Search

In this paper, we perform molecular dynamics simulations to investigate the performance of multilayer graphene slit membranes. Graphene slit membranes at a critical slit size have been found to be promising desalination membranes. In this contribution, it is shown that multilayer slit membranes have the potential to provide significantly better permeability while retaining outstanding salt rejection. Improved permeability of the membrane is achieved by using slits of widths larger than the critical slit size required to reject salt through size exclusion, and desalination of sea water is performed by increased resistance to salt passage through the multilayering. To facilitate the design process of future multilayer membranes, we analyze the flow resistance of the membrane as a combination of electrical resistors in series and show that this analogy works for membranes where the layers possess the same slit size, as well as membranes with layers of different slit sizes. Comparing with single layer graphene membranes, it was shown that it is possible to obtain 55% improvement in permeability without loss in salt rejection capabilities through multilayering. This opens up possibilities for membrane designers to be free from the restrictions of using a single layer graphene slit membrane with a fixed slit width.

The expanding capabilities and decreasing costs of additive manufacturing have resulted in the increased adoption of micro-unmanned aerial vehicles (micro-UAVs) among professionals and hobbyists. Due to safety regulatory requirements of UAV operations, weight is generally the overriding design features of interest for micro-drones, but it often comes as a trade-off against the durability, loading constraints, and other subsystem equipment. Nevertheless, ultra-lightweight structures can be realized through the adoption of both 3D-printing and topology optimization without compromising the structural integrity and overall strength and this article explores the use of these two technologies for designing and manufacturing optimized ultralight micro-UAVs. First, material properties of Nylon 12 (PA12) manufactured using selective laser sintering (SLS) were accurately characterized via mechanical testing and ultrasonic means. These properties were verified by comparing the mechanical response of 3-point and 4-point bending tests with corresponding finite element (FE) simulation. Next, topology optimization was performed to produce an optimized structure of a Z-split configured lightweight micro-quadcopter. The optimized design is then 3D-printed and subsequently validated through a load test for verification against the optimized FE simulation-based design. A close correlation was obtained between the numerical and experimental data, suggesting that topology optimization with 3D printing can be safely and reliably adopted for the design and rapid prototyping of micro-UAVs, whilst catering to different specifications and requirements. National Research Foundation (NRF) Submitted/Accepted version This research is jointly supported by the National Research Foundation, Prime Minister’s Office, Singapore under its Medium-Sized Centre funding scheme, and ST Engineering Aerospace Ltd., under project titled ‘3D Printing of micro UAV’.

It has been shown that the nanoplastic particles present in graphene membranes have a high tendency to cause fouling in them due to the high affinity between graphene and nanoplastic molecules. This poses a significant challenge for the use of graphene membranes for desalination. In this paper, we introduce a double-layer graphene slit membrane as a viable solution to significantly reduce fouling caused by the presence of nanoplastic particles in graphene membranes. The molecular dynamics (MD) simulations performed in this work show that when fouling occurs in a single-layer membrane, the presence of nanoplastics reduces the average permeability by close to 40%, from 1877 LMBH to 1148 LMBH, with a large standard deviation of 26% between runs. With the addition of the secondary membrane, the average permeability increases by 17%, with a significantly reduced standard deviation of 7%. These suggest that the secondary layer acts as a sacrificial shield, attracting the nanoplastic contaminants and preventing them from coming into close proximity with the primary membrane, thus preventing fouling at the primary rejection layer. Furthermore, due to the affinity of the nanoplastic particles with the secondary graphene membrane, this membrane design points toward an effective and efficient way of extracting nanoplastic particles for further analysis or processing.

In the current practice, the one-dimensional nonlocal constitutive relations are always employed to model beam-type structures, regardless of the inherent three-dimensional interactions between atoms, resulting in inaccurately predicted nonlocal structural behaviors. To improve modeling, the present work first reveals and establishes how the nonlocal interactions in beams’ width and height directions substantially affect the bending behaviors of nanobeams. Based on the new concept of a general three-dimensional nonlocal constitutive relation, a three-dimensional nonlocal Euler–Bernoulli beam model is developed for bending analysis. Closed-form solutions for the deflections of simply supported and cantilevered beams are derived. The cross-section effect on the nonlocal stress, the bending moment and the deflection is explored. Moreover, an effective calibration method is proposed for the determination of the intrinsic characteristic length based on molecular dynamics simulations. It has been found that the actually nonlocal physical dimension of beam-type structures is not in agreement with that from the one-dimensional geometric direction of classical beam mechanics, and the beams’ width and height must be taken into consideration, instead of just employing the beam’s length, a practice which has been somewhat blindly undertaken in the current practice without indeed much theoretical support and understanding. A beam problem at nanoscale has to be viewed and treated as a “three-dimensional” geometric and physical problem due to its geometric dimension in length and its physical dimensions in both the thickness and the width. When its intrinsic characteristic length is comparable to its width or thickness, a rigidity-softening effect of a beam can be observed. Further, it is established that the nonlocal effect is more sensitive in the thickness direction, as compared with the width direction.

Some nonlinear systems possess innate capabilities of enhancing weak signal transmissions through a unique process called Stochastic Resonance (SR). However, existing SR mechanism suffers limited signal enhancement from inappropriate entraining signals. Here we propose a new and effective implementation, resulting in a new type of spectral resonance similar to SR but capable of achieving orders of magnitude higher signal enhancement than previously reported. By employing entraining frequency in the range of the weak signal, strong spectral resonances can be induced to facilitate nonlinear modulations and intermodulations, thereby strengthening the weak signal. The underlying physical mechanism governing the behavior of spectral resonances is examined, revealing the inherent advantages of the proposed spectral resonances over the existing implementation of SR. Wide range of parameters have been found for the optimal enhancement of any given weak signal and an analytical method is established to estimate these required parameters. A reliable algorithm is also developed for the identifications of weak signals using signal processing techniques. The present work can significantly improve existing SR performances and can have profound practical applications where SR is currently employed for its inherent technological advantages.

