Singapore

Prediction of mechanical properties using multiscale manufacturing simulation and micromechanical models is the focus of this paper. An integrated platform is designed to cover the whole chain of analysis and perform the properties&#39; calculation and transfer them between the modules in a smooth mapping procedure. The manufacturing step is the particular problem of coupled dual-scale deformation-flow process such as the one arising in RTM, Vacuum Assisted Resin Infusion (VARI) and Vacuum Bag Only (VBO) prepregs. A finite element formulation of porous media theory framework is employed to predict the element-wise local volume fractions and the deformation of a preform in a press forming process. This formulation considers coupling effects between macro-scale preform processes and mesoscale ply processes as well as coupling effects between the solid and fluid phases. The outcome of the manufacturing simulation is the deformed mesh and element wise volume fractions to be used in micromechanical model to calculate mechanical properties. Structural performance of the &quot;deformed&quot; geometry is then evaluated in mechanical analysis. Structural performance of composites is directly dependent on its ingredient&#39;s properties, ply configurations and the manufacturing effects. These are evaluated in the integrated simulation chain and are defined as inputs. The contribution is concluded with a numerical example, where a compression-relaxation test of a planar fluid filled prepreg at globally un-drained condition is considered followed by a mechanical loading analysis. The development is user friendly and interactive and is established to enable design and optimization of composites.
This research is supported by the Singaporean Agency for Science, Technology and Research (A*STAR) under Polymer Matrix Composites Programme (PMCP) project (WP1-1) through the grant no. R265000673305.


MATERIALS FOR HUMANITY (MH 21)

Integrated simulation chains for composites mechanical performance and manufacturing effect of defects

**Please view our Welcome to MH21 video.**

<center>
<figure class="video_container">
	 <iframe src="https://screencast-o-matic.com/watch/cr1UYHV1pcz?v=6&ff=1&title=0&controls=1" width=826 height=526 frameborder="0" allowfullscreen="true"> </iframe>
</figure>
	</center>


**Walkthrough Video**

Here is a Walkthrough showing how to navigate the conference and network with others 

<center>
<figure class="video_container">
	 <iframe src="https://screencast-o-matic.com/watch/cr1vnrV1paE?v=6&ff=1&title=0&controls=1" width=826 height=526 frameborder="0" allowfullscreen="true"> </iframe>
</figure>
	</center>




	 Organized by
 
<a href="https://mrs.org.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/95d5789689d5f900b90e9fe54ea3cf0d.jpg"</a>

 
	
	 In association with
 
<a href="http://www.ntu.edu.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/334cd171ad2fdee033c635b368e5e160.png"</a>
	<a href="https://www.nus.edu.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/32e6f82c0440f7bb24c590f9f67821bd.png"</a>
		<a href="https://www.a-star.edu.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/72557865b589cfa3d2a0243ffce4a270.png"</a>
	
 


	 Supported by
 
		 <a href="https://www.visitsingapore.com/mice/en/about-us/about-secb/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/6e679842ead62be522ba6a630b82d5c9.png"</a>

To access the MH 21 Conference, [please register](https://mh21.mrs.org.sg/event/da3f9651-429f-4763-81b2-ebd7311b6bc8/websitePage:206f7605-1731-41f9-8421-56d3890f398f)

MH 21 is arranged around 6 themes that focus attention on pressing needs that should be addressed to enhance the well-being of communities and provide a forum for presenting breakthrough discoveries.

Prediction of mechanical properties using multiscale manufacturing simulation and micromechanical models is the focus of this paper. An integrated platform is designed to cover the whole chain of analysis and perform the properties' calculation and transfer them between the modules in a smooth mapping procedure. The manufacturing step is the particular problem of coupled dual-scale deformation-flow process such as the one arising in RTM, Vacuum Assisted Resin Infusion (VARI) and Vacuum Bag Only (VBO) prepregs. A finite element formulation of porous media theory framework is employed to predict the element-wise local volume fractions and the deformation of a preform in a press forming process. This formulation considers coupling effects between macro-scale preform processes and mesoscale ply processes as well as coupling effects between the solid and fluid phases. The outcome of the manufacturing simulation is the deformed mesh and element wise volume fractions to be used in micromechanical model to calculate mechanical properties. Structural performance of the "deformed" geometry is then evaluated in mechanical analysis. Structural performance of composites is directly dependent on its ingredient's properties, ply configurations and the manufacturing effects. These are evaluated in the integrated simulation chain and are defined as inputs. The contribution is concluded with a numerical example, where a compression-relaxation test of a planar fluid filled prepreg at globally un-drained condition is considered followed by a mechanical loading analysis. The development is user friendly and interactive and is established to enable design and optimization of composites.
This research is supported by the Singaporean Agency for Science, Technology and Research (A*STAR) under Polymer Matrix Composites Programme (PMCP) project (WP1-1) through the grant no. R265000673305.


