|Current Workshop | Overview||Back | Home | Search ||
Nanothermodynamics: For Equilibrium and Non-Equilibrium
Ralph V. Chamberlin - The big world of nanothermodynamics
Small-system thermodynamics was originally developed by Terrell Hill to describe the properties of isolated nanoparticles and individual molecules. Now we find that this “nanothermodynamics” is also crucial for characterizing nanometer-sized fluctuations inside bulk materials. A key result is that if these fluctuations are to obey the second law of thermodynamics, a nonlinear correction must be included in Boltzmann’s factor. I will show that this correction improves agreement between standard models and the measured behavior of liquids, glasses, polymers, and crystals. Indeed, the nonlinear correction provides a common explanation for several empirical formulas including stretched-exponential relaxation, super-Arrhenius activation, non-classical critical scaling, and 1/f noise. I will conclude with some thoughts regarding the life and legacy of Terrell Hill.
Anatoly B. Kolomeisky - How to Understand Molecular Transport through Channels: The Role of Interactions
The motion of molecules across channels and pores is critically important for understanding mechanisms of many biological, chemical and physical processes. Here we investigate the role of different types of interactions in the molecular transport through pores by analyzing exactly solvable discrete stochastic models. According to this approach, the channel transport is a non-equilibrium process that can be viewed as a set of chemical transitions between discrete spatially separated states. It provides a full dynamic description of the translocation process. It is shown that the strength and the spatial distribution of molecule/channel interactions can strongly modify the particle currents. Our analysis indicates that the most optimal transport is achieved when the binding sites are near the entrance or near the exit of the pore depending on the sign of interaction potential. These observations agree with single-molecule experiments on translocation of polypeptides through biological channels. We also hypothesize that intermolecular interactions during the channel transport might also significantly influence the translocation dynamics. Our explicit calculations show that the increase in the flux can be observed for some optimal interaction strengths. The relevance of these results for biological systems is discussed. The physical-chemical mechanisms of these phenomena are discussed from the microscopic point of view.
Lionel Mercury - (Meta)Stability of water in micrometric and nanometric geometries: Is there a size effect?
The reducing geometries is very often evoked to explain experimental shifts of water properties occluded inside small cavities. At nanometric level, the deviation of water properties can be modelled and measured and both results agree that from 5 nm (10 nm at max.) to smaller confinement, the finite size effects play a growing role to control the water thermodynamics. In the meantime, collecting large-scale evidences, like the phase transitions and chemical saturations states observed/measured in natural aquifers, the threshold between macroscopic and nanoscopic domains appears much more unclear, extending to 0.1 μm or more. The question thus arises as for the existence of intermediate mesoscopic behaviors, between truly nanoscopic and truly macroscopic domains.
Recently, we accumulated a large number of measurements (vibrational and X spectroscopies, phase transitions) on water occluded in different types of cavities (closed cavities in quartz = fluid inclusions, open nano-channels patterned inside silicon wafers, pores membranes, capillaries), and sometimes on the container. These measurements tend to demonstrate that the properties of the water body can change as a function of the distance to interfaces (over 1-3 micrometers !) and as a function of the confinement degree from around 35-40 nm.
These measurements open the discussion to the surface-to-volume competition in thermodynamics, especially when addressing two-phases (or more) systems. Thermodynamics of small systems appears attractive to deal with the evident size effect in confinement regime, though it is visible at sizes larger than usually defined. Another challenge is to thermodynamically describe biphasic frontiers using surfacial formalism, possibly requiring, in certain cases, the definition of an interfacial phase with finite width, instead of an abrupt interface between two bulk phases. In both cases, the characteristic threshold between bulk and “others” behaviors should be calculated and generalized on thermodynamic grounds.
Hong Qian - Hill's nanothermodynamics and stochastic thermodynamics
Two of the major contributions of T.L. Hill follow nicely Gibbs' two achievements: macroscopic chemical thermodynamics and statistical mechanics of fluctuating ensembles. The latter has been significantly superseded by the current theory of stochastic thermodynamics based on the notion of entropy production along a trajectory, first noted by Hill and his coworker Y.-D. Chen in 1975! Stochastic thermodynamics has found numerous applications in nonequilibrium chemistry of living cells. I shall give a brief overview of this in the first half of my talk. A unification of nanothermodynamics and stochastic thermodynamics, thus, is in order.