Objective. To investigate the safety and efficacy of the Herbal Medicine C-117 (C-117) formula in the treatment of carotid atherosclerotic vulnerable plaques. Methods. This was a prospective, single-centre, randomized, double-blind study. A total of 120 eligible patients were randomly divided into two groups to receive the C-117 formula or placebo. As the basic treatment, both groups were treated according to the Guidelines for Secondary Prevention of Ischemic Stroke/Transient Ischemic Stroke in China using statins to regulate blood lipids, blood pressure lowering drugs, drugs for controlling blood sugar, and antiplatelet drugs according to the indications. The primary outcomes were the change in stability, the mean change of the plaque Crouse score, and the area and number of bilateral carotid artery plaques before and after 6 months of treatment. The secondary outcomes were the total number of cardiocerebrovascular events during the treatment and follow-up and the mean changes of lipid levels. Result. After 180 days of treatment, the plaque Crouse score(95% CI, 0.39 (0.01-0.77), P=0.046) and plaque area (95% CI, 2.14 (-10.10-14.39), P=0.727) were lower in the C-117 formula group than that before treatment. The plaque Crouse score of the control group (95% CI, 0.17 (-0.24-0.57), P=0.417) was lower than that before treatment, while the plaque area (95% CI, -0.35 (-9.35-8.65), P=0.938) increased, but without statistical significance. There was no significant difference in the reduction of the intima-media thickness (IMT), plaque Crouse score, or plaque area between the two groups after treatment (P>0.05). Subgroup analysis of patients whose Lipitor medication time ≥ 20% of the 6-month treatment showed that the levels of total cholesterol, triglycerides, and low-density lipoprotein were lower in the two groups after treatment than before, and the low-density lipoprotein levels in the C-117 formula group significantly decreased (95% CI, 2.99 (-0.08-0.39), P=0.005), but there was no statistical difference between the two groups after treatment (P>0.05). No serious adverse events occurred in the two groups after 180 days of treatment. Conclusion. The C-117 formula may be antiatherosclerotic by strengthening statins to reduce the low-density lipoprotein levels and reducing the carotid plaque Crouse scores. Clinical trials with large sample sizes, long-term interventions, and follow-up are needed to investigate the efficacy of the C-117 formula. Clinical Trials Registration. This trial is registered with clinicaltrials.gov identifier: NCT03072225 (registered retrospectively on 1st March 2017).

Dynamic characterizations of fractional vibration systems have recently attracted significant research interest. Increasingly, successful applications of fractional derivatives have been found to the modeling of mechanical damping, vibration transmissions, improved fractional vibration controls and nonlinear vibration analyses. To facilitate further development, the eigenvalue problem including its derivatives, which are the central issues of vibration analysis, have to be fully established. This paper examines how eigenvalue and eigenvector derivatives of fractional systems can be derived when system matrices become functions of physical design parameters. First, new important orthonormal constraints are proposed since the modes are no longer orthonormal to the mass matrix, in this case due to its complex and frequency dependent nature. Next, new methods of eigenvector derivatives are developed for distinct eigenvalues for the cases of complete, incomplete and single mode modal data. Realistic and practical FE models incorporating fractional derivatives in the form of viscoelastic supports are employed to demonstrate the numerical accuracy and computational efficiency of the proposed methods. However, when repeated eigenvalues are considered due to structural spatial symmetries, the eigenvector space degenerates and further differentiation of system matrices are required in order to uniquely determine the eigenvector derivatives. Consequently, a new and effective general method is developed which can be applied to compute eigenvector derivatives of repeated eigenvalues with any multiplicity m. A simplified turbine bladed disk vibration model which is known to have repeated eigenvalues due to its cyclic symmetry, is then used to demonstrate the accuracy and salient features of the proposed method.

Fractional derivative is increasingly being deployed to improve existing mathematical models due to its unique capability in describing anomalous behavior and memory effects which are common characteristics of natural phenomena. Improved vibration analysis has been accomplished by introducing fractional derivatives to model viscoelastic damping and vibration propagation through complex media and much research has been carried out to date. However, much of these existing research efforts have been sporadic to the best and there remains a pressing need to develop a consistent and systematic theoretical framework for vibration analysis of fractional systems to synergize for more productive and coordinated efforts in the area. This paper seeks to address some fundamental issues to facilitate further development such as the definition of general form of a fractional vibration system, its eigenvalue problem and methods of solution, definition of frequency response functions and applicability of conventional modal analysis, equivalent eigensystem and its more efficient eigensolution. With these important issues being resolved to clear the myth, vibration studies of fractional systems can be encouraged and expected to grow in a more fruitful direction. New methods developed and concepts discussed in the paper have all been validated through realistic numerical case studies based on a practical GARTEUR structure with viscoelastic supports modeled using fractional derivatives.

Aeroelastic structural systems are intrinsically nonlinear and accurate predictions of dynamic responses of nonlinear aeroelastic systems have become of paramount importance since these directly affect the accuracy and reliability of subsequent stability analyses. Such nonlinear systems can be generally represented with Volterra series whose kernels have been found to be effective in their dynamic characterizations. This paper examines how first- and second-order Volterra kernels of nonlinear aeroelastic systems can be accurately identified and then incorporated into the theoretical models of aeroelastic analyses such as predictions of dynamic response and onset of aeroelastic flutter. A novel identification method based on correlation analysis to extract frequency components has been developed which can be applied to general nonlinear aeroelastic systems to obtain accurately the required Volterra transfer functions. The method is very accurate and extremely robust against measurement noise contaminations in both input and output signals, due to the correlation scheme which effectively filters uncorrelated signal components. Detailed aeroelastic behavior of a representative pitch-plunge airfoil dynamic model with nonlinear pitch stiffness has been examined. Its Volterra transfer functions are then identified which are found to be close to their exact analytical counterparts, though interaction between kernels becomes apparent as input level increases. Once inverse Fourier transformed, these identified Volterra kernels are then included in the modeling of the dynamics of aeroelastic systems for vibration response and flutter. Extensive numerical simulation results have demonstrated that the proposed method is very accurate and resilient to measurement errors when applied to the identification of second-order Volterra kernels, and the improvement in predictions of vibration response and flutter become significant when the contributions of these second-order Volterra kernels are included in the overall aeroelastic system dynamics. The identification and subsequent inclusion of second-order Volterra kernels into system dynamics model offer improved design capabilities of nonlinear aeroelastic structural systems.

This discussion seeks to address some of the inconsistencies and concerns regarding fractional derivatives when applied to vibration analysis of viscoelastically damped beams [1]. Detailed numerical assessments of the major fractional derivatives of concern have been carried out with their relationships numerically examined. Numerical results have shown that two of the key equations of [1] are mathematically inappropriate and their applications could lead to serious modelling errors. A simplified yet representative nonlinear system with fractional viscoelastic damping and cubic stiffness nonlinearity is used to illustrate the existence of periodic steady-state solutions, as well as amplitude frequency characteristics of such fractional nonlinear systems.