technical paper

The toughness of epoxy resin has a significant influence on the performance of fiber-reinforced composites. Pure epoxy resin has poor toughness and usually needs to be toughened by adding elastomer. However, the epoxy resin is usually a liquid with very low viscosity at the mixing temperature. When the high molecular weight elastomer is directly used, the viscosity of the two materials is quite different, and it is difficult for the shear force to be transmitted from the matrix phase to the dispersion phase. On the other hand, when using a liquid rubber type with low viscosity to toughen, not only the toughening effect is not good, but also the heat resistance and strength of epoxy resin will be significantly reduced. Therefore, to prepare high-performance fiber-reinforced composite materials, we use dynamic covalent bonds to construct a reversible molecular weight polyurethane elastomer toughening agent to achieve reversible changes in the viscosity of the elastomer at different temperatures and its applying for epoxy resins toughening. The reversible bond polyurethane elastomer has low viscosity at higher temperature and is easy to disperse in epoxy resin; after the temperature decreases, the toughening agent returns to high molecular weight elastomer. The phase separation of the composite material system can be controlled on a larger scale by changing the temperature. This technology is expected to be further promoted and applied in multiphase composite material systems.

Adaptable elastomer and its application in toughening epoxy resin based on dynamic bonds

Applications of fiber-reinforced composites in engineering structures sometimes require joining with other materials such as metal components or other composites. Various joining methods such as adhesive bonding and bolting can be adopted depending on the required loading conditions, types of materials, surface conditions, and environmental factors. This study aims to model and analyze mechanical joints between carbon fiber reinforced thermoplastic composite plates with metallic materials under out-of-plane load. Conical metallic inserts are used to transfer the load to the composite plate. The influence of the insert geometry and the adhesion parameters to the strength of the joint were studied. It should be noted that in reality the adhesion parameters, which include the strength and toughness of the interface between the composite surface and the metallic insert, can potentially be modified by using various surface treatments to the metallic insert. The results show that both the geometry of the insert and the adhesion parameters influences the strength of the joint, indicating that the mechanism of load transfer is not only through mechanical locking between the two but also through their adhesion.

Analysis of hybrid joints between carbon fiber reinforced thermoplastic composite plates and metallic inserts

Coherent interaction of light beams in the presence of a thin absorber allows to operate between the regimes of total light absorption and total light transmission. It was demonstrated that such absorbers (coherent perfect absorber, CPAs) provide nonlinear input-output signal dependencies and enables light-by-light control without requiring intrinsic nonlinearity. All-optical logical gates, signal amplification and dark pulse generation were demonstrated with CPAs and classical (laser) light. Beyond this, recent experimental studies showed that CPAs enable manipulation of quantum states of light including single photon and NOON states. 
Here, we extend quantum regimes of CPAs by interfering bosonic and fermionic states of entangled photons. In a single experiment we compare interference of bosonic and fermionic states in the absence and presence of dissipation. Lossless interference results in the Hong-Ou-Mandel effect where bosons coalesce and fermions anti-coalesce. In contrast, lossy interference with CPAs has the opposite outcome with anti-coalescent bosons and ‘coalescent’ fermions. We provide analysis of this ‘reversed’– anti-Hong-Ou-Mandel effect, and show that it originates from the phenomena of probabilistic two-photon absorption for bosonic pairs and deterministic one-photon absorption for fermionic pairs. The ability to manipulate and control the behaviour of photons and their output statistics with CPAs offers novel approaches for quantum information and metrology applications.

