with sentiment polarity (ALSA). Although distinct, these two tasks are highly correlated. The work primarily hypothesize that transferring knowledge from a pre-trained AE model can benefit the performance of ALSA models. Based on this hypothesis, word embeddings are obtained during AE and subsequently, feed that to the ALSA model. Empirically, this work show that the added information significantly improves the performance of three different baseline ALSA models on two distinct domains. This improvement also translates well across domains between AE and ALSA tasks. △ Less

Submitted 3 May, 2020; originally announced May 2020.

Computational Modeling and Analysis of Flow-induced Vibration of an Elastic Splitter Plate Using a Sharp-interface Immersed Boundary Method

Authors: Anup Kundu, Atul K. Soti, Hemanshul Garg, Rajneesh Bhardwaj, Mark C. Thompson

Abstract: We present the development and benchmarking of an in-house fluid-structure interaction (FSI) solver. An implicit partitioned approach is utilized to couple a sharp-interface immersed boundary (IB) method based flow solver and a finite-element method based structural solver. In the present work, the coupling is accelerated using a dynamic under-relaxation scheme. The revised coupling is around two… ▽ More We present the development and benchmarking of an in-house fluid-structure interaction (FSI) solver. An implicit partitioned approach is utilized to couple a sharp-interface immersed boundary (IB) method based flow solver and a finite-element method based structural solver. In the present work, the coupling is accelerated using a dynamic under-relaxation scheme. The revised coupling is around two to three times faster and numerically stable, as compared to the one that uses a constant under-relaxation parameter. The solver is validated against two FSI benchmarks in which a thin, finite thickness, elastic splitter plate is attached to the lee side of a circular or square rigid cylinder, subjected to laminar flow. In these two-dimensional benchmarks, the flow induces a wave-like deformation in the plate, and it attains a periodic self-sustained oscillation. We employ the FSI solver to analyze the flow-induced vibration (FIV) of the plate in a uniform laminar free-stream flow for a wide range of mass ratio and bending stiffness at Reynolds number ($Re$) of 100, based on the diameter of the cylinder. At the given $Re$, two-dimensional numerical simulations show that the FIV of the plate effectively depends only on the mass ratio and bending stiffness. The largest displacement of the plate vibration is found to occur in the lock-in region, where the vortex shedding frequency of the coupled fluid-structure system is close to the natural frequency of the splitter plate. We briefly discuss wake structures and phase plots for different cases of mass ratio and bending stiffness. △ Less

Submitted 6 May, 2020; originally announced May 2020.

Evaporation of initially heated sessile droplets and the resultant dried colloidal deposits on substrates held at ambient temperature

Authors: Sanghamitro Chatterjee, Manish Kumar, Janani S Murallidharan, Rajneesh Bhardwaj

Abstract: The present study experimentally and numerically investigates the evaporation and resultant patterns of dried deposits of aqueous colloidal sessile droplets, when the droplets are initially elevated to a high temperature before being placed on a substrate held at ambient temperature. The system is then released for natural evaporation without applying any external perturbation. Infrared thermograp… ▽ More The present study experimentally and numerically investigates the evaporation and resultant patterns of dried deposits of aqueous colloidal sessile droplets, when the droplets are initially elevated to a high temperature before being placed on a substrate held at ambient temperature. The system is then released for natural evaporation without applying any external perturbation. Infrared thermography and optical profilometry were used as essential tools for interfacial temperature measurements and quantification of the coffee-ring dimensions, respectively. Initially, a significant temperature gradient exists along the liquid-gas interface as soon as the droplet is deposited on the substrate which triggers a Marangoni stress-induced recirculation flow directed from the top of the droplet towards the contact line along the liquid-gas interface. Thus, the flow is in the reverse direction to that seen in the conventional substrate heating case. Interestingly, this temperature gradient decays rapidly -- within the first 10% of the total evaporation time and the droplet-substrate system reaches thermal equilibrium with ambient thereafter. Despite fast decay of the temperature gradient, the coffee-ring dimensions significantly diminish, leading to an inner deposit. This suppression of the coffee-ring effect is attributed to the fact that the initial Marangoni stress-induced recirculation flow continues until the last stage of the evaporation, even after the interfacial temperature gradient vanishes. This is essentially a consequence of liquid inertia. Overall, together with a new experimental condition, the present investigation discloses a distinct nature of Marangoni stress-induced flow in the drying droplet and its role in influencing the associated colloidal deposits, which was not explored previously. △ Less

Submitted 7 July, 2020; v1 submitted 13 April, 2020; originally announced April 2020.

Large spin-Hall effect in Si at room temperature

Authors: Paul C. Lou, Anand Katailiha, Ravindra G. Bhardwaj, Tonmoy Bhowmick, W. P. Beyermann, Roger K. Lake, Sandeep Kumar

Abstract: Silicon's weak intrinsic spin-orbit coupling and centrosymmetric crystal structure are a critical bottleneck to the development of Si spintronics, because they lead to an insignificant spin-Hall effect (spin current generation) and inverse spin-Hall effect (spin current detection). Here, we undertake current, magnetic field, crystallography dependent magnetoresistance and magneto thermal transport… ▽ More Silicon's weak intrinsic spin-orbit coupling and centrosymmetric crystal structure are a critical bottleneck to the development of Si spintronics, because they lead to an insignificant spin-Hall effect (spin current generation) and inverse spin-Hall effect (spin current detection). Here, we undertake current, magnetic field, crystallography dependent magnetoresistance and magneto thermal transport measurements to study the spin transport behavior in freestanding Si thin films. We observe a large spin-Hall magnetoresistance in both p-Si and n-Si at room temperature and it is an order of magnitude larger than that of Pt. One explanation of the unexpectedly large and efficient spin-Hall effect is spin-phonon coupling instead of spin-orbit coupling. The macroscopic origin of the spin-phonon coupling can be large strain gradients that can exist in the freestanding Si films. This discovery in a light, earth abundant and centrosymmetric material opens a new path of strain engineering to achieve spin dependent properties in technologically highly-developed materials. △ Less

Submitted 1 April, 2020; originally announced April 2020.