In this work, an improved counterflow double-channel micro combustor is designed. Computational Fluid Dynamics software Fluent is used to conduct numerical investigations on the thermal performance comparison between the old and improved counterflow double-channel micro combustors. It is found that the improved counterflow double-channel micro combustor has much higher and more uniform wall temperature compared that of the old one under various hydrogen mass flow rates, hydrogen/air equivalence ratios and solid materials. To make a quantitative comparison, the main results are presented as follows: (a) The improved combustor achieves the largest thermal enhancement at the hydrogen mass flow rate of 5.25 × 10−7 kg/s when the hydrogen mass flow rate is ranged from 5.25 × 10−7 kg/s to 9.8245 × 10−7 kg/s. Namely, the mean temperature of upper and right wall is improved by about 9.66 K and 13.53 K, respectively, and the mean nonuniformity of upper and right wall temperature is reduced by about 23.24% and 26.79%, respectively. (b) The improved combustor achieves the largest thermal enhancement at the hydrogen/air equivalence ratio of 0.6, when the hydrogen/air equivalence ratio is ranged from 0.9 to 0.5. Namely, the mean temperature of upper and right wall is improved by about 27.55 K and 29.55 K, respectively, and the mean nonuniformity of upper and right wall temperature is reduced by about 23.24% and 19.51%, respectively. (c) The improved combustor achieves the largest thermal enhancement at the solid material of silicon carbide when the solid material is changed from quartz to silicon carbide. Namely, the mean temperature of upper and right wall is improved by about 16.77 K and 18.38 K, respectively, and the mean nonuniformity of upper and right wall temperature is reduced by about 26.47% and 28.41%, respectively. Finally, some guidelines are proposed for applications of the improved counterflow double-channel micro combustor in the micro-thermophotovoltaic system.

We investigate the effect of varying carbon nanotube (CNT) size on the desalination performance through slit confinements formed by horizontally aligned CNTs stacked on top of one another. By increasing the CNT size, the results obtained from this study indicate a corresponding increase in the water flow rate, accompanied by a slight reduction in salt rejection performance. However, due to the increase in the membrane area with CNT size, the permeability performance is observed to reduce as the CNT size increases. Nevertheless, a comparison with nanoporous 2D membranes shows that the permeability of an outer-wall CNT slit membrane remains significantly higher for all CNT sizes considered. This indicates that precise dimensions of the CNTs are not highly crucial for achieving ultra-high permeability performance in such membranes, as long as the critical slit size is maintained. In-depth analytical studies were further conducted to correlate the influence of curvature effects due to increasing CNT size on the flow characteristcis of the outer-wall CNT membrane. These include the analysis of the measured velocity profiles, oxygen density mapping, potential of mean force profile and friction profile. The present numerical results demonstrate the superb desalination performance of the outer-wall CNT slit membrane, regardless of the size of CNTs used. In addition, an extensive analysis conducted provides detailed characterization of how the curvature affects flow across outer-wall CNTs, and can be used to guide future design and fabrication for experimental testing.

Repeated vibration modes often occur in practice in structural vibration systems with general nonproportional viscous damping. However, the topic remains perhaps the least understood in vibration analysis. Some researchers have suggested that the inclusion of nonproportional viscous damping renders a system defective in the case of repeated eigenvalues, while others have assumed that a complete set of linearly independent eigenvectors can always be found regardless of the forms and magnitudes of viscous damping. This paper seeks to first establish that for light non-proportional viscous damping, a system with repeated eigenvalues generally does not become defective based on perturbation theory and realistic numerical examples since rigorous theoretical proof is believed to be difficult. Once non-defectiveness is confirmed, a method for computing the eigen derivatives with repeated eigenvalues in the case of general viscous damping is developed. Effect of mode truncation on numerical accuracy has been discussed. When the magnitude of non-proportional viscous damping becomes considerably high however, it is possible for a damped system to become defective when repeated modes occur. This can have profound practical implications where very high damping is desired as in the cases of vibration suspension and absorber designs, since system defectiveness can lead to major difficulties in the applications of conventional vibration analyses.

A novel numerical method has been developed and applied to the prediction of vibration modes of general stepped beams with arbitrary steps and general elastic supports. Each beam section is modelled as an element with input impedance at one end and output impedance at the other. These impedances are then coupled when the beam sections are joined to form an overall stepped beam structure. General elastic supports are modelled and their effects on impedance are examined. The method is theoretically exact assuming each beam section can be modelled as Euler-Bernoulli beam and is computationally very efficient since it does not involve any matrix operations of large dimension. Vibration modes of a number of stepped beams of different configurations have been computed using the proposed method and are compared with existing results in literature. Experimental investigations have also been carried out to validate the practical usefulness of the method. The method is also ideally suited for forced vibration applications since it not only establishes vibration modes, but also frequency response functions. In addition, a general purpose software has since been developed based on the method which can be very useful for structural dynamics design of general stepped beam structures.

Since the frequency response function (FRF) sensitivity method was first proposed [26], it has since become a most powerful and practical method for analytical model updating. Nevertheless, the original formulation of the FRF sensitivity method does suffer the limitation that the initial analytical model to be updated should be reasonably close to the final updated model to be sought, due the assumed mathematical first order approximation implicit to most sensitivity based methods. Convergence to correct model is not guaranteed when large modelling errors exist and blind application often leads to optimal solutions which are truly sought. This paper seeks to examine all the important numerical characteristics of the original FRF sensitivity method including frequency data selection, numerical balance and convergence performance. To further improve the applicability of the method to cases of large modelling errors, a new novel function-weighted sensitivity method is developed. The new method has shown much superior performance on convergence even in the presence of large modelling errors. Extensive numerical case studies based on a mass-spring system and a GARTEUR structure have been conducted and very encouraging results have been achieved. Effect of measurement noise has been examined and the method works reasonably well in the presence of measurement uncertainties. The new method removes the restriction of modelling error magnitude being of second order in Euclidean norm as compared with that of system matrices, thereby making it a truly general method applicable to most practical model updating problems.

