7th International Workshop on Mechanisms of Vacuum Arcs (MeVArc 2018)

20 May 2018, 08:00 → 24 May 2018, 18:00 America/Puerto_Rico

Other Institutes

Antonio De Lorenzi (Consorzio RFX), Chris Moore (Sandia), Flyura Djurabekova (University of Helsinki), Matthew Hopkins (Sandia National Laboratories), Sergio Calatroni (CERN), Walter Wuensch (CERN), Yinon Ashkenazy (The Hebrew University of Jerusalem (IL))

Introduction

Vacuum arcs are a concern in nearly every vacuum electronic device, consequently they are present in a very wide range of applications. Sometimes they form the basis for device operation, but all too often they are the primary failure mode.

Understanding the physical processes of a vacuum arc requires expertise from many disciplines – material science, surface physics, and plasma physics. Applications include high-voltage electronics, RF accelerators, and vacuum interrupters. The purpose of this workshop series is to bring together scientists and engineers from many different disciplines and application areas to discuss the latest efforts in understanding vacuum arcs. We cover theory, simulation and experiments.

The workshop will run from May 20 to May 24 2018 in San Juan, Puerto Rico. An opening reception will occur on the evening of Sunday, May 20, followed by oral sessions Monday through Thursday and a poster session. Additionally, there will be a banquet on Wednesday night at Oceano:

Participants requiring a letter of invitation for visa purposes should contact the organizers. Visiting Puerto Rico is the same as visiting the USA.

What to Expect in Puerto Rico

We want to assure participants that the workshop location is both convenient and safe! May is not hurricane season and while there are certainly continuing challenges across the full island from hurricane Maria, we are in Old San Juan, the first area to return to full operation. Unfortunately, news programs tend to focus on the negative. Our hotel suffered no damage. The immediate area is heavily tied to tourism, including the cruise ship port nearby. There will be restaurants and bars to walk to in a great historical venue and an overall wonderful environment! In addition to having a great time at the workshop, our participation (and our financial footprint) will help the local community, and Puerto Rico as a whole, recover!

Links referenced above: http://puertoriconow.seepuertorico.com/ and http://status.pr/.

Prior Workshops:

Presentations from prior workshops can be found at:

• 6th MeVARC
• 5th MeVARC
• 4th MeVARC
• 3rd MeVARC
• 1st MeVARC

 

Funding

The organizers wish to acknowledge the following institutions for providing funding for the workshop:

• Sandia National Laboratories
• CERN/CLIC

Sunday, 20 May

Welcome Reception

Monday, 21 May

Conference Activity: Welcome and Opening Remarks

MeVARC 2019 site presentation.pdf

MeVARC 2019 site presentation.pptx

Overview/Foundational

1 Applications of High-Gradient Accelerators

Prototype accelerating structures are now being built and operated routinely with gradients in excess of 100 MV in test stands which are part of the CLIC TeV-range electron positron collider project. This represents a factor of approximately three beyond the highest gradient linacs in operation today, and is due to a significantly improved understanding of how to design rf structures for high fields. The success of these structure tests, along with the performance of the X-band rf systems which power them, is inspiring the spread of the high-gradient technology to numerous other projects including Inverse Compton Sources, X-ray Free Electron Lasers and medical linacs. A survey of selected applications is presented.

Speaker: Walter Wuensch (CERN)