Surpringly, there is a problem to establish a statistical foundation for Hill's theory, which requires the Gibbs potential function G = E-TS+PV. We show that without the assumption of macroscopic limit, algebraic Legendre transform and integral Legendre transform do not agree perfectly. Possible resolution to this paradox will be discussed.
Guillaume Galliero - From local transport properties to shear induced swelling in nanopores
During this workshop we will give some examples of the impact of nanoporous slit on the dynamic behaviour of confined fluids.
Using molecular dynamics simulation on simple systems, we will show how the local variations of the transport properties of the confined fluid (diffusion , viscosity ), induced by density inhomogeneities, may be described by a simple local average density model combined with a new weight function . In addition, we will discuss the couplings that occur between shear and swelling at the pore scale because of the fluid confinement .
1. H. Hoang and G. Galliero, J. Chem. Phys. 136, art n° 184702 (2012).
2. H. Hoang and G. Galliero, Phys. Rev. E. 86, art n° 021202 (2012).
3. H. Hoang and G. Galliero, J. Phys. Cond. Matter 25, art n° 485001 (2013).
4. H. Hoang and G. Galliero, Poromechnaics V, pp. 685-692 (2013).
Øivind Wilhelmsen - The Curvature Dependence of Heat and Mass Transfer through Interfaces
Heat and mass transfer through interfaces is central in nucleation, nanotechnology and many other fields of research. Nanotechnology is rapidly developing, and demands knowledge of heat and mass transfer at the nanoscale.[1-5]
We present a method to obtain the curvature dependence of interface resistivities, such as the Kapitza resistance. The method combines the power of nonequilibrium molecular dynamics with the flexibility of density functional theory. We use the method to, for the first time, obtain the terms in the curvature expansion of the interface resistivities up to second order.
Mass and heat transfer to bubbles and droplets occurs very differently. Fluctuations in the shape of the bubble/droplet can enhance transport across the interface, leading to a highly heterogeneous initial growth-phase in nucleation processes.
1] Keblinski P., Phillpot S.R., Choi S.U.S., Eastman J.A., Int. J. of Heat
and Mass Transfer, 2002, 45, 855.
Sofia Calero - Molecular Simulations in Porous Materials: Environmental and Technological Applications
With the increasing demand for efficient,
environmentally friendly, and energy saving procedures, porous materials with
tailored structures and tuneable surface properties
are of wide spread use and this is only expected to intensify in the future.
Knowledge of the adsorption and diffusion behaviour
as a function of molecular composition and morphology is essential for an
informed choice of material for a given application. Molecular Simulations are
ideally suited to identify improved structures for adsorption applications and
to guide the design of new materials. Efficient simulation methodology and
accurate force field potentials are essential to reach this goal. Much progress
has been made in the search of fast and efficient simulation and screening
methods, but the development of good, transferable force fields, and accurate
models for flexible structures is still a formidable challenge. We perform
molecular simulations to study the adsorption and diffusion behaviour
of a variety of molecules on metal-organic frameworks. These are a relative new
class of porous materials with potential applications in molecular adsorption,
gas separation, and storage/release applications due to their good stability,
large void volumes, and well-defined cavities of uniform size. As an example we
can study structural deformation upon adsorption, identify the preferential
adsorption sites as a function of pressure, temperature and humidity, study the
effect of functionalization of open metal site find non accessible places by
comparing our results with experimental data, analyse
molecular diffusion inside the pores, or study the framework composition and
the mechanisms behind the instability of MOFs in humid environments.
Cees van Rijn - Liquid Threads: a case study of non-equilibrium fluid mechanics
Sondre Schnell - Small Systems Thermodynamics to Determine Kirkwood-Buff Coefficients From Equilibrium Molecular Dynamics Simulations
Elif Genceli Guner - Non-Equilibrium Thermodynamics of Water-Ice Interface
To understand the crystallization phenomena better, this work aims to investigate the crystal growth from an aqueous solution on a cooled surface theoretically and experimentally in terms of irreversible thermodynamics aspect for (i) estimating the magnitude of the temperature jump at the crystal - liquid interface, (ii) quantifying the effect of the coupling of heat and mass transport and (iii) showing the error one may do using this assumption.
Thus, ice formation from pure water is taken as an example case. A quartz growth cell is designed and constructed for this crystallization experiment. The cell underneath surface is cooled via a heat exchanger. Data is collected during crystallization on the heat exchanger surface. Thermo Liquid Crystal is used to detect the temperature profile at a micro scale both in the solution and under the ice crystals.