Vibration modes with repeated eigenvalues often occur in engineering design and analysis practices due to geometrical symmetries of structural systems such as the cyclic symmetry of bladed disk assemblies of turbomachinery. However, the degeneration of eigenvector space and the resulting discontinuities have prevented useful applications of eigensensitivities to vibration analysis of a wide class of problems with repeated modes. To overcome such difficulties, a new concept of a global design variable is developed in which all intended multivariate design modifications are grouped into a single global design variable. Eigensensitivities have then been formulated for repeated eigenvalues from which accurate predictions of vibration characteristics can be made. Further, second order eigenvalue derivatives are also employed to further improve the accuracy of predicted vibration properties. Such newly formulated eigensensitivity analysis has been effectively applied for the first time to the prediction of mistuning effect of bladed disk assemblies. Numerical results from a realistic discrete parameter model of a bladed disk have demonstrated that not only natural frequencies and mode shapes can be predicted very accurately, but also blade vibration responses under engine order excitations. In addition, the proposed eigensensitivty analysis can predict the statistical variations of blade vibrations under random mistuning. Finite element modeling and vibration testing of a practical bladed disk structure have been carried out to demonstrate the practical potential of the proposed method to be possibly integrated into finite element analysis for structural modification predictions, especially for structures with repeated eigenvalues such as bladed disks.

In this paper, we investigate the potential issues low dimensional carbon material desalination membranes may encounter when nanoplastics are present in the systems. This study is performed via molecular dynamics simulations on desalination systems which comprise the nanoporous graphene membrane. It was found that there is high affinity between nanoplastics and graphene, indicating high tendency of the nanoplastic particles to foul the membranes. Furthermore, it was observed that the presence of nanoplastic molecules reduce the salt rejection rate of the membranes. Through this work, we hope to highlight the potential issues caused by the increasing prevalence of nanoplastics in the environment on carbon based desalination membranes, so as to aid the future design of these high performance membranes. Published version

The underlying mechanisms of surface phenomena are very complex and still not entirely clear. The aim of this work attempts to reveal the important contribution of the nonlocal integral theory of elasticity to surface elasticity, which is of fundamental scientific interest. By considering a uniform and isotropic half-space medium subjected to an arbitrary uniform strain, it is shown that the bulk is homogeneous, however, the whole medium is heterogeneous due to the bond loss near the free surface. The nonlocal effect cannot be observed at all in the homogeneous bulk, but the bond loss characterized by the nonlocal integral theory can show an important contribution to the surface elasticity. This in turn allows to propose a simplified surface model to replace the complex modeling of nonlocal integral theory. The thickness of surface zone can be evaluated from the intrinsic characteristic length used in the nonlocal integral theory. According to the intrinsic correlation and using the molecular dynamics simulations, the thickness of the surface layer of silicon films is 2.6911 A. Furthermore, with application of the simplified surface model to a thin film in tension, it has been noted that the geometric dimension of size-dependence is generally not that of traditional mechanics. For instance, in most situations, the main contribution to size-dependence has actually come from the thickness (or radius) direction of a rod-type structure, rather than its axial direction which is intuitively- and widely-used in the current practice in open literature.

Eigenvalue and eigenvector derivatives with respect to system design variables and their applications have been and continue to be one of the core issues in the design, control and identification of practical engineering systems. Many different numerical methods have been developed to compute accurately and efficiently these required derivatives from which, a wide range of successful applications have been established. This paper reviews and examines these methods of computing eigenderivatives for undamped, viscously damped, nonviscously damped, fractional and nonlinear vibration systems, as well as defective systems, for both distinct and repeated eigenvalues. The underlying mathematical relationships among these methods are discussed, together with new theoretical developments. Major important applications of eigenderivatives to finite element model updating, structural design and modification prediction, performance optimization of structures and systems, optimal control system design, damage detection and fault diagnosis, as well as turbine bladed disk vibrations are examined. Existing difficulties are identified and measures are proposed to rectify them. Various examples are given to demonstrate the key theoretical concepts and major practical applications of concern. Potential further research challenges are identified with the purpose of concentrating future research effort in the most fruitful directions. Ministry of Education (MOE) The first and third authors gratefully acknowledge the financial support from the Singapore Ministry of Education through the award of research project grant AcRF Tier 1 RG183/17.

In this paper, comparisons of the adhesive strengths of two commercially available adhesives, epoxy and cyanoacrylate, on 3D printed plastic materials, Acrylonitrile Styrene Acrylate (ASA) and Nylon 12 Carbon Fiber (NCF) were carried out. The single lap shear test is used to determine the adhesive properties of the specimens with and without post-curing at elevated temperature. A comparison is made with fully printed, non-bonded specimens to give a relative gauge of the performance of the adhesives. It was found that for ASA and NCF, the adhesive strength for cyanoacrylate (CA) is much higher than that of epoxy. ASA and NCF bonded with CA had average failure load of 1810 kN and 2310 kN, respectively, as compared to those bonded with epoxy which had significantly lower failure load of 470 kN and 860 kN, respectively. It was observed that although heat treatment and surface treatment improve the adhesive strength of epoxy with both adherend materials, the improved adhesive strength of epoxy is still observed to be significantly weaker than that of CA. Accepted version

A novel and computationally efficient iterative method is proposed for the exact eigenvalues and eigenvectors of nonviscously damped vibration systems. General nonviscous damping model is assumed in which damping forces depend on the past motion history via convolution integrals over exponentially decaying kernel functions. The presence of nonviscous damping leads to complex frequency dependent eigenvalue problem whose solution requires either extensively augmented state-space formulation or iterative full complex eigensolutions on mode by mode basis, both of which are computationally expensive when practical systems with large dimensions are considered. By simply solving the eigenvalue problem of the underlying undamped vibration system, the real eigenvalues and eigenvectors can then be combined with the nonviscous damping matrix to develop an iterative procedure from which required complex eigenvalues and eigenvectors of the damped system can be computed. First order perturbation is employed to further improve the starting estimates of the desired eigenvalues and eigenvectors and the convergence is generally very fast. The proposed method is mathematically developed based on progressive eigensensitivity analysis and convergence is ensured when the norm of the damping matrix is of second order when compared with that of the stiffness matrix. Representative numerical examples of discrete mass spring damping model as well as practical finite element model with nonviscous damping are given to demonstrate and validate the accuracy and efficiency of the proposed iterative method.