applications 2018-5-11.pdf

applications 2018-5-11.pptx

2 Review of high voltage breakdown: from multipactor to ionization discharge*

High-voltage breakdown in the vicinity of a dielectric or conducting surface is examined across a wide range of conditions using theoretical and experimental treatments. DC and RF power sources are considered, across a wide pressure range. DC multipactor along an insulating surface can lead to local heating-driven gas desorption and ultimately to gaseous breakdown. In microwave driven systems, at low pressure, a single-surface multipactor absorbs about 2% of the microwave energy and has a mean energy of hundreds of eV. At 10-50 Torr for L-band radiation, a transition occurs from a single surface multipactor to a detached ionization discharge. Above 50 Torr, the multipactor disappears and the discharge forms a typical sheath, with mean electron energy below 10 eV. Simple scaling laws fit results in the low and high pressure regimes for several gases. Experimental results demonstrate a variable long statistical delay time, followed by a rapid breakdown. UV illumination of the dielectric surface reduces the statistical delay time, making onset of breakdown more consistent. Experiments recently demonstrated arrays of plasma filaments aligned along electric field lines, spaced ≤ ¼ wavelengths at low pressure, with filaments coalescing into more continuous diffuse plasmas at higher pressure. A 1D drift-diffusion fluid model combined with an analytic model for EM wave propagation though plasma slabs of arbitrary profile was able to demonstrate the propagation and filament spacing mechanisms, including decreasing spacing with increasing microwave power, as well as the diffuse plasma transition at higher pressure. Finally, an open breakdown reference platform and validated models are proposed to standardize breakdown and mitigation studies.

* Research supported by the US AFOSR MURI grant FA9550-18-1-0062, and an MSU Foundation Strategic Partnership Grant.

Speaker: Prof. J. Verboncoeur (Michigan State University)

Verboncoeur_Breakdown May2018.pdf

Verboncoeur_Breakdown May2018.ppt

3 Advancing electrode models for PIC-DSMC simulation of discharge between real electrodes

We have developed a stochastic surface model for use in Particle-In-Cell Direct Simulation Monte Carlo (PIC-DSMC) simulations of vacuum discharge. In the present work, we simulate breakdown between two parallel Pt plates in which the modelled electrode surface elements are given a local work function and field enhancement factor (β) by drawing values from probability density functions based on the electrode material and preparation/conditioning. Presently, PhotoEmission Electron Microscopy (PEEM) measurements are used to inform the work function distribution. The distribution for the field enhancement factor is obtained by meshing the surface topology obtained via Atomic Force Microscopy (AFM), solving for the roughness-resolved (~nm’s) electric field and field emission, and then computing an effective emission area and β at PIC-DSMC (~μm) element scales that yields comparable field emission. The framework of the model is extensible to the case where the properties vary in time (e.g. contaminant sublimation or electromigration); however, the current results assume the local properties are constant.

Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Speaker: Chris Moore (Sandia National Labs)

Moore_Advancing electrode models for PIC-DSMC simulation of discharge.pdf

Moore_Advancing electrode models for PIC-DSMC simulation of discharge.pptx

Coffee Break

Field Emission

4 Nanoscopic and Surface Aspects of Field Emission and Vacuum Arcs

The performance and efficiency of the future linear accelerators is determined by the quality of the RF structures used to accelerate the particle beams. This quality is affected by the inner surface of the structures which are subject to field emission of electrons and even cause vacuum breakdowns at the desired high acceleration gradients. Better understanding of the mechanism of the forming of the field emitting spots under the electric field is needed to mitigate the breakdown process. We are addressing that issue with an atomic scale study of the surface of the accelerator structure by measuring breakdown, pre-breakdown and Fowler-Nordheim tunneling curves in the scanning electron microscope (SEM). Our in-SEM high voltage system setup measures the tunnel current in a DC mode between a sharp needle and surface under vacuum conditions. The needle and the sample is positioned using in-situ piezo stage which provides nanometer lateral resolution. To understand the structural and chemical modifications at the mono-layer of the surface in the breakdown region we use high resolution SEM and X-ray photon electron spectroscopy. We present the latest results and progress of the project.