Journal ref: Phys. Rev. B 101, 094435, 2020

HyperSched: Dynamic Resource Reallocation for Model Development on a Deadline

Authors: Richard Liaw, Romil Bhardwaj, Lisa Dunlap, Yitian Zou, Joseph Gonzalez, Ion Stoica, Alexey Tumanov

Abstract: Prior research in resource scheduling for machine learning training workloads has largely focused on minimizing job completion times. Commonly, these model training workloads collectively search over a large number of parameter values that control the learning process in a hyperparameter search. It is preferable to identify and maximally provision the best-performing hyperparameter configuration (… ▽ More Prior research in resource scheduling for machine learning training workloads has largely focused on minimizing job completion times. Commonly, these model training workloads collectively search over a large number of parameter values that control the learning process in a hyperparameter search. It is preferable to identify and maximally provision the best-performing hyperparameter configuration (trial) to achieve the highest accuracy result as soon as possible. To optimally trade-off evaluating multiple configurations and training the most promising ones by a fixed deadline, we design and build HyperSched -- a dynamic application-level resource scheduler to track, identify, and preferentially allocate resources to the best performing trials to maximize accuracy by the deadline. HyperSched leverages three properties of a hyperparameter search workload over-looked in prior work - trial disposability, progressively identifiable rankings among different configurations, and space-time constraints - to outperform standard hyperparameter search algorithms across a variety of benchmarks. △ Less

Submitted 7 January, 2020; originally announced January 2020.

Flexoelectric effect mediated spin-to-charge conversion at amorphous-Si thin film interfaces

Authors: Ravindra G Bhardwaj, Anand Katailiha, Paul C. Lou, W. P. Beyermann, Sandeep Kumar

Abstract: Interfacial spin to charge conversion arises due to an electric potential perpendicular to the interface. The electric potential can be artificially induced, for example, using ferroelectric and piezoelectric thin films at the interface. An alternate way to induce the electric potential could be flexoelectric field. The flexoelectricity can be observed in all the material that either have or lack… ▽ More Interfacial spin to charge conversion arises due to an electric potential perpendicular to the interface. The electric potential can be artificially induced, for example, using ferroelectric and piezoelectric thin films at the interface. An alternate way to induce the electric potential could be flexoelectric field. The flexoelectricity can be observed in all the material that either have or lack inversion symmetry, additionally no large gate bias is needed. In this experimental study, we report large spin to charge conversion (spin-Hall angle- 0.578) at Ni80Fe20/amorphous-Si interfaces attributed to flexoelectricity mediated Rashba spin-orbit coupling. The flexoelectricity at the interface also gave rise to interlayer spin-acoustic phonon or flexo-magnetoelastic coupling. In addition to spin-charge conversion, the strained interfaces also led to almost three-fold increase in anomalous Nernst effect. This strain engineering for spin dependent thermoelectric behavior at room temperature opens a new window to the realization of spintronics and spin-caloritronics devices. △ Less

Submitted 19 June, 2020; v1 submitted 6 January, 2020; originally announced January 2020.

Comments: 28 pages, 3 main figures and 9 supplementary figures

Colloidal Deposit of an Evaporating Sessile Droplet on a Non-uniformly Heated Substrate

Authors: Laxman K. Malla, Rajneesh Bhardwaj, Adrian Neild

Abstract: The pattern and profile of a dried colloidal deposit formed after evaporation of a sessile water droplet containing polystyrene particles on a non-uniformly heated glass are investigated experimentally. In particular, the effects of temperature gradient across the substrate and particles size are investigated. The temperature gradient was imposed using Peltier coolers, and side visualization, infr… ▽ More The pattern and profile of a dried colloidal deposit formed after evaporation of a sessile water droplet containing polystyrene particles on a non-uniformly heated glass are investigated experimentally. In particular, the effects of temperature gradient across the substrate and particles size are investigated. The temperature gradient was imposed using Peltier coolers, and side visualization, infrared thermography, optical microscopy, and optical profilometry were employed to collect the data. On a uniformly heated substrate, a ring with an inner deposit is obtained, which is attributed to axisymmetric Marangoni recirculation and consistent with previous reports. However, the dimensions of the ring formed on a non-uniformly heated substrate are significantly different on the hot and cold side of the substrate and are found to be a function of the temperature gradient and particles size. In the case of smaller particle size, the contact line on hot side depins and together with twin asymmetric Marangoni recirculations, it results in a larger ring width on the cold side as compared to the hot side. In contrast, the contact line remains pinned in case of larger particles, and the twin asymmetric Marangoni recirculations advect more particles on the hot side, resulting in a larger ring width at the hot side. A mechanistic model is employed to explain why the depinning is dependent on the particle size. A larger temperature gradient significantly increases or decreases the ring width depending on the particle size, due to a stronger intensity recirculation. A regime map is proposed for the deposit patterns on temperature gradient-particle size plane to classify the deposits. △ Less

Submitted 22 September, 2019; originally announced September 2019.

A Combined Computational and Experimental Investigation on Evaporation of a Sessile Water Droplet on a Heated Hydrophilic Substrate

Authors: Manish Kumar, Rajneesh Bhardwaj

Abstract: We numerically and experimentally investigate evaporation of a sessile droplet on a heated substrate. We develop a finite element (FE) model in two-dimensional axisymmetric coordinates to solve coupled transport of heat in the droplet and substrate, and of the mass of liquid vapor in surrounding ambient while assuming diffusion-limited, quasi-steady evaporation of the droplet. The two-way coupling… ▽ More We numerically and experimentally investigate evaporation of a sessile droplet on a heated substrate. We develop a finite element (FE) model in two-dimensional axisymmetric coordinates to solve coupled transport of heat in the droplet and substrate, and of the mass of liquid vapor in surrounding ambient while assuming diffusion-limited, quasi-steady evaporation of the droplet. The two-way coupling is implemented using an iterative scheme and under-relaxation is used to ensure numerical stability. The FE model is validated against the published spatial profile of the evaporation mass flux and temperature of the liquid-gas interface. We discuss cases in which the two-way coupling is significantly accurate than the one-way coupling. In experiments, we visualized side view of an evaporating microliter water droplet using a high-speed camera at different substrate temperatures and recorded temperature of the liquid-gas interface from the top using an infrared camera. We examine the dependency of inversion of the temperature profile across the liquid-gas interface on the ratio of the substrate thickness to the wetted radius, the ratio of the thermal conductivity of the substrate to that of the droplet and contact angle. A comparison of measured evaporation mass rate with the computed values at different substrate temperature show that the evaporation mass rate increases non-linearly with respect to the substrate temperature, and FE model predicts these values close to the experimental data. Comparisons of time-averaged evaporation mass rate obtained by the previous and present models against the measurements suggest that the evaporative cooling at the interface and variation of diffusion coefficient with temperature should be taken into account in the model in order to accurately capture the measurements. △ Less

Submitted 31 March, 2019; originally announced July 2019.