This paper investigates the orthotropic properties of Fused Deposition Modeling (FDM)-printed Acrylonitrile Styrene Acrylate (ASA) material with different raster configurations. The elastic properties were determined using a non-destructive ultrasonic technique. This technique allows us to deduce the orthotropic elastic constants from the material density and the velocities of the longitudinal and shear waves propagating through the material along different directions. Tensile tests were performed in addition to ultrasonic tests to obtain the yield properties of the ASA material and to validate the elastic constants determined by the ultrasonic tests, which have shown very close correspondence. Finally, numerical verification was performed by comparing the experimental results of the three-point and four-point bending tests with the finite element simulation results which have as input the material properties from the ultrasonic testing. The simulation results have shown excellent agreement with the experimental results, implying that the material properties obtained from the ultrasonic testing were highly accurate comparing to the actual orthotropic elastic properties of the 3D-printed ASA material. NRF (Natl Research Foundation, S’pore) Accepted version

Although single layer transverse flow carbon nanotube (CNT) membrane (TFCM) has been shown to be ultra-permeable with high salt rejection, its physical fabrication with sub-nanometre slits remains a significant challenge to its development. This work presents the multilayer TFCM, which resembles vertically aligned CNT arrays, as an alternative candidate for efficient desalination. Using molecular dynamics, this work shows that multilayer TFCM can provide permeability and salt rejection on par with its single layer counterpart. By multilayering, the slit size between neighbouring CNTs can be increased to nanometre range while still maintaining high salt rejection. The increase in slit size counteracts the reduction in permeability due to multilayering, thereby allowing multilayer TFCM to achieve permeability performance comparable to the single layer TFCM. The effects of the number of layers n and other design parameters (interlayer distance d , CNT diameter D C N T , offset h ) on the desalination performance of multilayer TFCM are investigated thoroughly using results from non-equilibrium and equilibrium molecular dynamics. It was found that the desalination performance is not sensitive to variations in d , D C N T or h . Finally, this work provides computational evidence that the multilayer TFCM, which could be fabricated using techniques for current dense vertically aligned CNT arrays, can make an efficient design for future low dimensional materials membrane.

Many different methods have been developed since the pioneering works of Mills-Curran (1988) and Dailey (1989) on eigenvector derivatives with repeated eigenvalues. In spite of the increasing mathematical complexities witnessed in many of the newly emerged methods, some underlying fundamental theories governing the eigenvector derivatives have neither been much discussed, nor fully established to date. The present approach seeks to fill such an outstanding theoretical gap and to lay down the necessary theoretical foundation on which existing methods can be mathematically unified and further improved in numerical accuracy and computational efficiency. The particular solutions of eigenvector derivatives generally required have been derived in terms of modal properties, thereby avoiding the computationally expensive and potentially erroneous procedure of solving a set of algebraic equations of system dimension. The contributions of higher unavailable modes have been theoretically derived, enhancing the practical applicability of the proposed method to the general case where only partial eigensolutions are made. To avoid degeneration of eigenvector space in the case of repeated eigenvalues, a concept of global design variable is developed in which all intended multivariate design modifications are grouped into a single global variable to which eigenvector derivatives are derived, rendering real major applications of the proposed method to the predictions of structural design modifications. A discrete parameter model of a turbine bladed disk assembly, which is known to have many pairs of repeated eigenvalues due to its cyclic symmetry, as well as a finite element model of a cantilevered beam with large DOFs have been employed. Numerical results have demonstrated the accuracy and the practical applicability of the proposed new theoretical developments, as well as the proposed new method.

The widespread use of low dimensional carbon membrane for desalination and gas separation is limited by the difficulty to physically realise such membrane designs on a meaningful scale. This review aims to bring together results achieved in this field, hoping to inspire new designs or developments that could bridge this technical challenge. The focus of this paper is on sub-nanometer separation operations such as desalination or gas separation. This is because such operations consume the most energy, and there is thus much interest to reduce this cost. Three groups of low dimensional carbon materials are considered: graphene, carbon nanotubes (CNT) and graphene oxide (GO). Graphene and CNT membranes have the advantage of high permeability but are difficult to manipulate to form membranes that separate efficiently. GO, on the other hand, has the advantage of ease of fabrication but suffers in terms of separation performance. This review dives deep into the innovative ideas proposed for these low dimensional carbon membrane design, deliberating their strengths and weaknesses, in a consolidated effort to generate new ideas for further advancements.

The use of two-dimensional (2D) materials as ultra-thin desalination membrane has garnered much interest in the scientific community. Various 2D materials membrane designs have been tested both numerically and experimentally. However, a detailed direct comparison between different 2D materials membrane is lacking. This study uses molecular dynamics (MD) to characterize and compare the desalination performance of graphene, borophene, molybdenum disulphide (MoS2) and MXene slit membranes. It was found that monoelemental graphene and borophene have the best desalination performance. They achieve 100% monovalent salt rejection with permeability 25% higher than MoS2 membranes and 87% higher than MXene membranes. Three factors are found to influence the desalination characteristics. Firstly, the hydrophobicity of the material which is found to increase salt rejection but reduce permeability. Second the material's inherent interaction parameters, and third the shape of the slits formed by atomic arrangement of the 2D material. Furthermore, although desalination performance varies across 2D materials membrane, all the membranes tested have permeability two orders of magnitude higher than current desalination membrane. All in all, this study highlights the varying characteristics of membrane made of different 2D materials and relate the material's properties with the resulting desalination performance.

In this paper, we perform molecular dynamics simulations to propose a novel bio-inspired nanopumping mechanism that is achieved through the rotation of graphene nanoribbons. Due to the rotation and interaction with water, the graphene nanoribbons undergo morphological transformation. It is shown that with appropriate geometrical and spatial parameters, the resulting morphology is twisted ribbon, which is efficient in pumping of water through a channel. This mimics the propulsive behavior of bacterial flagella through continual rotation at the base and causing morphology of the geometry into twisted ribbons, thus driving flow. It was observed that the maximum flux rate decreases upon reaching the optimal configuration even with increasing rotational speed and graphene width. This is due to the development of cavitation near the region of the nanoribbon with tip velocities approaching the speed of sound in water. The simulation shows promising results where the flux rate of the driven flow outperforms various nanopump configurations that have been reported in recent literature by more than one order.