To define the fluxes and forces of the system, the excess entropy production rate for heat and mass transport into, out of and across the interface (between ice surface and liquid) is used. The method describes the interface as a separate (two-dimensional) phase in local equilibrium. Coupled heat and mass flux equations from non-equilibrium thermodynamics are defined for crystal growth and solved taking the advantage of the Onsager equations. A temperature jump at the ice-water interfaces is detected up to 1.7 °C. This observation is the experimental proof for the existence of an interfacial temperature jump during liquid-solid phase transition.Coupling coefficients and all transfer resistivities at the interface of a growing ice on a cooled surface is defined.
This knowledge can improve film- or fugacity models for the interface, change current modelling of phase transitions and eventually help the prevention of crystal growth at unwanted locations (such as scale layer formation on heat exchanger surfaces) in process industry saving considerable capital and operational costs.
Non-equilibrium molecular simulations offer a powerful tool to interrogate the non-equilibrium response of materials at the nanoscale using a microscopic approach. It provides a route to visualize transient processes and uncover novel non-equilibrium behaviour, hence complementing the predicting power of theoretical approaches and providing a powerful tool to rationalize experimental observations
I will discuss recent work performed in our lab that focuses on computational studies of heat and mass transport in nanomaterials, manipulation of nanoparticles with external fields and the investigation of tunable friction in nanopores.
Thuat Trinh - Thermodynamics of CO2 adsorbed on a Graphite Surface from Molecular Dynamics Simulation and Small System Method
A method to calculate thermodynamic properties of macroscopic systems by extrapolating properties of systems of molecular dimensions was recently developed (Small System Method)1 Here we apply the method for a graphite surface system with CO2 gas adsorbed2,3 in two distinct layers (Figure 1). From Equilibrium Molecular Dynamics simulation data, we predicted the chemical potential and activity coefficient of the surface in various temperature and pressure conditions.3 We will present a recent implementation of the method into a popular simulation package LAMMPS. An extension to experimental data of CO2 and CH4 adsorbed on activated carbon material will be discussed.
Markus Hutter - Effect of nanoscopic confinement on the viscoelasticity of filled elastomers
This paper deals with the viscoelastic effects in hard-particle filled elastomer nanocomposites. Since the fillers are hard and the matrix is purely elastic, the emergence of rate-dependent behavior (viscous-like) is of particular interest. It is discussed that this behavior originates from the immobilization of network chain-segments close to the filler particles, which effectively results in a locally increased glass-transition temperature. We discuss how such aspects can be built into a concurrent two-scale model for the macroscopic viscoelastic behavior of such nanocomposites, using nonequilibrium thermo-dynamics.
Jan van Eijkel - Nanothermodynamics in nanochannels, sensors and actuators: experiments and interpretation
The first part of this talk concerns acid/base chemistry in small confinements.
It is standard textbook theory to predict the response of a potentiometric sensor to its potential determining ion by a thermodynamic description of the solid/solution interface. Such a description can also be used to predict the surface potential of an oxide to the solution pH. In this talk we will further expand the use of this theory to explain transient pH changes observed in glass nanochannels filling with aqueous solutions. We will subsequently show how this knowledge can be used to make a electrochemical proton actuator.
The second part of this talk concerns water behavior in small confinement.
Here we will show how thermodynamics can be used to qualitatively predict the surprisingly high drying speed of nanochannels
David Reguera - Nanothermodynamics of nucleation
Phase transitions such as condensation, cavitation, or crystallization are omnipresent phenomena in nature. The mechanism triggering most of them is the formation of a rare embryo or nucleus of the new phase by random fluctuations.
This initiating mechanism is known as nucleation and its understanding and control is of utmost interest in a wide range of scientific and technological disciplines. However, despite intense investigations for more than a century, nucleation is still an unsolved problem. One of the main reasons for that is the fact that these nuclei are non-equilibrium entities consisting on just a few molecules, whose properties are difficult to characterize and predict.
In this talk, we will discuss our investigations on the nucleation of nanodroplets and nanobubbles, combining theory with simulations. We will see how, remarkably, the properties of these nanosized entities can be precisely described in the context of nanothermodynamics, and how the results of these studies provide useful insight to control and predict accurately the occurrence of this important phenomenon.