Journal ref: International Journal of Heat and Mass Transfer, Volume 122, pages 1223-1238, 2018

Analysis of Profile and Morphology of Colloidal Deposits obtained from Evaporating Sessile Droplets

Authors: Laxman K. Malla, Rajneesh Bhardwaj, Adrian Neild

Abstract: We experimentally investigate the profile and morphology of the ring-like deposits obtained after evaporation of a sessile water droplet containing polystyrene colloidal particles on a hydrophilic glass substrate. In particular, the coupled effect of particle size and concentration are studied. The deposits were qualitatively visualized under an optical microscope and profile of the ring was measu… ▽ More We experimentally investigate the profile and morphology of the ring-like deposits obtained after evaporation of a sessile water droplet containing polystyrene colloidal particles on a hydrophilic glass substrate. In particular, the coupled effect of particle size and concentration are studied. The deposits were qualitatively visualized under an optical microscope and profile of the ring was measured by an optical profilometer. The profile of the ring resembles a partial torus-like shape for all cases of particles size and concentration. The cracks on the surface of the ring were found to occur only at smaller particle size and larger concentration. We plot a regime map to classify three deposit types - discontinuous monolayer ring, continuous monolayer ring, and multiple layers ring - on particles concentration - particle size plane. Our data shows a possible existence of a critical concentration (particle size) for a given particle size (concentration) at which the monolayer ring forms. For the larger particle sizes, the immersion capillary forces between the particles dominate, aiding the formation of a monolayer ring of the particles. The relative mass of the particles accumulated in the ring is lesser in cases of the monolayer ring. We measure the width and height of the ring and show that they scale with particle concentration by a power law for the multiple layers ring. This scaling corroborates with an existing continuum based theoretical model. We briefly discuss the effect of the interaction of growing deposit with shrinking free surface on the ring dimensions and profile. The present results aid understanding of the ring formation process and will be useful in guiding the design of self-assemblies of the colloidal particles formed by the evaporating droplets. △ Less

Submitted 31 March, 2019; originally announced April 2019.

Journal ref: Colloids and Surfaces A: Physicochemical and Engineering Aspects, Volume 567, pages 150-160, 2019

Recent Developments on Colloidal Deposits Obtained by Evaporation of Sessile Droplets on a Solid Surface

Authors: Nagesh D. Patil, Rajneesh Bhardwaj

Abstract: Understanding flow patterns and coupled transport phenomena during evaporation of droplets loaded with colloidal particles is central to design technical applications such as organizing proteins/DNA on a solid surface. We review recent reports on evaporating sessile droplets of colloidal suspensions on a solid surface. Starting from the classical mechanism of formation of a ring-like deposit, we d… ▽ More Understanding flow patterns and coupled transport phenomena during evaporation of droplets loaded with colloidal particles is central to design technical applications such as organizing proteins/DNA on a solid surface. We review recent reports on evaporating sessile droplets of colloidal suspensions on a solid surface. Starting from the classical mechanism of formation of a ring-like deposit, we discuss the influence of several problem parameters. Notably, thermal or solutal Marangoni effect, particle size, particle concentration, particle shape, substrate wettability, pH of the suspension etc have been found important in controlling the deposition pattern. The deposit pattern complexity and shape have been attributed to the underlying coupled transport phenomena during the evaporation. We discuss important regimes maps reported for different types of deposit, which allow us to classify the deposits and coupled physics. We also present studies that have demonstrated particles sorting in an evaporating bi-dispersed colloidal suspensions on a solid surface. Finally, some remarks for the future research opportunities in this arena are presented. △ Less

Submitted 31 March, 2019; originally announced April 2019.

Journal ref: Journal of Indian Institute of Science, 2019

Analysis of an Evaporating Sessile Droplet on a Non-Wetted Surface

Authors: Rajneesh Bhardwaj

Abstract: We investigate evaporation of a sessile droplet on a non-wetted surface in the framework of diffusion-limited and quasi-steady evaporation. We extend previous models and numerically solve Laplace equation for the diffusion of liquid vapor in ambient. We propose a unified, simple and accurate expression of the evaporation mass flux valid for 90^o < theta < 180^o, where theta is the equilibrium cont… ▽ More We investigate evaporation of a sessile droplet on a non-wetted surface in the framework of diffusion-limited and quasi-steady evaporation. We extend previous models and numerically solve Laplace equation for the diffusion of liquid vapor in ambient. We propose a unified, simple and accurate expression of the evaporation mass flux valid for 90^o < theta < 180^o, where theta is the equilibrium contact angle. In addition, using the derived expression of the evaporation mass flux, we propose a simple and accurate expression of the evaporation mass rate for a non-wetted surface, which does not exhibit singularity at theta = 180^o. Finally, using the scaling analysis, the expression of the evaporation mass flux is utilized to estimate the direction and magnitude of the characteristic evaporation-driven flow velocity inside the droplet on a non-wetted surface. The predicted flow direction is found to be consistent with the previous measurements. △ Less

Submitted 7 July, 2018; originally announced July 2018.

Self-sorting of Bi-dispersed Colloidal Particles near Contact Line of an Evaporating Sessile Droplet

Authors: Nagesh D. Patil, Rajneesh Bhardwaj, Atul Sharma

Abstract: We investigate deposit patterns and associated morphology formed after the evaporation of an aqueous droplet containing mono- and bi-dispersed colloidal particles. In particular, the combined effect of substrate heating and particle diameter is investigated. We employ high-speed visualization, optical microscopy and scanning electron microscopy to characterize the evaporating droplets, particle mo… ▽ More We investigate deposit patterns and associated morphology formed after the evaporation of an aqueous droplet containing mono- and bi-dispersed colloidal particles. In particular, the combined effect of substrate heating and particle diameter is investigated. We employ high-speed visualization, optical microscopy and scanning electron microscopy to characterize the evaporating droplets, particle motion, and deposit morphology, respectively. In the context of mono-dispersed colloidal particles, an inner deposit and a typical ring form for smaller and larger particles, respectively, on a nonheated surface. At larger substrate temperature, a thin ring with inner deposit forms, explained by the self-pinning of the contact line and advection of the particles from the contact line to the center of the droplet due to Marangoni flow. In the context of bi-dispersed colloidal particles, self-sorting of the colloidal particles within the ring occurs at larger substrate temperature. The smaller particles deposit at the outermost edge as compared to the larger diameter particles and this preferential deposition in a stagnation region near the contact line is due to the spatially-varying height of the liquid-gas interface above the substrate. The sorting occurs at a smaller ratio of the diameter of the smaller and larger particle. At the larger substrate temperature and a larger ratio, the particles do not get sorted and mix into each other. Our measurements show that there exists a critical substrate temperature as well as a diameter ratio in order to achieve the sorting. We propose regime maps on substrate temperature-particle diameter and substrate temperature-diameter ratio plane for mono- and bi-dispersed solutions, respectively. △ Less

Submitted 7 July, 2018; originally announced July 2018.

