Department of Mechanical Engineering

IIT Madras is a public technical and research university located in Chennai, Tamil Nadu. Founded in 1959, it is recognised as an Institute of National Importance.

IIT Madras has been ranked as the top engineering institute in India for four years in a row by the National Institutional Ranking Framework of the MHRD

It currently offers undergraduate, postgraduate and research degrees across 16 disciplines in Engineering, Sciences, Humanities and Management. About 596 faculty belonging to science and engineering departments and centres of the Institute are engaged in teaching, research and industrial consultancy.

IIT Madras

Hand rehabilitation requires intensive training of various gross and fine movements. Robotic devices have been developed and tested for implementing intense training of different hand functions in the current literature. Current hand rehabilitation robots can be broadly grouped into two categories: (a) simple robots with one or two degrees-of-freedom (DOF) that train only one or two hand functions; or (b) complex robots with several DOF capable of training a wide range of functions. Thus, to train different hand functions, we either need a set of simple robots or one complex robot, both of which are not economically viable solutions. A potential solution is to have a single DOF robot that can train a wide range of functions, i.e. a robot with a single actuator, along with a set of easily pluggable passive mechanisms for different hand functions. In this paper, we present the design and analysis of various kinematic mechanisms for a single DOF hand rehabilitation robot plug and train hand rehabilitation robot (PLUTO). Three different kinematic mechanisms have been designed and analyzed to train four different functions: wrist flexion/extension, wrist ulnar and radial deviation, forearm pronation/supination, and gross hand opening-closing. Proper alignment of the robot and human joints is essential for safe and appreciate interaction with a human. Misalignment in the robot and human joints can cause undesired forces on the human joint, makes the interaction potentially unsafe. The problem of misalignment was addressed in the proposed mechanisms through the use of appropriate passive DOFs between the robot and human joints. This paper presents the design and analysis of the three mechanisms for the different hand functions. The analysis of the mechanisms were carried out by considering variations in human hand anthropometry, and uncertainty in the robot-human axis alignments. © 2021, Springer Nature Singapore Pte Ltd.

World University root node

Lecture Notes in Mechanical Engineering

Design of Single Degree-of-Freedom Mechanisms for Hand Neurorehabilitation

Rotating blades are widely used in turbomachinery applications. In these applications, blades fail frequently because of large amplitude vibrations. These large amplitude vibrations are related to the resonances of the blade structure. Thus, accurate estimation of natural frequencies of such rotating structures is of utmost importance. As documented in the literature, a widely used model of such blade structures consists of rotating beams or plates. The objective of this work is to determine the natural frequencies of rotating plate structures. Starting from the non-dimensional governing equation of the rotating staggered cantilever plate as presented by Sun et al. (J Sound Vib 332(5):1355–1371, 2013) [1], the corresponding weighted residual statement is derived. In the present study, a Galerkin-based methodology is implemented to obtain the modal characteristics. Towards this end, we choose six linearly independent shape functions in the form of product of Clamped-Free (C-F) and Free-Free (F-F) beam mode shapes. The characteristic equation for the determination of the natural frequencies is formulated. This sixth-order polynomial equation is solved numerically to obtain the natural frequencies of the rotating staggered cantilever plate. The results for the non-dimensional natural frequencies as a function of the non-dimensional rotating speed are presented in the form of natural frequency maps. Further, these results are validated using Finite Element (FE) simulation. Towards this end, a pre-stressed modal analysis is performed in ANSYS. It is noted that the results obtained by the two methods are in good correlation. © Springer Nature Singapore Pte Ltd. 2021.

Natural frequencies of rotating cantilever plates

Isobaric-isothermal flash calculations are at the heart of any process calculation. In this paper, we propose two new methods for flash calculations at a specified temperature and pressure. The governing equations derived can not only be solved rapidly but also converge correctly even very close to the critical point with the methods presented in this paper. The derived equations show that simplified equations have been used in the past by other workers. The use of full equations results in faster convergence. Two different approaches, the pressure and density-based, are presented in this work. The density-based approach is particularly useful in reducing the computational time substantially in the case of multiparameter equations of state such as the PC-SAFT and the Helmholtz energy equations of state. Many examples are provided to demonstrate the efficacy of the methods. The methods presented have immediate use in process simulation software and thermodynamic programs. © 2021 Elsevier B.V.