Speaker: Marek Jacewicz (Uppsala University (SE))

MeVArc18_jacewicz.pdf

MeVArc18_jacewicz.pptx

5 CHARACTERIZING THE INFLUENCE OF ELECTRODE SURFACE MORPHOLOGY ON FIELD EMISSION

Field emission (FE) from high-voltage electrode surfaces governs arc breakdown processes in the limit of microscale electrode spacings. In this regime, microscopic features of the electrode surface exert an influence on emitted electrons via local field enhancement at high-aspect-ratio features and work function variation. However, the tendency for surface charge distributions to alter the apparent work function locally (e.g., from stepped surfaces, physi-/chemisorbed layers) has been less a feature of breakdown models that incorporate FE processes. To improve these models, morphology and local work function of polycrystalline Pt (poly-Pt) thin films were characterized using an array of techniques, including transmission electron microscopy (TEM), electron backscatter diffractometry (EBSD), atomic force microscopy (AFM), and photoemission electron microscopy (PEEM). Where work function values of clean, single-crystal Pt(111) approach 6 eV, grain tilt and surface morphology measurements suggested a lowering of the poly-Pt work function by 0.42 +/- 0.20 eV. This was confirmed by local work function maps acquired with PEEM, which showed an average work function of 5.60 eV, with a standard deviation of 30 meV. Measured work function distributions and surface morphology were incorporated into discharge simulations to model field emission based on local surface properties.

Speaker: Morgann Berg (Sandia National Laboratories)

Berg_201805_7th_MeVArc_Berg_oral_draft_05142018_rev.pdf

Berg_201805_7th_MeVArc_Berg_oral_draft_05142018_rev.pptx

Lunch Break

Modeling and Simulations

6 Do Cracks and Unipolar Arcs explain vacuum breakdown and gradient limits?

Although discovered 118 years ago, the understanding of vacuum arcs has proceeded very slowly, due to a number of factors. We believe a simple model based on surface cracks and unipolar arcs gives the clearest picture of this process. We will describe how our model, based almost entirely on our own experience with 805 MHz RF data at Fermilab and mechanisms not mentioned in the arcing literature before 2001, seems appropriate to applications as different as power grid losses, fusion devices, integrated circuit failure modes, laser ablation, vacuum microelectronics, classic plasma physics, as well as gas breakdown. We also compare its predictions with experimental data, various applications and with other models.

Speaker: Dr Jim Norem (Nano Synergy Inc)

Norem San Juan talk.pdf

7 Simulations of Multi-Phase Processes of Arc Interaction with Electrodes

Although Explosive Electron Emission (EEE) is well known for vacuum arcs [1], effects of gas pressure on EEE processes have been rarely studied so far [2]. Often, anode erosion associated with crater formation and liquid metal pool on the anode surface has been observed. In our presentation, we will report results of an ongoing work to develop computational models of arc interaction with electrodes for a wide range of gas pressures and arc currents. Simulations of craters on electrode surface, formation of plasma jets, acceleration of ions to high energies, and other phenomena associated with arc interaction with electrodes require coupling plasma physics with multi-phase flow science.
We have developed a Unified Flow Solver (UFS) with dynamically adaptive Cartesian mesh and multi-scale simulation capabilities [3]. The Adaptive Mesh and Algorithm Refinement (AMAR) methodology of UFS is being extended to add multi-phase capabilities for arc interactions with electrodes and electrode erosion processes. We use the Volume of Fluid model to dynamically trace solid-liquid interface and liquid molten pool dynamics, as well as expansion of plasma jet in vacuum and ambient gas.
We have simulated development of a liquid pool of molten Cu surface and formation of splashes and droplets. The dynamically adapted grids capabilities were found to be crucially important to properly resolve the freely moving gas-liquid interface and the electric fields around the sharp edges of the interface. Current work includes simulations of single ectons on tip of Taylor cones formed on liquid cathodes. EEE results in the formation of plasma jets expanding from the cathode. During jet expansions, a non-ideal plasma turns into an ideal plasma. We use hybrid fluid-kinetic models of UFS to analyze effects of ambient gas pressure on the dynamics of EEE-induced plasma processes.

Acknowledgments
This work is supported by the DOE SBIR Project DE-SC0015746 and by the NSF EPSCoR project OIA-1655280.

References
[1] G. A. Mesyats, Plasma Phys. Control. Fusion 47 (2005) A109–A151.
[2] A. Anders, Cathodic Arcs: From Fractal Spots to Energetic Condensation, Springer (2008)
[3] V.I. Kolobov, R.R. Arslanbekov, J. Comput. Phys. 231 (2012) 839.