Ivan Latella - Hill's method for systems with long-range interactions
Systems with long-range interactions are characterized by an interaction potential whose range is of the order of the size of the system. The energy due to interactions between different parts of the system cannot be neglected if these interactions are long-ranged, causing the system to be intrinsically nonadditive. Although the system may contain a large number of particles, since the range of the interactions is of the order of its size, the system itself can be regarded as small, in a thermodynamic sense. Thus, the nonadditivity in these systems persists even in the limit of large number of particles.
Furthermore, the nonadditive character, manifested in the nonextensivity of thermodynamic quantities, is also a remarkable property of systems with a small number of particles and it constitutes the basis on which Hill's approach is built. We will exploit this connection to show that the formalism of nanothermodynamics can also be applied to large systems with long-range interactions.
Niels Tas - Capillarity in Silicon Based Nanochannels
Elasto-capillary deformation of thin capped nanochannels can be used to measure the pressure in isolated confined liquid plugs. These measurements confirm the existence of capillary negative pressure, in accordance with the Young-Laplace equation. Despite the tension in the liquid, plugs are stable on a time-scale of minutes. The role of the confinement in the (meta)stability will be discussed.
Dynamic capillarity measurements, following the model of Washburn, reveal the existence of (apparent) viscosity effects in water flowing in sub-100 nm nanochannels. The ionic content of the water plays a crucial role in understanding these effects.
Christophe Labbez - Chemical potential of charged species at solid/liquid interfaces
In this talk I will discuss and criticize the concept of “surface pH”. This concept was introduced early in the last century by J. Danielli to explain the change in interfacial tension of interfacial films formed by long chain aliphatic acids in conjunction with their degree of dissociation. Today it has spread to a large scientific community including analytical chemistry, electrochemistry and colloid and surface chemistry to rationalize surface charge formation, adsorption and transport of charge species, corrosion, particle growth and so forth. Using simple thermodynamic arguments I will show that the notion of “surface pH” is simply wrong as long as systems are at equilibrium. This will be further illustrated with the application of fluctuation theory on a model charged plate immersed in a chloride acid solution. Finally, I will conclude with the potential application of fluctuation theory to the detection of traces of charged species (proteins, heavy metals, nanoparticles …) in aqueous solutions.
Giancarlo Franzese - Thermodynamics of nanoconfined water and for water at biological interfaces
We study, by simulations and analytic approach, the behavior of water in hydrophobic nanoconfinement or at the interface of proteins [1,2]. We analyze the metastability of supercooled liquid water, between hydrophobic walls, with respect to ice formation and clarify how the interplay between breaking of hydrogen bonds (HB) and cooperative rearranging regions of 1-nm size gives rise to diffusion extrema and offers a possible explanation for the ultrafast transport of water in sub-nanometer channels [3,4]. We discuss the relevance at low temperature of two isobaric loci of increasing fluctuations and of two associated dynamic crossovers due to the HB dynamics, consistent with experiments for hydrated proteins [5,6,7]. We study the fluctuations at cryopreservation temperatures and for confinement in disordered nanochannels we find a dramatic decrease of the average compressibility, relevant for the preservation of food and organic materials .
However, detailed studies of anomalous liquids, including water, confined in disordered nanoporous matrices show a large inclrease of local compressibility near hydrophobic interfaces . Furthermore, we observe that in a slit-pore geometry, depending on how strong is the confinement, the fluctuations increase from 2D-like to 3D-like, something unexpected based on studies for simple liquids. We ascribe this result to the strong cooperativity and the low coordination number of the HB network . If the time will allows us, we will discuss the relevance of these results for the protein folding, superdiffusion in nanoconfinement  and for nanomedicine 
 G. Franzese, V. Bianco, S. Iskrov, Food Biophysics 6 ,186 (2011).
 G. Franzese, and V. Bianco, Food Biophysics, 8, 153 (2013).
 F. de los Santos, and G. Franzese, J. Phys. Chem. B 115, 14311 (2011).
 F. de los Santos, and G. Franzese, Phys. Rev. E 85, 010602(R) (2012).
 K. Stokely, M. G. Mazza, H. E. Stanley, and G. Franzese, Proc. Natl.Acad. Sci. USA 107, 1301-1306 (2010).
 M. G. Mazza, K. Stokely, S. E. Pagnotta, F. Bruni, H. E. Stanley, and G. Franzese, Proc. Natl. Acad. Sci. USA 108, 19873 (2011).