Fluid Phase Equilibria

New pressure and density based methods for isothermal-isobaric flash calculations

Literature reveals that machining a silicon wafer is very arduous due to its high brittle nature and low fracture toughness. Researchers have attempted to machine features on a silicon wafer using both conventional and non-conventional processes and both have their merits &amp; demerits. Abrasive jet machining (AJM) is one of the efficient non-conventional machining methods for machining hard and brittle materials. Removal of material occurs by impact erosion and is free from Heat Affected Zone (HAZ) and resolidification. These advantages make AJM a potential process for machining silicon wafers. Therefore an attempt has been made to machine channels on silicon wafers using a Tailor-made AJM setup. The process parameters selected for machining channels on a silicon wafer are air pressure, Feed rate (FR), and Standoff distance (SOD). The output responses considered for characterization are material removal rate (MRR), channel characteristics like width, depth, and Surface Roughness (Ra). The machined Channels are found to have sufficient depth and good surface finish and are free from heat affected zone and resolidification. X-Ray Diffraction (XRD) analysis has indicated the presence of embedded SiC particles in the channel. GRA method is utilized to achieve an optimal condition of higher depth and MRR, smaller width, and Ra. Also, the significance and percentage contribution of process parameters are found out through ANOVA and obtained results are presented in the paper. © 2021, Springer Nature B.V.

Silicon

Machining, Characterization and Optimization: A Novel Approach for Machining Channels on Silicon Wafer Using Tailor-Made Micro Abrasive Jet Machining

For defense applications, rapid X-ray inspection of propellant samples is essential for the identification and assessment of defects. Automation of this process using artificial intelligence is possible by properly training a neural network model. Convolution Neural Networks (CNNs) have recently demonstrated excellent success in both the tasks of image recognition and localisation using an adequate amount of data. In real-world, it’s not an easy task to produce the correct amount of experimental data required for the deep neural network to operate. In this work, we propose a method for producing synthetic radiographic data that is supported by ray tracing based radiographic simulations for the deep learning algorithms to automatically detect anomaly in X-ray images. The simulation results, which are then supplemented by noise extracted from the experimental data, show a good comparison with the measurements. This Simulation assisted Automatic Defect Recognition (Sim-ADR) system simultaneously perform defect detection and defect instance segmentation. The accuracy of the defect detection system is more than 87% on a testing set included 416 images. © 2021, The Author(s), under exclusive licence to Springer Science+Business Media, LLC part of Springer Nature.

Department of Humanities and Social Sciences

Journal of Nondestructive Evaluation

Automated Defect Recognition on X-ray Radiographs of Solid Propellant Using Deep Learning Based on Convolutional Neural Networks

Nacre, a composite layer present in sea-shells, exhibits a remarkable combination of toughness, strength, and stiffness through its brick–mortar micro-structure, acting as a template for novel materials. Strength, one of nacre's important properties is highly variable due to distribution of underlying material's properties as well as various defects present in its micro-structure. Presently, researchers assume the Weakest-link hypothesis and consequently use the Weibull distribution to model this variability. However, this assumption is theoretically unproven for biological materials such as Nacre and extrapolation of the same to predict rare but catastrophic behavior would be incorrect. In this article, the suitability of Weibull distribution to account for the variability of strength in nacre is tested, using multi-scale models, developed using FEM. Through Monte Carlo based numerical experiments and Renormalization Group (RG) based arguments, it is shown that the Weakest-link hypothesis, which is commonly used to justify the use of the Weibull distribution, does not seem to hold for nacre. Micromechanics and non-local homogenization based distributions such as the one suggested by Luo and Bazant (2019) might be more appropriate for accurate extrapolation to the low probability tail. © 2021 Elsevier Ltd