Speaker: Dr Vladimir Kolobov (CFD Research Corporation)

Kolobov MeVarc-Kolobov_2018_rev5.pdf

Kolobov MeVarc-Kolobov_2018_rev5.pptx

Coffee Break

Experiments and Diagnostics

8 Electrical vacuum breakdowns in large-gap DC-setup: experiment versus theory

Vacuum breakdowns in DC conditions between the electrodes with large gap (several mms) are believed to initiate at anode. We have investigate the origin of anodic plasma by setting up DC experiments with the fast intensified charge-coupled device camera (ICCD) and combining our observations with theoretical calculations. We observe the evolution of light emission, analyze the typical waveforms of voltage and current during the breakdown and conclude that origin of plasma is not at anode.

The presentation will discuss the new observations and give the overview of the theoretical efforts ongoing at University of Helsinki on a multiscale simulation of vacuum breakdown phenomenon.

Speaker: Flyura Djurabekova (University of Helsinki)

Djurabekova_MeVARC2018.pdf

Djurabekova_MeVARC2018.pptx

9 Spectroscopic study of high-current vacuum arcs considering anode activity

Development of vacuum circuit breakers for high-voltage applications leads to continuous basic research in the field of vacuum arcs. Investigations of the arc plasma properties is an important instrument for understanding of basic phenomena and parameter optimization for corresponding applications. It is well known that the high-current anode phenomena have a distinct impact on contact erosion and interrupting capability, because they lead to injection of atomic vapour into the interelectrode gap causing the lowering of dielectric strength.

The contribution presents the results of optical measurements using emission and absorption spectroscopy techniques. Emphasis is put on the determination of spatial distributions of line emission from various plasma species – atoms, single and double charged ions – during different high-current modes. The driving current pulse was supplied by a high-current generator that produces an AC waveform at 50 Hz or 100 Hz at several kA maximum current. The high-current anode modes are observed by means of high-speed camera imaging and can be correlated with changes in the arc voltage. Video spectroscopy was applied during the active phase, i.e. before current zero, for determination of temporal dynamics of copper spectral lines for various anode modes – diffuse, intense, anode spot mode 1 and 2. The results show significant changes in the intensities of the atomic and the ionic Cu lines near the anode during the transition to the anode spot mode. Broad band absorption spectroscopy was used for determination of the vapour density (chromium) close to the current zero crossing and in the early post-arc phase. The temporal evolution of the Cr ground state density is presented and discussed.

Speaker: Dr Sergey Gorchakov (Leibniz Institute for Plasma Science and Technology)

Gortschakow_MeVArc2018.pdf

10 Photon-assisted Plasma Breakdown Studies in Plane to Plane Discharges

We present an overview of studies that examine the role of photo-assisted breakdown. Emphasis is placed on an experimental platform that subjects well-characterized platinum surfaces to pulsed UV illumination as functions of UV illumination and UV wavelengths spanning from 260 nm (4.77 eV) to 400 nm (3.11 EV), well below the 6 eV values often used for platinum. Application UV illumination is synchronized to voltage pulses applied across the anode-cathode gap to control the initiation of the breakdown event. Current-voltage characteristics measured during breakdown in modest pressures of helium gas (10-100 Torr) are compared to computational simulations.
Research reported in this publication was supported by the Office of Defense Nuclear Nonproliferation Research and Development. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA000352

Speaker: Ed Barnat (Sandia National Laboratories)

Poster Session

Profatilova_MEVARC2018_DC_system.pdf

Profatilova_MEVARC2018_Poster_Marx Generators.pdf

Tuesday, 22 May

Modeling and Simulations

11 Dynamic Coupling of a Finite Element Solver to Large-Scale Atomistic Simulations

We propose a method for efficiently coupling the finite element method with atomistic simulations like molecular dynamics or kinetic Monte Carlo. It is a continuation of our earlier attempt to develop a tool to perform multiscale-multiphysics simulations about evolution of nanostructures under high electric field. Our method enables to dynamically build an unstructured mesh with optimized density that follows the geometry defined by atomistic data. On this mesh, multiphysics problems can be solved to obtain distributions of physical quantities of interest, which can be fed back to the atomistic system. The simulation flow is optimized to maximize computational efficiency while maintaining good accuracy.