Giant enhancement in Rashba spin-Seebeck effect in NiFe/p-Si thin films

Authors: Ravindra G. Bhardwaj, Paul C Lou, Sandeep Kumar

Abstract: The spin-Seebeck effect mediated thermoelectric energy conversion can provide efficient alternative to traditional thermoelectrics for waste heat recovery. To achieve this goal, efficient spin to charge conversion using earth-abundant materials is essential. Proximity induced Rashba effect arises from the charge potential mediated by structural inversion asymmetry, which has been reported in Si th… ▽ More The spin-Seebeck effect mediated thermoelectric energy conversion can provide efficient alternative to traditional thermoelectrics for waste heat recovery. To achieve this goal, efficient spin to charge conversion using earth-abundant materials is essential. Proximity induced Rashba effect arises from the charge potential mediated by structural inversion asymmetry, which has been reported in Si thin films and can be manipulated by controlling the thickness of Rashba layer. We demonstrate a giant Rashba spin-Seebeck effect in NiFe/p-Si (polycrystalline) bilayer thin films. The bilayer thin film specimens have p-Si layer thickness of 5 nm, 25 nm and 100 nm while keeping the NiFe layer thickness at 25 nm. The Rashba spin Seebeck coefficient has been estimated to be 0.266 μV/K for 100 nm p-Si, and increases by an order of magnitude to 2.11 μV/K for 5 nm p-Si. The measured spin-Seebeck coefficient in 5 nm p-Si specimen is one of the largest coefficient ever reported. The measured voltage of 100.3 μV is one of the largest reported spin-Seebeck voltage, with smallest area of ~160 X 10 μm2 used in any spin-Seebeck measurement. This scientific and technological breakthrough using earth abundant elements brings the spin mediated thermoelectric energy conversion for waste heat recovery closer to reality. △ Less

Submitted 4 April, 2018; originally announced April 2018.

Journal ref: physica status solidi (RRL) Rapid Research Letters 2018

Spin Seebeck effect and thermal spin galvanic effect in Ni80Fe20/p-Si bilayers

Authors: Ravindra G Bhardwaj, Paul C Lou, Sandeep Kumar

Abstract: The development of spintronics and spin-caloritronics devices need efficient generation, detection and manipulation of spin current. The thermal spin current from spin-Seebeck effect has been reported to be more energy efficient than the electrical spin injection methods. But, spin detection has been the one of the bottlenecks since metals with large spin-orbit coupling is an essential requirement… ▽ More The development of spintronics and spin-caloritronics devices need efficient generation, detection and manipulation of spin current. The thermal spin current from spin-Seebeck effect has been reported to be more energy efficient than the electrical spin injection methods. But, spin detection has been the one of the bottlenecks since metals with large spin-orbit coupling is an essential requirement. In this work, we report an efficient thermal generation and interfacial detection of spin current. We measured a spin-Seebeck effect in Ni80Fe20 (25 nm)/p-Si (50 nm) (polycrystalline) bilayers without heavy metal spin detector. The p-Si, having the centosymmetric crystal structure, has insignificant intrinsic spin-orbit coupling leading to negligible spin-charge conversion. We report a giant inverse spin-Hall effect, essential for detection of spin-Seebeck effect, in the Ni80Fe20/p-Si bilayer structure, which originates from Rashba spin orbit coupling due to structure inversion asymmetry at the interface. In addition, the thermal spin pumping in p-Si leads to spin current from p-Si to Ni80Fe20 layer due to thermal spin galvanic effect and spin-Hall effect causing spin-orbit torques. The thermal spin-orbit torques leads to collapse of magnetic hysteresis of 25 nm thick Ni80Fe20 layer. The thermal spin-orbit torques can be used for efficient magnetic switching for memory applications. These scientific breakthroughs may give impetus to the silicon spintronics and spin-caloritronics devices. △ Less

Submitted 31 January, 2018; originally announced February 2018.

Comments: 28 pages

Journal ref: Appl. Phys. Lett. 112, 042404 (2018)

Harnessing Electrical Power from Vortex-Induced Vibration of a Circular Cylinder

Authors: Atul Kumar Soti, Mark C. Thompson, John Sheridan, Rajneesh Bhardwaj

Abstract: The generation of electrical power from Vortex-Induced Vibration (VIV) of a cylinder is investigated numerically. The cylinder is free to oscillate in the direction transverse to the incoming flow. The cylinder is attached to a magnet that can move along the axis of a coil made from conducting wire. The magnet and the coil together constitute a basic electrical generator. When the cylinder undergo… ▽ More The generation of electrical power from Vortex-Induced Vibration (VIV) of a cylinder is investigated numerically. The cylinder is free to oscillate in the direction transverse to the incoming flow. The cylinder is attached to a magnet that can move along the axis of a coil made from conducting wire. The magnet and the coil together constitute a basic electrical generator. When the cylinder undergoes VIV, the motion of the magnet creates a voltage across the coil, which is connected to a resistive load. By Lenz's law, induced current in the coil applies a retarding force to the magnet. Effectively, the electrical generator applies a damping force on the cylinder with a spatially varying damping coefficient. For the initial investigation reported here, the Reynolds number is restricted to Re < 200, so that the flow is laminar and two-dimensional (2D). The incompressible 2D Navier-Stokes equations are solved using an extensively validated spectral-element based solver. The effects of the electromagnetic (EM) damping constant xi_m, coil dimensions (radius a, length L), and mass ratio on the electrical power extracted are quantified. It is found that there is an optimal value of xi_m (xi_opt) at which maximum electrical power is generated. As the radius or length of the coil is increased, the value of xi_opt is observed to increase. Although the maximum average power remains the same, a larger coil radius or length results in a more robust system in the sense that a relatively large amount of power can be extracted when xi_m is far from xi_opt, unlike the constant damping ratio case. The average power output is also a function of Reynolds number, primarily through the increased maximum oscillation amplitude that occurs with increased Reynolds number at least within the laminar range, although the general qualitative findings seem likely to carry across to high Reynolds number VIV. △ Less

Submitted 5 December, 2017; originally announced December 2017.

