Seminars 2017

Tight-binding model is a simple yet powerful method that can be used to study electronic band structures in solid-state physics. The method considers only the hopping, or the coupling, among the most adjacent atomic wavefunctions. In this talk, I will show that tight-binding model can be faithfully reproduced by using coupled acoustic resonators. Based on such a convenient platform, we are able to experimentally investigate novel physics that arises therein. Two examples will be discussed. As the first example, we consider a 4-by-4 non-Hermitian matrix, and study the complex behaviors of the eigenfrequencies, including formation and coalescence of multiple exceptional points upon the variation of system parameters [1]. Second, we construct a Su-Schrieffer-Heeger (SSH) dimerized chain using the similar acoustic cavities. Band inversion, together with a change in (quantized) Zak phase can be realized. An interface system can be constructed using two SSH chains with different Zak phases. By coupling two of such interface systems together; we obtain a new system with a five-site unit cell. For such a system, a specific separation of the 5-by-5 Hamiltonian into a topological subspace and a non-topological subspace can be found. By tuning the system parameters, the topological interface state, which exists in the topological subspace, can be embedded into a band continuum, which resides in the non-topological subspace. This gives rise to a new type of bound states in the continuum [2].

[1] ‍ Ding, K., Ma, G., Xiao, M., Zhang, Z.-Q. and Chan, C. T. Emergence, coalescence, and topological properties of multiple exceptional points and their experimental realization. Phys. Rev. X 6, 021007 (2016).
[2] ‍ Xiao, Y., Ma, G., Zhang, Z.-Q. and Chan, C. T. Topological subspace induced bound states in continuum. Phys. Rev. Lett., accepted.

The development of high-throughput techniques for gene expression profiling over the last two decades has rapidly led to the accumulation of vast amounts of datasets, which are available in public repositories. This has enabled large-scale meta-analyses of combined data to provide new biological insights, including the identification of new cancer genes. We compiled a human gene expression dataset from over 40000 microarrays. After strict quality control and data normalisation the data was quantified in an expression matrix of ~20,000 genes and ~28,000 samples, where we identified groups like normal tissues, neoplasmic tissues, cell lines and incompletely differentiated cells. Several unsupervised analyses of the data confirmed a global structure of the gene expression, which was consistent with earlier analyses, but with more details revealed due to the increased resolution. We suggested a suitable mixed-effects linear model for gene expression, which was used to further investigate it in solid tissue tumours, and to compare these with the respective healthy solid tissues. Our analysis identified 1285 genes with a systematic expression change in cancer. The list was significantly enriched with known cancer genes from large, peer-reviewed databases, whereas the remaining ones were proposed as new cancer gene candidates.

In marked difference from incompressible fluid flows, microscale gas flows commonly couple the dynamic and thermodynamic fluid descriptions, through the combination of bulk-flow evolution and external boundary conditions. Yet, while the thermal boundary conditions have a significant effect on the generated flows, traditional studies on rarefied gas systems have been limited to gas-surface interactions where the surfaces temperatures are prescribed. Such an assumption, however, seems of little practical value, as the surface temperature at any experimental setup can only be imposed indirectly through a direct prescription of the boundary heat-flux. To examine this observation, the present work demonstrates the impact of replacing an isothermal surface condition with a heat-flux condition in a variety of one-dimensional and unsteady micro-flow setups. These include gas-flow animation problems, acoustic wave propagation, and active flow control applications. Possible extensions to higher-dimension configurations, including shear-driven flows and hydrodynamic instability problems, are also reviewed.

Granular materials are considered as a complex media when elastic waves propagates in the bulk. This complexity comes from their micro-inhomogeneous character and from the highly nonlinear behavior of the contact between the particles. Compared to the classical elastic solids, one major difference is due to the importance of the rotational degrees of freedom of each particle for the description of the elastic behavior of the media. In this work, the effects of the rotational degrees of freedom are shown through theoretical predictions, numerical simulations and experimental results. Nonlinear wave propagation phenomena also happen in granular materials. Two examples are presented. The first one is the asymmetry of the nonlinear wave generation in a granular crystal submitted to gravity. The second one is the study of one single contact sphere-plane as a nonlinear resonator.