On the first stage of development, the code solved dynamically the Laplace’s equation in 3D domain and used the solution to obtain electron emission currents, charges and electrostatic forces for surface atoms. By taking into account Joule and Nottingham heating and solving 1D heat equation, we obtained also atomistic velocity perturbation.

Over the last year we have extended and improved the capabilities to the code. By incorporating particle-in-cell technique, we obtained the possibility to include space charge in electric field calculation. Replacing 1D heat equation solver with a 3D one increased the overall accuracy and extended the spectrum of possible applications of the code. Several improvements in the algorithm base have significantly improved the tolerance against the geometry under scope.

The tests have shown remarkable overlapping with an analytical solution and proved the code to be efficient and robust enough to simulate large-scale thermal runaway processes.

Speaker: Mihkel Veske (University of Helsinki)

Veske_MeVArc18.pptx

12 Thermal runaway of intensively field-emitting metal tips.

One of the most common hypotheses attempting to explain the ignition of a vacuum arc is the thermal runaway of field emitting tips residing on the metal electrodes. Here we use multi-scale simulations in order to explore this hypothesis, investigate the conditions under which a field emitter is driven to thermal runaway and how this can lead to the ignition of a vacuum arc.

Our simulation method concurrently couples Molecular Dynamics (MD) with the Finite Element Method (FEM) which is used to solve the electrostatic field and thermal diffusion equations. Furthermore, we combine the above with a novel method for the calculation of electron emission from sharp emitters in the general thermal-field regime and the Particle-In-Cell (PIC) method to include the space-charge effects.

We find that in case the emitter geometry is considered static, thermal runaway cannot appear if the tip does not have a minimum size of micrometer order, regardless of its aspect ratio. However, our simulation approach of dynamically coupling MD with FEM allows to follow the shape evolution of a nanotip at an atomistic level. These dynamic simulations show that the field-induced
force elongates and sharpens the hot apex of the nanotip, which initiates a positive feedback process and eventually causes evaporation of large fractions of the tip. We find that the total evaporation rate exceeds the one required to ignite arc plasma as found by recent PIC simulations of the plasma onset. Therefore the above mechanism can explain the ignition of a vacuum arc as an
intrinsic response of a nano-scale metal tip to the application of a high
electric field.

Speaker: Andreas Kyritsakis

Kyritsakis_MeVarc_2018.pdf

13 Stochastic model of breakdown nucleation under intense electric fields

Plastic response due to dislocation activity under intense electric fields is proposed as a source of breakdown. A model is formulated based on stochastic multiplication and arrest under the stress generated by the field. A critical transition in dislocation population is suggested as the cause of protrusion formation leading to subsequent arcing. The model is studied using Monte Carlo simulations and theoretical analysis, yielding a simplified dependence of breakdown rates on the electric field.

After presenting the underlying principles of the model, we discuss its latest developments. Sample analysis performed by our research group is used to determine the values of constants in the model, and results from the model are compared to experimental breakdown times in order to calibrate the unknown remaining parameters of the model and then to validate it.

A number of experimental setups are proposed, which can further test the validity of the model and its predictions. this provides an opportunity to establish the model in order to develop methods of limiting dislocation mobility, as well as of establishing prebreakdown warning signals through the evolution of the dislocation population.