Mechanics of Materials

Testing Weibull as a viable statistical strength distribution for Nacre

To avoid excessive tissue removal and collateral damage, the high-power density laser is apt for dental surgery also need to have high precision. For high-precision dental surgery with minimal tissue damage, the present work frames a method to predict laser ablation profile based on surface morphology and chemical composition of dentin. The surface morphology and chemical composition were studied on different dentin samples using scanning electron microscope (SEM) and Energy Dispersive X-ray Analysis (EDAX), respectively. The key laser ablation parameters (ω0, Deff, and Fth) were determined by laser irradiation study using 800 nm, Ti:Sapphire femtosecond laser at processing condition of 100 fs, 10 kHz and 10 mm/s. The dentin samples show a strong linear correlation between physiochemical characteristics and laser ablation parameters. The surface morphology exhibits a negative linear correlation with threshold fluence, whereas the converse is true for chemical composition. The laser ablation parameters of a random dentin sample are derived from the knowledge of linearity data. From the obtained laser ablation parameters, the complete theoretical ablation profile is constructed and validated with experimental ablation profile. Even though the surface morphology of dentin shows high linearity, the concentration of Ca and P can be used as the most feasible probe in clinical settings. Furthermore, the laser ablation rate and ablation efficiency are predicted by the method to optimize the laser processing condition for any specific teeth. The versatility of the method overcomes the problem of heterogeneity on various teeth and simplifies the method of finding optimal laser processing condition for immaculate laser surgery. © 2020 Wiley Periodicals LLC.

Journal of Biomedical Materials Research - Part B Applied Biomaterials

Physiochemical characteristics: A robust tool to overcome teeth heterogeneity on predicting laser ablation profile

The introduction of foreign cargo into living cells with high delivery efficiency and cell viability is a challenge in cell biology and biomedical research. Here, we demonstrate a nanosecond pulse laser-activated photoporation for highly efficient intracellular delivery using titanium dioxide (TiO2) microspikes as a substratum. The TiO2microspikes were formed on titanium (Ti) substrate using an electrochemical anodization process. Cells were cultured on top of the TiO2microspikes as a monolayer, and the biomolecule was added. Due to pulse laser exposure of the TiO2microspike-cell membrane interface, the microspikes heat up and induce cavitation bubbles, which rapidly grow, coalesce and collapse to induce explosion, resulting in very strong fluid flow at the cell membrane surface. Thus, the cell plasma membrane disrupts and creates transient nanopores, allowing delivery of biomolecules into cells by a simple diffusion process. By this technique, we successfully delivered propidium iodide (PI) dye in HeLa cells with high delivery efficiency (93%) and high cell viability (98%) using 7 mJ pulse energy at 650 nm wavelength. Thus, our TiO2microspike-based platform is compact, easy to use, and potentially applicable for therapeutic and diagnostic purposes. © The Royal Society of Chemistry 2021.

Department of Engineering Design

Fulltext

RSC Advances

Fabrication of TiO2microspikes for highly efficient intracellular delivery by pulse laser-assisted photoporation

This paper aims to develop a technique for estimating an aluminium/epoxy/aluminium lap shear joint stiffness. Shear horizontal guided waves are employed using periodic permanent magnet electromagnetic acoustic transducers (PPM EMATs). Based on the source influence of the EMATs and sensitivity of SH-like mode to adhesion, an appropriate transduction regime is selected. Different modality of inspection, such as reflection and transmission mode, is explored. The arrival time of the received signal in transmission mode depends on the adhesion level. A procedure for calculating the energy velocity in the bond and inverse calculation of bond stiffness is proposed. Experiments were performed on samples of different surface treatments. Based on the proposed method, the stiffness of the samples was estimated. Results show the technique's potential to quantify the adhesion level rather than merely classifying as good, weak, and bad bonds. © 2020 Elsevier Ltd

International Journal of Adhesion and Adhesives

Bond stiffness estimation with shear horizontal guided waves generated using PPM-EMATs

Polymeric gels, when made into thin films, are reported to depict a reversible folding behavior on exposure to favorable solvents as a result of the coupled diffusion–deformation mechanism. In this work, we develop a model based on the coupled diffusion–deformation mechanism which can predict the large deformation caused during this reversible folding behavior of a polymeric thin film. The formulation incorporates a concentration dependent Young's modulus variation. It is found that the incorporation of concentration dependent Young's modulus of the film in the formulation is essential to predict the folding curvatures accurately. The results obtained using this model are validated with the experimental results of the Chitosan–Water system by comparing the folding characteristics — intermediate folding stages and curvature, while accurately explaining the mechanism of reversible folding. The model is applied to other material systems such as PVA–Water and PDMS–Hexane systems. Furthermore, we demonstrate that the developed model successfully predicts the buckling and wrinkling phenomenon observed experimentally in stimuli-responsive polymer thin films. © 2020 Elsevier Ltd