It is well known that encryption algorithms developed based on Logistic map suffer from limited key space due to the narrow regions of system parameters which can be used, potential risk of security in the presence of numerous periodic windows within the key space, and weakness in known-plain-text attack due to the inherent correlation among the chaotic sequence used for encryption. To overcome these existing problems, this paper presents a secure image encryption algorithm based on a new highly nonlinear discrete dynamical system with ideal chaotic characteristics. Transcendental functions are introduced together with modulo operations which physically represent discontinuous time-varying nonlinearities, leading to extremely complex chaotic behavior that is highly sensitive to system parameters and initial conditions, both of which are considered as the key for the cryptosystem. Extensive numerical experiment results have revealed that the proposed image encryption algorithm offers advantages of unlimited key space and high-level security, since those problematic periodic windows are no longer present within the key space, and it is extremely robust against known-plain-text attack, since the chaotic sequence generated bears no correlation whatsoever due to the folding effect of modulo operation. The algorithm makes truly efficient yet highly secure image encryption based on chaotic systems a reality. Published version

Many modern structural systems are found to contain localized areas, often around structural joints and boundaries, where the actual dynamic behavior is far from linear. Such nonlinearities need to be properly identified so that they can be incorporated into the improved mathematical model used for design and operation. One approach to this task is to use higher-order frequency response functions (FRFs). Identification of structural nonlinearities using higher-order FRFs is currently an active and growing emerging research area and to date, much research has been conducted. However, most existing research seems to be confined to very simple nonlinear system models such as SDOF mass-spring-damper models or to the most, MDOF mass-spring chain models. For more general continuous nonlinear structures with discrete localized nonlinearities however, analytical derivation of higher-order FRFs remains unknown and as a result, identification of nonlinearities of such systems using measured higher-order FRFs becomes impossible to achieve. This missing link between higher-order FRFs and physical parameters of nonlinearities of continuous structures has to be established before any real progress can be possibly made. In this paper, a new novel method is developed which can be used to derive analytical higher-order FRFs of continuous structural systems with discrete localized nonlinearities. The method is generally applicable and theoretically exact, and it serves exactly as that missing link. Having established analytical higher-order FRFs which serve as the theoretical foundation for any subsequent identification, method of parameter identification of nonlinearity is then further developed. Important characteristics of higher-order FRFs are discussed, some of which are revealed the first time since there has never been higher-order FRFs derived from nonlinear continuous structures. Various numerical aspects on how to improve accuracy of identified nonlinear system parameters are discussed.

Multiple single-channel meso-scale combustors assembling into a multi-channel meso-scale planar combustor is a good solution for elevating the output power of the micro-thermophotovoltaic system. Consequently, how to assembly a multi-channel meso-scale planar combustor for achieving higher output power comes to be an interesting issue. In this work, four representative four-channel meso-scale planar combustors, namely, combustor A, combustor B, combustor C and combustor D are designed. Extensive numerical investigations are conducted to compare the wall temperature of the four combustors under various hydrogen mass flow rates, hydrogen/air equivalence ratios and solid materials. Two variables, namely, mean wall temperature and nonuniformity coefficient of wall temperature are defined for comparing the wall temperature of the four meso-scale combustors. It is found that the purely counterflow four-channel meso-scale planar combustor D has more uniform and higher wall temperature compared with other combustors. Furthermore, the detailed heat transfer mechanisms of four different four-channel meso-scale planar combustors are analyzed and presented. In addition, some critical values are proposed for keeping high and uniform wall temperature on the purely counterflow four-channel meso-scale planar combustor. This work offers us great reference value for the design of multi-channel meso-scale planar combustor.

ObjectiveTo evaluate the safety and efficacy of removing blood stasis (RBS) herbal medicine for the treatment of acute intracerebral haemorrhage (AICH) within a 6-hour time window.Study designA randomised, multicentre, double-blind, placebo-controlled study performed in 14 hospitals in China.Participants and interventionsPatients with AICH were randomly assigned to receive a placebo, the ICH-1 (Intracerebral Haemorrhage) formula (eight herbs, including the RBS herbs hirudo and tabanus) or the ICH-2 formula (six herbs without the RBS herbs hirudo and tabanus) within 6 hours of ICH onset.OutcomesThe primary safety outcome was the incidence of haematoma enlargement at 24 hours and at 10 days after treatment. The secondary outcome was the incidence of poor prognosis (mortality or modified Rankin Scale score ≥5) assessed at 90 days after symptom onset.ResultsA total of 324 subjects were randomised between October 2013 and May 2016: 105 patients received placebo; 108 patients received the ICH-1 formula; and 111 patients received the ICH-2 formula. The incidence of haematoma enlargement at 24 hours was 7.8% in the placebo group, 12.3% in the ICH-1 group and 7.5% in the ICH-2 group; the incidence of haematoma enlargement on day 10 was 1.1% in the placebo group, 1.1% in the ICH-1 group, and 3.1% in the ICH-2 group, with no significant differences among the groups (P>0.05). The mortality rates were 3.8% in the placebo group, 2.8% in the ICH-1 group, and 0.9% in the ICH-2 group; the incidences of poor prognosis were 7.1% in the placebo group, 6.0% in the ICH-1 group and 4.8% in the ICH-2 group at 3 months, with no significant differences among the groups (p>0.05). However, the overall frequency of treatment-emergent adverse events in the ICH-1 group (12.1%) was higher among the three groups (5.8% and 2.8%, respectively, p<0.05). All three cases of serious adverse events were in the ICH-1 group.ConclusionsUltra-early administration of ICH-1 formula for AICH patients did not exert significant beneficial effects on clinical outcomes but increased the risk of bleeding, which probably resulted from the inclusion of RBS herbal medicines in ICH-1.Trialregistration numberNCT01918722.

A nonlinear differential quadrature based method is developed and successfully applied to structural and vibration analyses of general multi-layered graphene sheets in the presence of geometrical nonlinearities and nonlinear van der Waal forces. Important nonlinear displacement characteristics of multi-layered graphene sheets have first been established. Due to relatively large magnitudes of van der Waal forces, different layers of graphene undergo similar deflections under external loadings and nonlinearities in van der Waal forces do not contribute much to the overall nonlinear structural behaviour. In the case of vibration analysis, two groups of modes each having its own unique characteristics have been identified and are defined as the lower classical bending modes and the higher van der Waal enhanced modes. For lower classical bending modes, the layers vibrate in phase and their natural frequencies are only affected by geometrical nonlinearities but decoupled from nonlinear van der Waal forces. For higher van der Waal enhanced modes however, the nonlinear vibration characteristics are dictated by the nonlinear van der Waal forces. These observations are valid for different boundary conditions and different layers of graphene sheets that have been investigated. Van der Waal forces and their effects are properly modelled and examined, together with effects of key physical parameters. The results presented for the first time provide accurate and wholesome studies on the nonlinear structural and vibration characteristics of graphene sheets. These results are important to structural designs of graphene sheets which are increasingly being deployed for innovative engineering applications such as nano-electro-mechanical systems (NEMS).