Journal ref: Journal of Fluids and Structures, Volume 70, April 2017, Pages 360-373

Role of Skin Friction Drag during Flow-Induced Reconfiguration of a Flexible Thin Plate

Authors: Awan Bhati, Rajat Sawanni, Kaushik Kulkarni, Rajneesh Bhardwaj

Abstract: We investigate drag reduction due to the flow-induced reconfiguration of a flexible thin plate in presence of skin friction drag at low Reynolds Number. The plate is subjected to a uniform free stream and is tethered at one end. We extend existing models in the literature to account for the skin friction drag. The total drag on the plate with respect to a rigid upright plate decreases due to flow-… ▽ More We investigate drag reduction due to the flow-induced reconfiguration of a flexible thin plate in presence of skin friction drag at low Reynolds Number. The plate is subjected to a uniform free stream and is tethered at one end. We extend existing models in the literature to account for the skin friction drag. The total drag on the plate with respect to a rigid upright plate decreases due to flow-induced reconfiguration and further reconfiguration increases the total drag due to increase in skin friction drag. A critical value of Cauchy number ($Ca$) exists at which the total drag on the plate with respect to a rigid upright plate is minimum at a given Reynolds number. The reconfigured shape of the plate for this condition is unique, beyond which the total drag increases on the plate even with reconfiguration. The ratio of the form drag coefficient for an upright rigid plate and skin drag coefficient for a horizontal rigid plate ($λ$) determines the critical Cauchy number ($Ca_{cr}$). We propose modification in the drag scaling with free stream velocity ($F_{x}$ ${\propto}$ $U^{n}$) in presence of the skin friction drag. The following expressions of $n$ are found for $0.01 \leq Re \leq 1$, $n = 4/5 + λ/5$ for 1 $\leq$ $Ca$ $<$ $Ca_{cr}$ and $n = 1 + λ/5$ for $Ca_{cr} \leq Ca \leq 300$, where $Re$ is Reynolds number. We briefly discuss the combined effect of the skin friction drag and buoyancy on the drag reduction. An assessment of the feasibility of experiments is presented in order to translate the present model to physical systems. △ Less

Submitted 1 December, 2017; originally announced December 2017.

Droplet Bouncing and Breakup during Impact on Microgrooved Surface

Authors: Laxman K. Malla, Nagesh D. Patil, Rajneesh Bhardwaj, Adrian Neild

Abstract: We experimentally investigate impact dynamics of a microliter water droplet on a hydrophobic microgrooved surface. The surface is fabricated using photolithography and high-speed visualization is employed to record the time-varying droplet shapes in transverse as well as longitudinal direction. The effect of the pitch of the grooved surface and Weber number on droplet dynamics and impact outcome a… ▽ More We experimentally investigate impact dynamics of a microliter water droplet on a hydrophobic microgrooved surface. The surface is fabricated using photolithography and high-speed visualization is employed to record the time-varying droplet shapes in transverse as well as longitudinal direction. The effect of the pitch of the grooved surface and Weber number on droplet dynamics and impact outcome are studied. At low pitch and Weber number, the maximum droplet spreading is found to be greater in the longitudinal direction than the transverse direction to the grooves. The preferential spreading inversely scales with the pitch at a given Weber number. In this case, the outcome is no bouncing (NB); however, this changes at larger pitch or Weber number. Under these conditions, the following outcomes are obtained as function of the pitch and Weber number - droplet completely bounces off the surface (CB), bouncing occurs with droplet breakup (BDB) or no bouncing due to Cassie to Wenzel wetting transition (NBW). In BDB and NBW, the liquid penetrates the grooves partially or completely beneath the droplet due to the wetting transition. The former results in droplet breakup alongside bouncing while the latter suppresses the bouncing. These outcomes are demarcated on Weber number - dimensionless pitch plane and the proposed regime map suggests the existence of a critical Weber number or pitch for the transition from one regime to the other. CB and BDB are quantified by plotting coefficient of restitution of the bouncing droplet and volume of daughter droplet left on the surface, respectively. The critical Weber number needed for the transition from CB to BDB is estimated using an existing mathematical model and is compared with the measurements. The comparison is good and provides insights in mechanism of the liquid penetration into the grooves. △ Less

Submitted 28 August, 2017; originally announced August 2017.

A Fluid-Structure Interaction Solver for Compressible Flows with Applications in Blast Loading on Thin Elastic Structures

Authors: Shantanu Bailoor, Aditya Annangi, Jung Hee Seo, Rajneesh Bhardwaj

Abstract: We report development and application of a fluid-structure interaction (FSI) solver for compressible flows with large-scale flow-induced deformation of the structure. The FSI solver utilizes partitioned approach to strongly couple a sharp-interface immersed boundary method based flow solver with an open-source finite-element structure dynamics solver. The flow solver is based on a higher-order fin… ▽ More We report development and application of a fluid-structure interaction (FSI) solver for compressible flows with large-scale flow-induced deformation of the structure. The FSI solver utilizes partitioned approach to strongly couple a sharp-interface immersed boundary method based flow solver with an open-source finite-element structure dynamics solver. The flow solver is based on a higher-order finite-difference method on Cartesian grid and employs ghost-cell methodology to impose boundary conditions on the immersed boundary. A higher-order accuracy near the immersed boundary is achieved by combining the ghost-cell approach with a weighted least-square error method based on a higher-order approximate polynomial. The second order spatial accuracy of the flow solver is established by performing a grid refinement study. The structure solver is validated with a canonical elastostatics problem. The FSI solver is validated with published measurements and simulations for the large-scale deformation of a thin elastic steel panel subjected to blast loading in a shock tube. The solver correctly predicts oscillating behavior of the tip of the panel with reasonable fidelity and computed shock wave propagation is qualitatively consistent with the published results. In order to demonstrate the fidelity of the solver and to investigate coupled physics of the shock-structure interaction for a thin elastic plate, we employ the solver for simulating 6.4 kg TNT blast loading on the thin elastic plate. The initial conditions of the blast are taken from field tests reported in the literature. Using numerical schlieren, the shock front propagation, Mach reflection and vortex shedding at the tip of the plate are visualized during the shock wave impact on the plate. We discuss coupling between the non-linear dynamics of the plate and blast loading. △ Less

Submitted 23 May, 2017; originally announced May 2017.