Speaker: Eli Engelberg (The Hebrew University of Jerusalem)

main.pdf

Coffee Break

Experiments and Diagnostics

14 CURRENT SIGNALS AND X-RAYS SPECTRA ANALYSIS FOR A VACUUM HIGH VOLTAGE HOLDING EXPERIMENT

The High Voltage Padova Test Facility (HVPTF) is an experimental device for investigating HV insulation in vacuum, in support of the realization of MITICA, the prototype neutral beam injector for ITER. The experiments here described aim at understanding the physical phenomena underlying voltage holding in vacuum and specifically the electrode "conditioning" process. Two stainless steel electrodes are positioned inside a high vacuum chamber, separated by a few cm gap. When HV is applied (up to 400 kV), the breakdown voltage typically increases in time and achieves a saturation value in about ten hours. A highly time-resolved X-rays diagnostic has been recently installed in the device, measuring single X-ray emission events produced during the "conditioning" phase. High energy peaks have been observed in correspondence to each current peak collected by the electrodes. This work presents a characterization of the current and the X-rays events during the conditioning phase, in terms of frequency, amplitude and their occurrence with respect to the voltage between the electrodes.

Speaker: Dr silvia spagnolo (consorzio rfx)

spagnolo_MeVarc2018_def.pdf

15 Updated Results of Breakdown Study for a 509-MHz CW Accelerating Cavity based on Direct In-situ Observation

Normal-conducting RF accelerating structures are hearts of many modern accelerators, where vacuum arcs in the structures could lead to breakdowns, that might limit accelerator performance. However, we do not know what the real source of breakdowns is, i.e., the breakdown-trigger mechanism is not yet well understood.
Recently, based on direct in-situ observation method, we performed breakdown study of a normal-conducting 509-MHz continuous-wave (CW) single-cell cavity developed for SuperKEKB at KEK, and discovered that there were clear and stable bright spots on the inner surface of the cavity, which continued during high-power operation, and that such bright spots exploded and then disappeared at one-fifth of the breakdown events. We also found that a decrease in the number of stable bright spots after an explosion was a significant component of RF conditioning of accelerating structures. Furthermore, we observed sudden appearance of new bright spots shortly before breakdowns, that has stimulated our interest in the microscopic dynamics of the generation, growth, and explosion of bright spots and their correlation with RF conditioning effects and breakdown rates. Understanding the physical properties of such bright spots must be a key to elucidate the breakdown-trigger mechanism.
In this report, we present updated results of our breakdown study based on direct in-situ observation using higher-spec cameras, including spectra of the bright spots.

Speaker: Dr Tetsuo Abe (KEK)

MeVArc2018_CavBD_ABE_2018-05-22_v4.pdf

Lunch Break

Applications

16 Update on high-power testing of X-band RF structures at CERN

At the high-gradient X-band test facility at CERN, prototype accelerating structures for the Compact Linear Collider (CLIC) are tested at high powers of up to 50 MW to achieve accelerating gradients of over 100 MV/m and peak surface electric fields of over 220 MV/m, and are conditioned to reach the highest possible gradient at low breakdown rate. The setup of the test stands will be presented, as well as recent structure results including conditioning histories, breakdown localisation, field emission in the form of Faraday cup signals and radiation, as well as comparisons between different structure designs and conditioning strategies.

Speaker: Mr Jan Paszkiewicz (University of Oxford (GB))

CERN_Xboxes_Jan_Paszkiewicz.pdf

CERN_Xboxes_Jan_Paszkiewicz.pptx

17 Recent Results of X-Band High-Gradient Tests at KEK / Nextef

We have been performing high-gradient tests of X-band (11.4 GHz) normal-conducting accelerating structures at a test facility called Nextef in KEK, aiming at operational accelerating gradients of 100 MV/m or higher. In this report, we present recent results of high-gradient tests on the test structures, including a quadrant-type waveguide-damped single-cell test cavity.