 V. Bianco, M. G. Mazza, K. Stokely, F. Bruni, H. E. Stanley, and G.Franzese, in “Fragility of Glass-forming Liquids”, A. L. Greer, K. F.Kelton and S. Sastry Eds. (Hindustan Book Agency, New Delhi, 2014), pag.183-189.
 E. G. Strekalova, M. G. Mazza, H. E. Stanley, and G. Franzese, Phys.Rev. Lett. 106, 145701 (2011).
 E. G. Strekalova, J. Luo, H. E. Stanley, G. Franzese, S. V.Buldyrev, Phys. Rev. Lett. 109, 105701 (2012).
 V. Bianco and G. Franzese, Nature Sci. Rep. 4, 4440;DOI:10.1038/srep04440 (2014).
 Fabio Leoni and G. Franzese, J. Chem. Phys. 141, 174501(2014)
 P. Vilaseca, K. A. Dawson, and G. Franzese, Soft Matter 9,6978 (2013).
Jean-Marc Simon - Application of Hill’s thermodynamics of small system: The Small System Method
Before reaching the thermodynamic limit, different intensive thermodynamic properties change with the system size. This is particularly the case for molar quantities like the molar enthalpy or molar internal energy. This effect is mainly due to the fact that the system has a finite volume so it exhibits a surface, virtual or not, that affects the values of the properties. For molar quantities, it can also be seen like an effect of the volume of the particles on the limited volume. The thermodynamics of small system of T. L. Hill  gives a general background to systematically express these thermodynamic properties in terms of system size [2, 3]. Alternatively to Hill’s approach similar size effects were obtained using the Kirkwood-Buff’s formalism based on pair correlation functions [4, 5]. We applied these size properties on molecular simulations to calculate at the thermodynamic limit the thermodynamic factor, i.e. the derivative of the chemical potential with the concentration, in open systems [2, 3, 4, 5, 6]. Following the same approach, we also computed partial molar enthalpy and partial internal energy. This was done on systems were other methods fail to compute such thermodynamics properties. We call this new approach the small system method (SSM).
 T. L. Hill. Thermodynamics of small systems. Part 1, Benjamin, New York, 1963.
 Thermodynamics of small systems embedded in a reservoir: a detailed analysis of finite size effects S. K. Schnell, T. J. H. Vlugt, J.-M. Simon, D. Bedeaux, S. Kjelstrup, Mol. Phys. 110, 1069 (2012)
 Thermodynamics of a small system in a mT reservoir, S. K. Schnell, T. J. H. Vlugt, J.-M. Simon, D. Bedeaux, S. Kjelstrup, Chem. Phys. Let. 504, 199 (2011)
 Kirkwood-Buff Integrals for Finite Volumes, P. Kruger, S. K. Schnell, D. Bedeaux, S. Kjelstrup, T. J. H. Vlugt, J.-M. Simon, J. Phys. Chem. Lett., 4, 235 (2013)
 How to apply Kirkwood-buff theory of individual species in salt solutions, S. K. Schnell, P. Englebienne, J.-M. Simon, P. Kruger, S. P. Balayi, S. Kjelstrup, D. Bedeaux, A. Bardow, T. J. H. Vlugt, Chemical Physics Letters 582, 154 (2013).
 Fick diffusion coefficients of liquid mixtures directly obtained from equilibrium molecular dynamics, X. Liu, S. K. Schnell, J.-M. Simon, D. Bedeaux, S. Kjelstrup, A. Bardow, T. J. H. Vlugt, J. Phys. Chem. B, 115, 12921 (2011) + Correction J. Phys. Chem. B, 116, 6070 (2012)
Leonard M. C. Sagis - Developing coarse-grained models for mechanical behavior of structured interfaces in nanocapsules and microcapsules using GENERIC and simulations.
With Alan Luo, and Patrick Ilg.
In core-shell encapsulation systems interfacial properties often are crucial to the performance and stability of the system. Of particular importance are mechanical properties such as the surface shear modulus, the dilatational modulus, and the bending rigidity, and transport coefficients for mass and energy across the interface. A wide range of materials are currently being used to create shells, which include low molecular weight surfactants, (block co-) polymers, proteins, and colloidal particles. These tend to self-assemble at the interface into complex (quasi-) two-dimensional microstructures. Applied deformations, temperature gradients, and concentration gradients can cause changes in these microstructures, which often lead to a highly nonlinear response of the system to these perturbations. Constitutive models capable of describing this behavior are still scarcely available.