The Response of an Elastic Splitter Plate Attached to a Cylinder to Laminar Pulsatile Flow

Authors: Anup Kundu, Atul K. Soti, Rajneesh Bhardwaj, Mark C. Thompson

Abstract: The flow-induced deformation of a thin, elastic splitter plate attached to the rear of a circular cylinder and subjected to laminar pulsatile inflow is investigated. The cylinder and elastic splitter plate are contained within a narrow channel and the Reynolds number is mostly restricted to Re = 100, primarily covering the two-dimensional flow regime. An in-house fluid-structure interaction code i… ▽ More The flow-induced deformation of a thin, elastic splitter plate attached to the rear of a circular cylinder and subjected to laminar pulsatile inflow is investigated. The cylinder and elastic splitter plate are contained within a narrow channel and the Reynolds number is mostly restricted to Re = 100, primarily covering the two-dimensional flow regime. An in-house fluid-structure interaction code is employed for simulations, which couples a sharp-interface immersed boundary method for the fluid dynamics with a finite-element method to treat the structural dynamics. The structural solver is implicitly (two-way) coupled with the flow solver using a partitioned approach. This implicit coupling ensures numerical stability at low structure-fluid density ratios. A power spectrum analysis of the time-varying plate displacement shows that the plate oscillates at more than a single frequency for pulsatile inflow, compared to a single frequency observed for steady inflow. The multiple frequencies obtained for the former case can be explained by beating between the applied and plate oscillatory signals. The plate attains a self-sustained time-periodic oscillation with a plateau amplitude in the case of steady flow, while the superimposition of pulsatile inflow with induced plate oscillation affects the plateau amplitude. Lock-in of the plate oscillation with the pulsatile inflow occurs at a forcing frequency that is twice of the plate natural frequency in a particular mode and this mode depends on the plate length. The plate displacement as well as pressure drag increases at the lock-in condition. The percentage change in the maximum plate displacement, and skin-friction and pressure drag coefficients on the plate, due to pulsatile inflow is quantified. The non-linear dynamics of the plate and its coupling with the pulsatile flow are briefly discussed. △ Less

Submitted 9 November, 2016; originally announced November 2016.

Effects of Substrate Heating and Wettability on Evaporation Dynamics and Deposition Patterns for a Sessile Water Droplet Containing Colloidal Particles

Authors: Nagesh D. Patil, Prathamesh G. Bange, Rajneesh Bhardwaj, Atul Sharma

Abstract: Effects of substrate temperature, substrate wettability and particles concentration are experimentally investigated for evaporation of a sessile water droplet containing colloidal particles. Time-varying droplet shapes and temperature of the liquid-gas interface are measured using high-speed visualization and infrared thermography, respectively. The motion of the particles inside the evaporating d… ▽ More Effects of substrate temperature, substrate wettability and particles concentration are experimentally investigated for evaporation of a sessile water droplet containing colloidal particles. Time-varying droplet shapes and temperature of the liquid-gas interface are measured using high-speed visualization and infrared thermography, respectively. The motion of the particles inside the evaporating droplet is qualitatively visualized by an optical microscope and profile of final particle deposit is measured by an optical profilometer. On a non-heated hydrophilic substrate, a ring-like deposit forms after the evaporation, as reported extensively in the literature; while on a heated hydrophilic substrate, a thinner ring with an inner deposit is reported in the present work. The latter is attributed to Marangoni convection and recorded motion of the particles as well as measured temperature gradient across the liquid-gas interface confirms this hypothesis. The thinning of the ring scales with the substrate temperature and is reasoned to stronger Marangoni convection at larger substrate temperature. In case of a non-heated hydrophobic substrate, an inner deposit forms due to very early depinning of the contact line. On the other hand, in case of a heated hydrophobic substrate, the substrate heating as well as larger particle concentration helps in the pinning of the contact line, which results in a thin ring with an inner deposit. We propose a regime map for predicting three types of deposits namely, ring, thin ring with inner deposit and inner deposit - for varying substrate temperature, substrate wettability and particles concentration. A first-order model corroborates the liquid-gas interface temperature measurements and variation in the measured ring profile with the substrate temperature. △ Less

Submitted 19 October, 2016; originally announced October 2016.

Droplet Impact Dynamics on Micropillared Hydrophobic Surfaces

Authors: Nagesh D. Patil, Rajneesh Bhardwaj, Atul Sharma

Abstract: The effect of pitch of the pillars and impact velocity are studied for the impact dynamics of a microliter water droplet on a micropillared hydrophobic surface. The results are presented qualitatively by the high-speed photography and quantitatively by the temporal variation of wetted diameter and droplet height. A characterization of the transient quantitative results is a novel aspect of our wor… ▽ More The effect of pitch of the pillars and impact velocity are studied for the impact dynamics of a microliter water droplet on a micropillared hydrophobic surface. The results are presented qualitatively by the high-speed photography and quantitatively by the temporal variation of wetted diameter and droplet height. A characterization of the transient quantitative results is a novel aspect of our work. Three distinct regimes, namely, non-bouncing, complete bouncing and partial bouncing are presented. A critical pitch as well as impact velocity exists for the transition from one regime to another. This is explained with a demonstration of Cassie to Wenzel wetting transition in which the liquid penetrates in the grooves between the pillars at larger pitch or impact velocity. The regimes are demarcated on a map of pitch and impact velocity. A good agreement is reported between the present measurements and published analytical models. △ Less

Submitted 5 January, 2016; originally announced January 2016.

Comments: Experimental Thermal and Fluid Science, 2015

Computational Study of Bouncing and Non-bouncing Droplets Impacting on Superhydrophobic Surfaces

Authors: Prathamesh G. Bange, Rajneesh Bhardwaj

Abstract: We numerically investigate bouncing and non-bouncing of droplets during isothermal impact on superhydrophobic surfaces. An in-house, experimentally-validated, finite-element method based computational model is employed to simulate the droplet impact dynamics and transient fluid flow within the droplet. The liquid-gas interface is tracked accurately in Lagrangian framework with dynamic wetting boun… ▽ More We numerically investigate bouncing and non-bouncing of droplets during isothermal impact on superhydrophobic surfaces. An in-house, experimentally-validated, finite-element method based computational model is employed to simulate the droplet impact dynamics and transient fluid flow within the droplet. The liquid-gas interface is tracked accurately in Lagrangian framework with dynamic wetting boundary condition at three-phase contact line. The interplay of kinetic, surface and gravitational energies is investigated via systematic variation of impact velocity and equilibrium contact angle. The numerical simulations demonstrate that the droplet bounces off the surface if the total droplet energy at the instance of maximum recoiling exceeds the initial surface and gravitational energy, otherwise not. The non-bouncing droplet is characterized by the oscillations on the free surface due to competition between the kinetic and surface energy. The droplet dimensions and shapes obtained at different times by the simulations are compared with the respective measurements available in the literature. Comparisons show good agreement of numerical data with measurements and the computational model is able to reconstruct the bouncing and non-bouncing of the droplet as seen in the measurements. The simulated internal flow helps to understand the impact dynamics as well as the interplay of the associated energies during the bouncing and non-bouncing. △ Less

Submitted 5 January, 2016; originally announced January 2016.