Speaker: Tetsuo Abe (High Energy Accelerator Research Organization (KEK))

18 Autopsy of cryogenic normal conducting X-band cavity

Valery's abstract

Speaker: Valery Dolgashev (SLAC)

MevArc2018_Dolgashev_Single_Cell_Cryo_Autopsy_final_22May2018.pdf

MevArc2018_Dolgashev_Single_Cell_Cryo_Autopsy_final_22May2018.pptx

Special Topic: ITER status

BA_TDL.pdf

BA_TDL.pptx

DTT.pdf

DTT.pptx

fusion_slides_for_mevarc18.pdf

fusion_slides_for_mevarc18.pptx

Conference Activity: Social Excursion

MeVARC 2019 site presentation.pdf

MeVARC 2019 site presentation.pptx

Wednesday, 23 May

Experiments and Diagnostics

19 Constraints on stochastic plastic activity as a source for BD nucleation

It has been suggested that BD can nucleate due to stochastic plastic activity induced in the cathode under extreme fields. The nature and type of plastic activity can be constrained from experimental observations on post-BD electrodes. I will present estimates to such constraints from transmission electron microscopy of samples exposed to high fields, discussing properties like dislocations densities, denuded zones, lateral and depth correlation lengths. These together with understanding gained from effective mean filed model, are used to propose basis for advanced models and future experimental setups.
Such advanced models have the potential of explaining the role of electrode structure as well as operation procedure during conditioning.

Speaker: Yinon Ashkenazy (The Hebrew University of Jerusalem (IL))

mevarc_2018_huji.pdf

20 Copper Measurements with High Frequency Ultrasound Microscope

Field emission and electrical breakdowns play a central role in the RF conditioning process of particle accelerating structures. However, actual mechanical changes in the copper surface and just beneath the surface are unclear. Ultrasound microscopy is an advanced method to measure mechanical properties point-by-point from the sample surface. High lateral resolution, 3.5 μm, can be achieved by using focused 250 MHz ultrasound. By amplitude calibration both bulk and Young’s moduli can be extracted from ultrasound echoes, as well as from time of flight imagines that describe the topology of the sample surface.

To understand how breakdowns and the field emission affect the copper sample, we measured copper electrodes of the CERN DC-spark-system, before and after breakdowns. Similarly, we have studied the field emission effect.

We are currently developing a method to image subsurface features of the sample based on synthetic aperture focusing technique (SAFT). Whereas, traditional focused high frequency ultrasound is mostly limited only short focus distance, 50 μm – 700 μm and only suitable for surface imaging, SAFT allows one to move numerically the focus beneath the surface.

Speaker: Mr Antti Meriläinen (Helsinki Institute of Physics (FI))

AM_Mevarc2018.pdf

AM_Mevarc2018.pptx

21 Microscopy at CERN

The importance of microscopy analysis and observations have been proof over the last years over the post-mortem analysis of different structures. Here we will see as well its importance in other fields and pollution analysis, as it can be the origin of a breakdown, the analysis of bonded structures, to guarantee a good electrical contact and vacuum tightness.

Speaker: Enrique Rodriguez Castro (CERN)

MeVArc2018_ERCvIndico.pdf

MeVArc2018_ERCvIndico.pptx

Coffee Break

Modeling and Simulations

22 Modeling of Cathode Spot Crater Formation and Development on pure Cu and CuCr alloy in Vacuum Arc

The two-dimensional (2D) rotary axisymmetric model is used to describe the formation and development of cathode spot on pure Cu and CuCr alloy in vacuum arc. The model includes hydrodynamic equations and heat transfer equation. Parameters used in this model come from experiments and other researchers’ work. The influence of each parameter is analyzed and the simulation results are compared with pure metal simulation results. In the simulation, the process of cathode crater, and distribution of temperature are obtained,the depth of cathode crater is from 0.5μm to 1.1μm, the radius of cathode crater is from 1.6μm to 2.6μm, the maximum velocity of droplet is from 200m/s to 600m/s and the maximum temperature is from 3500K to 5000K which is located in the area with radius 0.5~1.5μm. Cathode spot appears on chromium grain only and the ablation on CuCr alloy is the smallest.
In the simulation, On CuCr alloy, cathode spot appears only on chromium grain and compared with pure metal, whether it is pure copper or pure chromium, the cathode crater on alloy is smaller which means under the same conditions, the ablation of the alloy is less than pure metal. The simulation results are in agreement with experiments and other researcher’s simulation on the crater size and liquid metal shape.
In our future works, more factors will be considered, such as thermo-field mitted electrons, backflow ion flux, back-diffused electrons and Lorentz force and so on.