In this presentation we will discuss how the GENERIC framework in combination with particle-based simulation methods (MC and NEMD) can be used to obtain constitutive models for the behavior of complex interfaces. We will illustrate this for a flat interface stabilized by hard ellipsoids, and will focus in particular on the mechanical properties of such an interface in steady and oscillatory deformations. The models we create in this way are based on the Gibbs dividing surface concept, and essentially macroscopic. The generalization of our and other Gibbs-based approaches to highly curved interfaces, such as those found in nanocapsules, is nontrivial, and we will discuss several important issues that arise in this generalization.
E. Lesniewska - ICB CNRS UMR 6303, CLIPP, University of Bourgogne, Dijon, France.
The presentation will focus onto recent developments: mode synthesizing atomic force microscopy, microwave microscopy, high-speed atomic force microscopy, and conception of new nanoprobes for the investigation of subsurface defects and the surface reactivity.
In a first part, we will present developments concerning three new nanoprobes designed to work in the range pH 2 - pH 14 for temperature, pH, and conductivity measurements. Nanoprobes were associated to atomic force microscopy (AFM) setup using either shear force control or inverted AFM configuration to control the distance between the sample and the nanosensor . The sensing area was limited to 100 nm2. The hydration of hydraulic phases let to an increase of the pH detected by the nanoprobe, due to the liberation of HO- ions in solution. The measuring scheme involves the time-resolved acquisition of pH evolution on reactive surfaces during their hydration . We have simultaneously performed temperature measurements on anhydrous phases of nanopowder using our nanothermocouple in the electrolytic solution . The challenge consisted in studying the first minutes of hydration. The results indicate that the nanothermocouple only can detect the temperature rise due to the exothermic reaction of anhydrous phases with water, contrarily to the macro-thermocouple, unable to detect anything at this scale. These results clearly show the high potentialities of this new temperature nanosensor. Recent developments in conductimetry at nanometer-scale resolution, allowed study of the diffusion processes involved at the beginning of hydration. These nanoprobe developments associated to standard AFM investigation allow understanding of surface reactivity.
In a second part, we will present new imaging modes. Atomic force microscopy (AFM) and other techniques derived from AFM mostly provide surface properties, while the observation of sub-surface nanoscale defects remains a challenge. Two techniques allow tomographic investigation of subsurface structures. We show that soft material can be probed using a nondestructive imaging method, called mode synthesizing atomic force microscopy (MS-AFM) [4-6]. MS-AFM is based on the interaction of two ultrasonic waves (5-10 MHz), one launched by the AFM probe, a second launched by the sample, and the resulting nonlinear frequency mixing. We will first illustrate their capabilities on study of nanofabricated calibration depth samples consisting of buried metallic patterns. Reconstruction of the depth profile of the sample could be performed, resolving the buried nanostructures with high fidelity. Moreover we applied this multi-frequency analysis to visualize with high resolution insulated polymers and lipid vesicles contained in bacteria. We have developed another non-invasive and innovative tool of characterization of material: make an adaptation of the scanning microwave microscope (SMM) to obtain tomography. Scanning Microwave Microscopy (SMM) uses microwave in 0.2 – 16 GHz range to measures electromagnetic interactions of the microwave from a sharp probe or aperture with the sample under test on a scale that is significantly less than the wavelength of the radiation [7-8]. Recent results on High Speed AFM will be presented to conclude in favor of dynamics of interface.
 J.P. Ndobo-Epoy et al. Ultramicroscopy, 103, 229 (2005).
 J.P. Ndobo-Epoy et al. Analytical Chemistry 79, 7560 (2007) 7560.
 E. Lesniewska. Temperature Nanosensor for Studies of Reactive Materials in Solution at a Nanoscale. Analytical Chemistry (submitted).
 L. Tetard et al. Nature Nanotechnology, 3 (8), 501 (2008).
 M. Ewald, Nanotechnology, 25, 295101(2014)
 P. Vitry et al. Appl. Phys. Lett, 105, 053110 (2014).
 C. Plassard et al., Phys. Rev. B, 83, 121409(2011)
 V. Optsanamu et al., Nanoscale DOI: 10.1039 (2014)