Anti-Hong-Ou-Mandel effect with coherent perfect absorbers

Silicon photonics is an enabling technology which has facilitated the drive towards continuous miniaturization of photonic integrated systems, by providing low cost and highly scalable solutions leading to faster and more compact devices. This has led to an increasing modal size mismatch between nanoscale on chip waveguides and standard optical fibers used in data communications today. As a result, one of the key challenges faced by current telecom industry has been to efficiently couple light between waveguides and external light sources such as fibres and free space optics, without additional fabrication post-processing and complex optical alignment. Among various coupling platforms, in-plane grating couplers (GC) have been widely preferred, mainly because of low fabrication costs, ease of alignment and high level of flexibility in circuit design. 
However, these couplers are passive photonic components, thus their coupling efficiency (CE) is fixed once the device is fabricated. Incorporating a switching functionality into an efficient GC design where the CE can be dynamically tuned, would allow for a multifunctional component enhancing its potential for a diverse reconfigurable range of applications as well as allowing post-fabrication trimming of silicon photonic circuits. 
Using inverse design techniques, we demonstrate CMOS compatible, reconfigurable phase change chalcogenide-on-insulator based apodised GC having maximized CE of more than 50% at λ=1580nm when the phase of the chalcogenide is in amorphous state. When the phase is switched to crystalline state, a near zero CE is shown allowing the design to be reversibly reconfigurable with highest transmission modulation contrast of more than 50db. 
We further demonstrate that the centre frequency for which the GC design has been optimized for maximum CE, can be shifted by varying geometrical parameters such as etch depth and thickness to tune the operational window between E to L telecom bands.

Inverse design of reconfigurable phase change chalcogenide grating couplers with ultrahigh modulation contrast

Label-free optical biosensors using nanotechnologies have made substantial progress in the last few years due to their ability for fast, compact, and cost-effective chemical detection devices [1]. High refractive index dielectric nanoresonators [2] have been found as promised candidates for the miniaturization of sensitive bio-sensing devices thanks to their low dissipative losses [3]. Integrated on-chip, such sensors allow instant monitoring of the concentration of target analytes, which becomes a necessary tool for many life sciences fields [4]. In this work, silicon nanoantennas arranged in a one- dimensional array are fabricated to produce a high-quality factor resonator (7500 measured experimentally) with high-sensitivity to optical modulation [5], allowing precise refractive index sensing of the surrounding medium. Here, the measurements for several glycerin dilutions in aqueous solutions show a high sensitivity of 265 nm/RIU and the highest figure of merit of 1400 RIU-1. With a small sensing area of 0.5×18.2 μm2, we believe this sensor design is an attractive candidate to realize lab- on-chip sensor arrays for label-free multiplexing in a cost-effective way due to the CMOS photolithography fabrication compatibility.

References
1. E. Luan, H. Shoman, D. Ratner, K. Cheung and L. Chrostowski, Sensors, 2018, 18, 3519.
2. A. I. Kuznetsov, A. E. Miroshnichenko, Y. H. Fu, J. Zhang and B. Luk’Yanchuk, Scientific
reports, 2012, 2, 492.
3. O. Yavas, M. Svedendahl and R. Quidant, ACS nano, 2019, 13, 4582-4588.
4. G. P. Hancke and G. P. Hancke Jr, Sensors, 2013, 13, 393-425.
5. L. Ding, D. Morits, R. Bakker, S. Li, D. Eschimese, S. Zhu, Y. F. Yu, R. Paniagua-
Dominguez and A. I. Kuznetsov, ACS Photonics, 2020, 7, 1001-1008.

One-dimensional high Q nanoparticle chain resonator for refractive index sensing

Optical metasurfaces have emerged as a new class of optical elements superior to the conventional ones owing to their unique ability to tailor the optical wavefront on a subwavelength scale. High efficiency, compactness and lightweight make metasurfaces a strong candidate for on-chip integrated optical devices, a core constituent of next-generation augmented reality displays, light ranging and detection (LiDAR) and optical computing. Here, utilizing the concept of Huygens metasurfaces modulated by liquid crystals (LCs), we demonstrate an all-solid-state tunable transmissive spatial light modulator (SLM) prototype. In contrast to conventional LC SLM, our device features reduced interpixel crosstalk, enabling small pixel size of ~1 um. By programming individually addressable electrodes, we demonstrate reconfigurable beam steering with high Side Mode Suppression Ratio (SMSR) of ~6 dB and transmission efficiencies >20 %. Combining the device with a two-photon polymerization assisted 3D printed doublet microlens, we expand the viewing angle up to 78º. The demonstrated characteristics make the proposed technology very promising for a variety of applications, such as tunable imaging, dynamic holography and LiDAR.

Large field-of-view all-solid-state transmissive metasurface spatial light modulator.

