Quantifying similarity of pore-geometry in nanoporous materials

Lee, Yongjin; Barthel, Senja; Dłotko, Paweł; Moosavi, Seyed Mohamad; Hess, Kathryn; Smit, Berend

doi:10.1038/ncomms15396

Cited by 160 publications

(145 citation statements)

References 27 publications

(39 reference statements)

Supporting

Mentioning

145

Contrasting

Order By: Relevance

“…Applications are not limited to atomic systems and can be applied to nanoparticle aggregates and even larger macroscopic porous media. This is an ideal tool to integrate into advanced computational workflows, such as creating fingerprints for topological data analysis or machine learning [59], since the porosity profile for each configuration is unique.…”

Section: Discussionmentioning

confidence: 99%

PorosityPlus: characterisation of defective, nanoporous and amorphous materials

et al. 2018

View full text Add to dashboard Cite

As both porous and amorphous semiconductors have different advantages the challenge becomes knowing how to select one over the other, and knowing how to anticipate the degree of crystallinity or the fraction of voids as a function of a controllable feature such as the density. These sorts of relationships can be modelled computationally but unambiguous characterisation of the porosity of complex and disordered structures requires specialist tools. In this paper we demonstrate the use of PorosityPlus to investigate porosity in the vacancy induced amorphisation of defective crystals of carbon, silicon and germanium. The PorosityPlus software allows for the identification of vacancies, twin vacancies and larger pores, along with their relative locations and their respective populations. Void migration and coalescence along with the associated density changes can also be calculated. We show that, with increasing initial vacancies (reduced density) carbon (diamond), silicon and germanium exhibit characteristic density-dependent porosity profiles, coupled with simultaneous amorphisation. This is an ideal tool for integration into advanced computational workflows, such as creating fingerprints for topological data analysis or machine learning, since the porosity profile for each configuration is unique.Although one often thinks of zeolites or metal organic frameworks when nanoporous materials are mentioned [1], pores in crystalline semiconductors can be produced in a wide range of geometries, giving rise to properties quite distinct from bulk materials and metamaterials. These materials have attracted considerable and increasing interest, lead by porous carbon materials that have been the topic of research for many decades [2]. Porous carbon can be obtained via mechanical rolling of thermally expanded graphite flakes, chemical vapour deposition and vacuum filtration of dispersions of graphene sheets or carbon nanotubes, or pyrolysis of thermosetting polymer precursors [3][4][5][6][7]. Such materials exhibit unique, tuneable physicochemical properties, and hold great promise in the fields of catalysis, water treatment, biofiltration, gas separation, fuel cells and optoelectronics due to their structural integrity, continuity and purity [8][9][10][11][12][13].Although it does not have the long history of porous carbon, porous silicon also exhibits unique properties suitable for a range of applications. Porous silicon can be obtained a variety of ways, including anodization [14], stain etching with acids [15, 16] and direct chemical synthesis from silicon tetrachloride using self-forming salt byproducts which are later removed [17]. Porous silicon has been used in sensing, tissue engineering, medical therapeutics and diagnostics, photovoltaics, rechargeable batteries, energetic materials, photonics, and micro electro mechanical systems [18]. Porous silicon was also found in 1995 to be either resorbable, bioactive or bioinert depending on the porosity, making it unique among the porous semiconductors [19].In contrast, porou...

show abstract

Section: Discussionmentioning

confidence: 99%

PorosityPlus: characterisation of defective, nanoporous and amorphous materials

et al. 2018

View full text Add to dashboard Cite

show abstract

“…A number of computational topology tools has been successfully applied to the multidimensional analysis of complex networks [6]. For instance, persistent homology [14] has been employed across fields, such as contagion maps [68] and materials science [69]. In neuroscience, it has also yielded quite impactful results [31,33,37,57,[70][71][72].…”