Comments: Theoretical and Computational Fluid Dynamics, 2015

Numerical study of Heat Transfer Enhancement by Deformable Twin Plates in Laminar Heated Channel flow

Authors: Rakshitha U. Joshi, Atul K. Soti, Rajneesh Bhardwaj

Abstract: Fluid-structure interaction (FSI) of thin flexible fins coupled with convective heat transfer has applications in energy harvesting and in understanding functioning of several biological systems. We numerically investigate FSI of the thin flexible fins involving large-scale flow-induced deformation as a potential heat transfer enhancement technique. An in-house, strongly-coupled fluid-structure in… ▽ More Fluid-structure interaction (FSI) of thin flexible fins coupled with convective heat transfer has applications in energy harvesting and in understanding functioning of several biological systems. We numerically investigate FSI of the thin flexible fins involving large-scale flow-induced deformation as a potential heat transfer enhancement technique. An in-house, strongly-coupled fluid-structure interaction (FSI) solver is employed in which flow and structure solvers are based on sharp-interface immersed boundary and finite element method, respectively. We consider twin flexible fins in a heated channel with laminar pulsating cross flow. The vortex ring past the fin sweep higher sources of vorticity generated on the channel walls out into the downstream - promoting the mixing of the fluid. The moving fin assists in convective mixing, augmenting convection in bulk and at the walls; and thereby reducing thermal boundary layer thickness and improving heat transfer at the channel walls. The thermal augmentation is quantified in term of instantaneous Nusselt number at the wall. Results are presented for two limiting cases of thermal conductivity of the fin - an insulated fin and a highly conductive fin. We discuss feasibility of flexible fins and effect of flow cycle-to-cycle variation on the Nusselt number. Finally, we investigate the effect of important problem parameters - Young's Modulus, flow frequency and Prandtl number - on the thermal augmentation. △ Less

Submitted 16 November, 2016; v1 submitted 13 September, 2015; originally announced September 2015.

Comments: arXiv admin note: text overlap with arXiv:1501.01807

Journal ref: Computational Thermal Sciences, Vol. 7, pp 1-10, 2015

Flow-Induced Deformation of a Flexible Thin Structure as Manifestation of Heat Transfer Enhancement

Authors: Atul Kumar Soti, Rajneesh Bhardwaj, John Sheridan

Abstract: Flow-induced deformation of thin structures coupled with convective heat transfer has potential applications in energy harvesting and is important for understanding functioning of several biological systems. We numerically demonstrate large-scale flow-induced deformation as an effective passive heat transfer enhancement technique. An in-house, strongly-coupled fluid-structure interaction (FSI) sol… ▽ More Flow-induced deformation of thin structures coupled with convective heat transfer has potential applications in energy harvesting and is important for understanding functioning of several biological systems. We numerically demonstrate large-scale flow-induced deformation as an effective passive heat transfer enhancement technique. An in-house, strongly-coupled fluid-structure interaction (FSI) solver is employed in which flow and structure solvers are based on sharp-interface immersed boundary and finite element method, respectively. In the present work, we validate convective heat transfer module of the in-house FSI solver against several benchmark examples of conduction and convective heat transfer including moving structure boundaries. The thermal augmentation is investigated as well as quantified for the flow-induced deformation of an elastic thin plate attached to lee side of a rigid cylinder in a heated channel laminar flow. We show that the wake vortices past the plate sweep higher sources of vorticity generated on the channel walls out into the high velocity regions, promoting the mixing of the fluid. The self-sustained motion of the plate assists in convective mixing, augmenting convection in bulk and near the walls, and thereby reducing thermal boundary layer thickness as well as improving Nusselt number at the channel walls. We quantify the thermal improvement with respect to channel flow without any bluff body and analyze the role of Reynolds number, Prandtl number and material properties of the plate in the thermal augmentation. △ Less

Submitted 8 January, 2015; originally announced January 2015.

Interfacial temperature measurements, high-speed visualization and finite-element simulations of droplet impact and evaporation on a solid surface

Authors: Rajneesh Bhardwaj, Jon P. Longtin, Daniel Attinger

Abstract: The objective of this work is to investigate the coupling of fluid dynamics, heat transfer and mass transfer during the impact and evaporation of droplets on a heated solid substrate. A laser-based thermoreflectance method is used to measure the temperature at the solid-liquid interface, with a time and space resolution of 100 μs and 20 μm, respectively. Isopropanol droplets with micro- and nanoli… ▽ More The objective of this work is to investigate the coupling of fluid dynamics, heat transfer and mass transfer during the impact and evaporation of droplets on a heated solid substrate. A laser-based thermoreflectance method is used to measure the temperature at the solid-liquid interface, with a time and space resolution of 100 μs and 20 μm, respectively. Isopropanol droplets with micro- and nanoliter volumes are considered. A finite-element model is used to simulate the transient fluid dynamics and heat transfer during the droplet deposition process, considering the dynamics of wetting as well as Laplace and Marangoni stresses on the liquid-gas boundary. For cases involving evaporation, the diffusion of vapor in the atmosphere is solved numerically, providing an exact boundary condition for the evaporative flux at the droplet-air interface. High-speed visualizations are performed to provide matching parameters for the wetting model used in the simulations. Numerical and experimental results are compared for the transient heat transfer and the fluid dynamics involved during the droplet deposition. Our results describe and explain temperature oscillations at the drop-substrate interface during the early stages of impact. For the first time, a full simulation of the impact and subsequent evaporation of a drop on a heated surface is performed, and excellent agreement is found with the experimental results. Our results also shed light on the influence of wetting on the heat transfer during evaporation. △ Less

Submitted 13 October, 2010; originally announced October 2010.