The flexural vibration of nano-meter sized single layer graphene sheets is studied using a REBO potential based atomistic structural model for a wide range of graphene sheets sizes and chiralities, and with consideration for different boundary conditions, as well as the effects of in-plane forces and environmental stiffness. The scale effects on the vibration of graphene sheets are examined and discussed. A novel nonlocal couple stress thin plate model is then proposed for single layer graphene sheet and is iteratively fitted to the earlier obtained atomistic results to obtain estimations of the scale effect parameters. In addition to the nonlocal parameter, the equivalent Poisson's ratio of the plate model of graphene sheet in flexure is also shown to be a scale effect parameter having mostly negative values. This phenomenon is discussed to corroborate the use of couple stress theory in developing the current thin plate model for graphene sheets.

Background Hypertensive intracerebral hemorrhage (HICH) is one of the most devastating forms of stroke. Currently, no specific therapies for HICH except general medical care. However, in China, medicine of promoting blood circulation (PBC) and removing blood stasis (RBS) are widely and efficiently used to treat HICH and become a potentially effective treatment for the secondary effects of HICH to alleviate brain injury, accelerate neuronal recovery, and improve the prognosis. In order to evaluate the safety and effect of PBC and RBS herbal drugs, we design a prospective, randomized, open, double-blind controlled clinical trial on the hematoma enlargement in HICH patients treating with PBC and RBS herbal medicine within 6 h time window from the symptom onset. Methods/design A multicenter, three-group, prospective, randomized, double-blinded and placebo-controlled clinical trial. Patients aged 18 or older with HICH confirmed by CT scan within 6 h from the onset are included. 360 patients will be randomized to 3 groups (PBC & RBS & Placebo) within 6 h of ictus. Stratified block randomization is undertaken using a sequentially numbered and opaque envelope. All subjects must take medicine within 6 h of ictus and have another CT scan at about 24 h to confirm hematoma expansion. A postal questionnaire to the patients to evaluate their recorvery at 3 months. Primary outcome is the percent change in the volume of hematoma at 24 h. Secondary outcomes include: mortality, disability, serious adverse events, etc. Conclusions The CRRICH Trial is expected to confirm the safety and effect of acute intracerebral hemorrhage treated within 6 h of ictus with “RBS” therapy and to determine whether the traditional therapy can cause hematoma growth after intracerebral hemorrhage. Discussions This is the first prospective, multicenter, randomized, placebo-controlled clinical trial to investigate herbal medicine whether can induce the incidence of hematoma enlargement of AICH patient within the 6 h time window from onset. We need the data to keep the herbal clinical usage safety. Trial registration clinicaltrials.gov: NCT01918722

So far, the industrial application of ultrasonic vibration assistance has been successful in continuous machining process such as turning process where ultrasonic vibration velocity is much higher than cutting velocity. Recently, vibration assistance has been experimentally investigated to the intermittent milling process mainly for feed and cross- feed directions. This paper focuses on the effect of ultrasonic vibration assistance in spindle axial direction for improvement of machined surface. With the designed ultrasonic vibration assisted milling process with 39.7 kHz and a few micro-meter amplitudes, workpiece vibrates along spindle axial direction while different RPMs and feed rates are applied. The axial directional vibration assistance acts as additional cutting motion which further reduces the leftover surface error. Experimental results validate that surface roughness can be improved from 20 % to 65 % for the tested conditions. Apparently chatter marks of the milling process are reduced with the help of the axial ultrasonic vibration assistance.

Many engineering structures exhibit frequency dependent characteristics and analyses of these structures lead to frequency dependent eigenvalue problems. This paper presents a novel perturbative iteration (PI) algorithm which can be used to effectively and efficiently solve frequency dependent eigenvalue problems of general frequency dependent systems. Mathematical formulations of the proposed method are developed and based on these formulations, a computer algorithm is devised. Extensive numerical case examples are given to demonstrate the practicality of the proposed method. When all modes are included, the method is exact and when only a subset of modes are used, very accurate results are obtained.

Detailed studies on the nanoscale vibration characteristics of multi-layered graphene sheets (MLGSs) that are embedded in an elastic medium are carried out using continuum-based modelling and Generalized Differential Quadrature (GDQ) method. Natural frequencies and their associated vibration modes of practical interest of single-layered and triple-layered graphene sheets, as well as general MLGSs that are embedded in an elastic medium are established. Numerical simulations are conducted to examine the effects of van der Waals (vdW) interactions, which are present as bonding forces between the layers, on nanoscale vibration natural frequencies and their mode shapes. The results show that for a general MLGSs embedded in an elastic medium, vibration modes can in general be classified into three families – lower classical synchronized modes which are independent of van der Waals forces and are somewhat sensitive to the surrounding elastic medium, middle van der Waals enhanced modes which are largely determined by the presence of van der Waals interactions and are hence less sensitive to the changes of the surrounding elastic medium, and higher mixed modes which are combinations of classical synchronized modes and van der Waals enhanced modes. Detailed characterizations of these modes from their derived mode shapes have been achieved for the typical case of an embedded triple-layered GSs, as well as general embedded MLGSs. Effects of Winkler modulus KW, the shear layer modulus Gb, different boundary conditions, aspect ratio β and the number L of graphene layers on nanoscale vibration properties have been examined in detail. The results presented in this paper, for the first time, provide accurate and wholesome studies and characterizations on the interesting nanoscale vibration properties of multi-layered graphene sheets embedded in an elastic medium and the results obtained will certainly be useful to those who are concerned with the dynamics of embedded graphene sheets which are increasingly being deployed for various innovative engineering applications such as nano-electro-mechanical systems (NEMS).