Section: Discussionmentioning

confidence: 99%

Topological phase transitions in functional brain networks

Santos

Raposo

Coutinho‐Filho

et al. 2018

Preprint

View full text Add to dashboard Cite

Functional brain networks are often constructed by quantifying correlations among brain regions.Their topological structure includes nodes, edges, triangles and even higher-dimensional objects.Topological data analysis (TDA) is the emerging framework to process datasets under this perspective. In parallel, topology has proven essential for understanding fundamental questions in physics. Here we report the discovery of topological phase transitions in functional brain networks by merging concepts from TDA, topology, geometry, physics, and network theory. We show that topological phase transitions occur when the Euler entropy has a singularity, which remarkably coincides with the emergence of multidimensional topological holes in the brain network. Our results suggest that a major alteration in the pattern of brain correlations can modify the signature of such transitions, and may point to suboptimal brain functioning. Due to the universal character of phase transitions and noise robustness of TDA, our findings open perspectives towards establishing reliable topological and geometrical biomarkers of individual and group differences in functional brain network organization. 1 2 I. INTRODUCTIONTopology aims to describe global properties of a system that are preserved under continuous deformations and are independent of specific coordinates, while differential geometry is usually associated with the system's local features [1]. As they relate to the fundamental understanding of how the world around us is intrinsically structured, topology and differential geometry have had a great impact on physics [2, 3], materials science [4], biology [5], and complex systems [6], to name a few. Importantly, topology has provided [7-11] strong arguments towards associating phase transitions with major topological changes in the configuration space of some model systems in theoretical physics. More recently, topology has also started to play a relevant role in describing global properties of real world, data-driven systems [12]. This emergent research field is called topological data analysis (TDA) [13,14].In parallel to these conceptual and theoretical advances, the topology of complex networks and their dynamics have become an important field in their own right [15,16]. The diversity of such networks ranges from the internet to climate dynamics, genomic, brain and social networks [15,16]. Many of these networks are based upon intrinsic correlations or similarities relations among their constituent parts. For instance, functional brain networks are often constructed by quantifying correlations between time series of activity recorded from different brain regions in an atlas spanning the entire brain [16].Here we report the discovery of topological phase transitions in functional brain networks.We merge concepts of TDA, topology, geometry, physics, and high-dimensional network theory to describe the topological evolution of complex networks, such as the brain networks, as function of their intrinsic correlation level. We consider tha...

show abstract

“…Indeed, the development of complex materials descriptors that include both chemical and geometric features 134 appears to improve the overall success of machine learning models. Notably, however, the model in Ref 138 .…”

Section: H1 Data Mining Approachesmentioning

confidence: 99%

Computational development of the nanoporous materials genome

2017

Self Cite

View full text Add to dashboard Cite

H1 Abstract.There is currently a large push towards big data and data mining in materials research to accelerate discovery. Zeolites, metal-organic frameworks (MOFs) and other related crystalline porous materials are not immune to this recent phenomenon, as evidenced by the proliferation of porous structure databases and computational gas adsorption screening studies over the past decade. The strive to identify the best materials for a variety of gas separation and storage applications has not only led to collections of thousands of synthesised structures, but the development of hypothetical material building algorithms.The materials databases assembled with these algorithms expand greatly on the range of complex pore structures that have been synthesised, with the rationale that we have discovered only a small fraction of realisable structures and expanding upon these will accelerate rational design. In this review, we highlight some of the methods developed to build these databases, and some of the important outcomes resulting from large-scale computational screening efforts. SummaryIn the field of nanoporous materials discovery, we are witnessing the emergence of big data analytics combined with traditional computational thermodynamics calculations. This review turns a critical eye on the current state of the art, with a focus on computational database generation and results from large-scale screening for gas separations. H1 Introduction.The discovery, in the late 19 th century, that zeolites could trap interesting and valuable particles in their pores opened up an entire field of research 1,2 . This seemingly simple phenomenon was an enormous boon to the oil and gas industry, seeing as how cheap, but effective porous materials could serve as a catalyst for hydrocarbon cracking. From their humble beginnings (being carved out of rock faces), zeolites, and their ability to selectively trap guests in their pores, have become integral in not only the oil and gas industry, but in detergents (as ion exchangers) and natural gas purification to name a few 1 .Zeolites have since enjoyed an almost exclusive dominance in porous materials research until the late 20 th century, when we began to observe the creation of more diverse materials in terms of chemistry, network connectivity, and physical properties.At present, there are a little over 200 known zeolite structures, which is only a small fraction of the total number of structures that have been predicted 3 . In addition, if we were also able to expand the chemical diversity of these materials beyond the conventional Si 4+ , O 2-, and Al 3+ ions found in zeolites, one could envision designing materials for virtually any gas separation application. Thus when the first articles arose in the 1990's characterising Metal-Organic Frameworks (MOFs) [4][5][6][7][8] , and later Covalent Organic Frameworks (COFs) 9,10 , Zeolitic Imidazolate Frameworks (ZIFs) 11 , and Porous Polymer Networks (PPNs) 12 the excitement was palpable. It was recognised early-on by Yaghi, O'Keeffe, ...

show abstract

Quantifying similarity of pore-geometry in nanoporous materials

Cited by 160 publications

References 27 publications

PorosityPlus: characterisation of defective, nanoporous and amorphous materials

PorosityPlus: characterisation of defective, nanoporous and amorphous materials

Topological phase transitions in functional brain networks

Computational development of the nanoporous materials genome

Contact Info

Product

Resources

About