Inverse design of photonic devices [1] (especially nanophotonics [2]) is challenging due to the large set of geometries andmaterial choices accessible to nanofabrication; Deep Learning is currently being explored [2] for these problems. Previous reportshave exclusively used traditional feed-forward networks with some papers using convolutional layers. With complicated datasets,increasing the number of layers is not effective as vanishing gradient makes it harder to train; resulting saturated or degradedpredictability. Residual learning (ResNet) [4] addresses this issue by identity shortcuts that propagate the gradient by skipping oneor more layers (Fig-1D-i). In DenseNet(Fig-1D-ii,iii), identity mappings are further exploited [5] by directly connecting all layers;that not only alleviates the vanishing-gradient but also strengthens feature propagation by reusing them. Our study explores thesedense network architectures for adequate global predictability to design 2-D extended unit-cell dielectric metagratings [3]. Eachextended unit cell consists of two non-overlapping elliptical nanoresonators whose positions, axes lengths and orientations are freeparameters (Fig-1A) and can lead to an output 3D tensor representing wavelengths, diffraction orders, and polarizations(Fig-1C).Fig-1Bshows the sample network designed for this learning; where the deconvolution layer with sigmoid activation works as thefinal layer. Geometrical parameters, bound by corresponding lower and upper limits, are uniformly sampled. After eliminatingstructures with overlapping ellipses and lattice overshoot, total 54980 feasible structures approximate the global mapping to belearned. Their diffraction spectra were calculated usingS4RCWA solver. Throughout the study, these data were considered fortraining and testing. As shown in Fig-1E, when trained with the complete dataset, all three network architectures were observed tobe giving better accuracies as compared to feed-forward network. By increasing the number of hidden layers, global predictabilitywas also improved.

Densely Connected Neural Networks for Nanophotonic StructureDiscovery

Optical fibers are the main photonic technology platform that serve as the crucial high-speed connections between network nodes and servers across large distances; however, current electro-optic signal conversions present an energy intensive, bandwidth and data rate limited bottleneck. To realized next-generation communication transfer speeds (beyond 5G), a change in hardware architecture is required: the implementation of an all-optical network where signal generation, processing and routing is brought into the optical domain. Recently, planar phase change chalcogenide metamaterials have demonstrated all-optical switching, capable of intensity and phase modulation with built-in memory functionality (non-volatility) at a fraction of the operation wavelength in size[1-5]; thereby presenting an energy efficient, reconfigurable, and small footprint platform for integration onto optical fibers to functionalize this traditionally passive light guiding component. Here we present metadevices made from reconfigurable phase change chalcogenide (specifically Ge2Sb2Te5 composition) metamaterials on both optical fiber-tip and side-polished configurations. Integrated on a commercial single mode fiber tip, the cuboid GST metasurface exhibits large transmission contrasts between amorphous and crystalline phases and both normal and anomalous dispersion profiles ranging across the telecom bands for applications in signal broadening compensation. The metasurface periodicity enables spectral position control of dielectric Mie resonances, allowing for tunable operation across the telecom band. We also demonstrate that side-polished fiber integrated metadevices exhibit two types of resonances: one arising from thin film Fabry-Perot modes and another originating from metamaterial grating modes. By varying the period and thickness of the nanograting, the two resonant modes can be spectrally shifted to address applications in multi-band filtering and wavelength division multiplexing.

Fiber integrated chalcogenide metasurface enabled phase and intensity modulation with built-in memory functionality

Multispectral Liquid Crystal Spatial Light Modulator based on Fabry-Perot Cavity Resonances

Conductance state of a single atom, which exclusively affects the charge transferring rate, plays a key role in determining the electronic properties of the single atom/molecule devices [1, 2]. Therefore, it is significantly important to precisely characterize and deliberately switch the conductance states of single atoms/molecules. In this study, we systematically investigated the dependence of the oxygen adatom (Oad) conductance switching on its charge states by noncontact atomic force microscopy (nc-AFM) and scanning tunneling microscopy (STM) at 78 K. We firstly clarify the Oad charge and conductance states on rutile TiO2(110) using simultaneous nc-AFM/STM measurement (Figure 1). It suggests that the Oad- has much higher conductance than Oad2-, and the Oad conductance switching between high and low levels can be controlled by deliberately manipulating the Oad charge states through ramping the bias voltage. The charge and conductance states switching of the Oad can be attributed to the tunneling electrons transferring from the tip into the Oad or from the Oad to the substrate polaron [3,4,5]. Our study provides a good reference for the systematic investigation of the atomic conductance, and has a potential overriding effect on revolutionizing the investigation of the single atom/molecule devices.

Premium content

Integrated simulation chains for composites mechanical performance and manufacturing effect of defects

Downloads

Next from MATERIALS FOR HUMANITY (MH 21)

Adaptable elastomer and its application in toughening epoxy resin based on dynamic bonds

Similar lecture

Advances in Recycling of Thermoplastics and Fiber-Reinforced Thermosetting Composites: Overview of West Virginia Approaches