In the conventional use of vibration assisted machining, vibratory motion is mostly applied to the continuous machining processes such as turning where the cutting speed is much lower than the vibration speed. Even the recent articles on vibration assisted milling processes are also quite limited to low spindle speed less than 3k RPM. This study investigates vibration assistance that is applied to the workpiece in a high speed micro/meso-scale intermittent milling system where the cutting speed is much higher than the vibration speed. In addition to this, the vibration effect is analyzed considering feed and cross-feed directional application separately, which gives an idea of a right vibration assistance direction for surface quality improvement. To validate this, a one-directional ultrasonic vibration assisted milling system with ultrasonic frequency at 40 kHz and with amplitudes of a few microns is designed and its effect on the machined surface quality is investigated at high spindle RPMs over 15k. As a result, cusp heights are found to be reduced with ultrasonic vibratory motion of cutting edge in high cutting speed. Furthermore, the machined surface quality clearly tells that feed directional vibration assistance is able to generate better surface quality with reduced wavy burrs than cross-feed directional vibration assistance.

Current modal analysis methods seek to identify the modal parameters of some or all of the modes in the measured frequency range of interest. In many applications however, it will be very useful if modal parameters of some of the out-of-range modes can be identified during modal analysis. Such a goal is obviously theoretically possible since the raw measured frequency response functions (FRFs), upon which modal analysis is performed, do contain adequate information about the out-of-range modes in the form of residue contributions. In this paper, a new method for the estimation of modal parameters using multiple FRFs analysis is presented. In the process of modal identification, the proposed method not only presents accurate modal parameters of the modes which are present in the measurement frequency range, but also quite accurately identifies some of the modes which are not measured. The method calculates the required modal parameters by solving eigenvalue problem of an equivalent eigensystem derived from those measured FRF data. All measured FRFs are used simultaneously to construct the equivalent eigensystem matrices from which natural frequencies, damping loss factors and modeshape vectors of interest are solved. Since the identification problem is reduced to an eigenvalue problem of an equivalent system, natural frequencies and damping loss factors identified are consistent. Applications of the method to both numerically simulated and practically measured FRF data are given to demonstrate the practicality of the proposed method and the results have shown the method is capable of accurately identifying modal parameters of out-of-range modes.

Carbon nanotube-based composite is becoming increasingly popular and offers great potential for highly demanding practical high strength and high damping applications. The excellent damping capacity of CNTs is primarily due to the interfacial friction between carbon nanotubes and polymer resins and the extremely large interfacial surface area over a given specific mass (specific area). In this paper, damping characteristics of carbon nanotube-based composites have been investigated, with an objective of developing an effective and accurate analytical model, which can be used as a design tool for the damping design of such materials. Based on the interfacial slips between the resin and nanotubes and between the nanotubes themselves, a micro stick-slip damping model has been developed. Such a physically derived model is believed to be appropriate and representative of the actual complex damping mechanism of the material system. The model, developed for the first time, is analytical and relates explicitly the material properties of the resin and nanotubes and the processing parameters to the overall material damping loss factor and hence it offers the possibility for material engineers to possibly optimize the damping for required applications. Due to the nonlinear force–displacement relationship derived under the micro stick-slip, a harmonic linearization method, the Describing Function method, has been employed to analyse its vibration characteristics and to derive the required damping loss factors. From the analytical formula, it can be seen that the damping loss factor of the material system depends on the individual material properties of the resin and the nanotubes, structural deformation, nanotube volume fraction and the critical shear stresses at which interfacial slips take place. By taking careful considerations of these design parameters, optimized carbon nanotube-based composites for advanced damping applications can be developed. Extensive numerical simulations have been carried out to establish the practical applicability of the proposed analytical model. Based on realistic material properties of carbon nanotubes and polymer resins, damping characteristics have been predicted which compare well with existing results from open literature. The results have shown that for a volume fraction as small as 1%, a damping loss factor as high as 20% can be achieved which is adequate for most practical applications. The model has been further developed to deal with bending vibrations where different parts of the material are subject to different vibration strain levels. A practical case of cantilevered beam vibration has been employed to demonstrate the practical application of the proposed model.

One-directional ultrasonic vibration assisted milling system is designed and its performance is investigated in terms of machined surface quality under 135,000 rpm. The ultrasonic vibration generator excites the workpiece with a frequency around 40 kHz and amplitude of a few micro meters. The milling tool’s cutting speed is controlled by an air-bearing spindle system. Both feed-directional and cross-feed directional ultrasonic vibration assistance are considered in order to understand the mechanism of ultrasonic vibration assistance for surface roughness generation. A comparison is done on a milled surface which is generated with and without ultrasonic vibration assistance. The experimental results show that ultrasonic vibration assistance can improve the machined surface quality which depended on the cutting edge radius and the feed per tooth.

In engineering practice, both viscous and hysteretic damping models are usually employed to characterize damped dynamic properties of linear mechanical systems, the identification of which has long been an active area of research in experimental modal analysis. This paper is devoted to investigate the relationship between viscous and hysteretic damping models. In order to show whether the arbitrary choice of damping model is reasonable, the relationship has been derived based on two normalization procedures and the modal parameter identification procedure. In case of proportional damping, there exists an exact relationship between hysteretic and viscous damping models as η r = 2 ξ r . As for non-proportional damping, the equivalent mode shapes and their counterparts of the original system are almost the same except differing by a complex scaling factor. The numerical cases based on the simulated modal identification procedure have demonstrated that the error in the estimation of modal parameters caused by arbitrarily choosing damping model is quite small. Furthermore, the equivalent damping matrix is physically sensible in the case of the system with distributed damping whereas, no physically sensible equivalent damping matrix exists in the case where the damping is localized. The validity of the relationship between the two damping models has been further verified by a practical experimental example.

Most model updating methods use measured frequency response function (FRF) data to update analytical models whereas only response functions under base excitation can be obtained in practical vibration test due to difficulties and constraints which prevent conventional FRFs from being measured accurately. This paper presents a new model updating method, which can employ measured response function data under base excitation directly for updating. Mathematical formulations using measured response function data under base excitation to identify mass and stiffness modeling errors, have been established. Through simulated numerical case studies based on a cantilever beam as well as a practical GARTEUR structure, it has been proved that the proposed method is feasible and effective when applied to the identification of mass and stiffness modeling errors. It is also shown that this method has considerable noise-resisting ability in the case where the measured response function data are contaminated by certain level measurement noise.