Journal ref: International Journal of heat and Mass transfer, Vol 53, pp 3733-3744, 2010

Self-assembly of colloidal particles from evaporating droplets: role of DLVO interactions and proposition of a phase diagram

Authors: Rajneesh Bhardwaj, Xiaohua Fang, Ponisseril Somasundaran, Daniel Attinger

Abstract: The shape of deposits obtained from drying drops containing colloidal particles matters for technologies such as inkjet printing, microelectronics and bioassay manufacturing. In this work, the formation of deposits during the drying of nanoliter drops containing colloidal particles is investigated experimentally with microscopy and profilometry, and theoretically with an inhouse finite-element cod… ▽ More The shape of deposits obtained from drying drops containing colloidal particles matters for technologies such as inkjet printing, microelectronics and bioassay manufacturing. In this work, the formation of deposits during the drying of nanoliter drops containing colloidal particles is investigated experimentally with microscopy and profilometry, and theoretically with an inhouse finite-element code. The system studied involves aqueous drops containing titania nanoparticles evaporating on a glass substrate. Deposit shapes from spotted drops at different pH values are measured using a laser profilometer. Our results show that the pH of the solution influences the dried deposit pattern, which can be ring-like or more uniform. The transition between these patterns is explained by considering how DLVO interactions such as the electrostatic and van der Waals forces modify the particle deposition process. Also a phase diagram is proposed to describe how the shape of a colloidal deposit results from the competition between three flow patterns: a radial flow driven by evaporation at the wetting line, a Marangoni recirculating flow driven by surface tension gradients, and the transport of particles towards the substrate driven by DLVO interactions. This phase diagram explains three types of deposits commonly observed experimentally, such as a peripheral ring, a small central bump, or a uniform layer. Simulations and experiments are found in very good agreement. △ Less

Submitted 12 October, 2010; originally announced October 2010.

Journal ref: Langmuir, Vol 26 (11), pp 7833-7842, 2010

Pattern formation during the evaporation of a colloidal nanoliter drop: a numerical and experimental study

Authors: Rajneesh Bhardwaj, Xiaohua Fang, Daniel Attinger

Abstract: An efficient way to precisely pattern particles on solid surfaces is to dispense and evaporate colloidal drops, as for bioassays. The dried deposits often exhibit complex structures exemplified by the coffee ring pattern, where most particles have accumulated at the periphery of the deposit. In this work, the formation of deposits during the drying of nanoliter colloidal drops on a flat substrate… ▽ More An efficient way to precisely pattern particles on solid surfaces is to dispense and evaporate colloidal drops, as for bioassays. The dried deposits often exhibit complex structures exemplified by the coffee ring pattern, where most particles have accumulated at the periphery of the deposit. In this work, the formation of deposits during the drying of nanoliter colloidal drops on a flat substrate is investigated numerically and experimentally. A finite-element numerical model is developed that solves the Navier-Stokes, heat and mass transport equations in a Lagrangian framework. The diffusion of vapor in the atmosphere is solved numerically, providing an exact boundary condition for the evaporative flux at the droplet-air interface. Laplace stresses and thermal Marangoni stresses are accounted for. The particle concentration is tracked by solving a continuum advection-diffusion equation. Wetting line motion and the interaction of the free surface of the drop with the growing deposit are modeled based on criteria on wetting angles. Numerical results for evaporation times and flow field are in very good agreement with published experimental and theoretical results. We also performed transient visualization experiments of water and isopropanol drops loaded with polystyrene microsphere evaporating on respectively glass and polydimethylsiloxane substrates. Measured evaporation times, deposit shape and sizes, and flow fields are in very good agreement with the numerical results. Different flow patterns caused by the competition of Marangoni loops and radial flow are shown to determine the deposit shape to be either a ring-like pattern or a homogeneous bump. △ Less

Submitted 12 October, 2010; originally announced October 2010.

Journal ref: New Journal of Physics, Vol 11, pp. 075020, 2009

A numerical investigation on the influence of liquid properties and interfacial heat transfer during microdroplet deposition onto a glass substrate

Authors: Rajneesh Bhardwaj, Jon P. Longtin, Daniel Attinger

Abstract: This work investigates the impingement of a liquid microdroplet onto a glass substrate at different temperatures. A finite-element model is applied to simulate the transient fluid dynamics and heat transfer during the process. Results for impingement under both isothermal and non-isothermal conditions are presented for four liquids: isopropanol, water, dielectric fluid (FC-72) and eutectic tin-lea… ▽ More This work investigates the impingement of a liquid microdroplet onto a glass substrate at different temperatures. A finite-element model is applied to simulate the transient fluid dynamics and heat transfer during the process. Results for impingement under both isothermal and non-isothermal conditions are presented for four liquids: isopropanol, water, dielectric fluid (FC-72) and eutectic tin-lead solder (63Sn-37Pb). The objective of the work is to select liquids for a combined numerical and experimental study involving a high resolution, laser-based interfacial temperature measurement to measure interfacial heat transfer during microdroplet deposition. Applications include spray cooling, micro-manufacturing and coating processes, and electronics packaging. The initial droplet diameter and impact velocity are 80 μm and 5 m/s, respectively. For isothermal impact, our simulations with water and isopropanol show very good agreement with experiments. The magnitude and rates of spreading for all four liquids are shown and compared. For non-isothermal impacts, the transient drop and substrate temperatures are expressed in a non-dimensional way. The influence of imperfect thermal contact at the interface between the drop and the substrate is assessed for a realistic range of interfacial Biot numbers. We discuss the coupled influence of interfacial Biot numbers and hydrodynamics on the initiation of phase change. △ Less

Submitted 12 October, 2010; originally announced October 2010.

Journal ref: International Journal of heat and mass transfer, Vol. 50, pp. 2912-2923, 2007

Non-isothermal wetting during impact of millimeter size water drop on a flat substrate: numerical investigation and comparison with high-speed visualization experiments

Authors: Rajneesh Bhardwaj, Daniel Attinger

Abstract: The objective of this work is to develop and validate a numerical model to study wetting during the impact of millimeter-size drops on a flat, smooth, solid substrate under isothermal or non-isothermal conditions. A finite-element modeling is used to simulate the transient fluid dynamics and heat transfer, considering Laplace forces on the liquid-gas boundary. The Lagrangian scheme allows a very p… ▽ More The objective of this work is to develop and validate a numerical model to study wetting during the impact of millimeter-size drops on a flat, smooth, solid substrate under isothermal or non-isothermal conditions. A finite-element modeling is used to simulate the transient fluid dynamics and heat transfer, considering Laplace forces on the liquid-gas boundary. The Lagrangian scheme allows a very precise tracking of the free surface deformation. In this work, the numerical model is extended to account for a temperature-dependent viscosity and for dynamic wetting at the contact line. Numerical results are presented to study the influence of the kinetic wetting parameter on the wetting incipience and behavior. Our results show the influence of wetting on the spreading and the transient drop shape. Also, numerical results are compared with high-speed visualization, for cases of isothermal and non-isothermal impact. Matching between simulations and high-speed visualization allows the determination of the value of the kinetic wetting parameter. Our main finding is that warm drops spread more than cold drops because of a reduction of viscous forces, and not because of an increase of wetting. △ Less

Submitted 12 October, 2010; originally announced October 2010.

Journal ref: International Journal of heat and fluid flow, Vol. 29, pp. 1422-1435, 2008