Speaker: Prof. Lijun Wang (Xi'an Jiaotong University)

Model of Cathode Spot Crater Formation and Development in Vacuum Arc.pdf

23 Droplet emission from non-refractory cathodes: From vacuum arcs to ultra-high-pressure arcs

Cathode erosion of vacuum arcs is due to emission of high-speed plasma jet of energetic ions and a liquid droplet ejection from a melted cathode surface. Most of the research (experimental and theoretical) has been devoted to studies of the ionic erosion. However, droplet emission could be as important (and even prevalent) mechanism responsible for the cathode erosion. According to a widely accepted model [1], the ejection of the droplets is caused by a very high pressure of vaporized cathode material at the foot of the cathode spot. That pressure expels the melt from the bottom of the cathode liquid pool thus creating a crater on the cathode surface. In many existing works, the expulsion rate has been calculated. However, at present, these is no model that would allow one to calculate what fraction of the expelled metal remains at the cathode in the form of a rim and what fraction leaves it in the form of droplets. In our presentation, we will suggest a simple model, which allows one to find that fraction. The central point of our model is the solidification of the molten metal extruded from the crater. No adjustable parameters are used in the model, which operates only with experimental data. In the first part of our presentation, we will describe the model and calculate the droplet erosion rate in a vacuum arc. The satisfactory agreement with experimental data for the erosion rate has been obtained. In the second part, the calculations of the erosion rates for high-pressure arcs will be presented. Comparison of simulation results with experimental data shows satisfactory agreement and may explain the observed non-monotonic dependence of the erosion rate on gas pressure.

Acknowledgment

This work is supported by the DOE SBIR Project DE-SC0015746.

References

[1] G W McClure. Journal of Applied Physics 45, 2078 (1974)

Speaker: Dr Valerian Nemchinsky (Keiser University)

Nemchinsky_Puerto Rico 2018.ppt

Lunch Break

Modeling and Simulations

24 Study of a low-pressure high voltage switch triggered by laser-surface interaction

Here, we present simulation of a low-pressure, high voltage switch triggered by a laser pulse. The switch is held at a nominally high voltage in an environment absent of sufficient gas particles such that breakdown does not occur. Application of a controlled laser pulse to the cathode surface initiates injection of neutral material and electrons into the low-pressure gap. The increase in pressure allows for more frequent electron-neutral collisions, specifically ionization such that the switch begins to close. Injection of cathode material (carbon) is modeled using a first-principles model of the laser-cathode interaction that tracks the dynamics of electron and hole populations in momentum-resolved conduction and valence band states including carrier-carrier and carrier-phonon scattering. This scattering results in electron and lattice heating and results in thermal-field emission and cathode ablation. This laser-surface model is used to generate time-dependent fluxes of neutral and charged material which is then coupled to a Particle-in-Cell (PIC), Direct Simulation Monte Carlo (DSMC) code that simulates the subsequent plasma formation. Using a 500 mW laser with a 1 microsecond pulsed duration, the laser-surface model predicts a carbon neutral injection flux of 1028 m-2s-1 and a space charge limited electron current density of 200 A/cm2. The low-pressure switch is assumed to be filled with nitrogen gas at 1 torr with an applied voltage of 10 kV. Simulating the plasma formation leading to switch closure shows the formation of a high carbon density behind a traveling shock resulting from the expanding sublimated carbon and its interaction with the background nitrogen. As time proceeds, the integrated density across the gap becomes is much higher than one torr. As a result, electron ionization increases significantly across the gap causing plasma formation and initiating gap closure.

Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Speaker: Andrew Fierro (Sandia National Laboratories)

Fierro_MeVarc_2018.pdf

Fierro_MeVarc_2018.pptx

25 Computing properties of Pt surfaces for understanding electron emission

Electron emission from a surface, and ultimately electrical breakdown, will be governed by local heterogeneity in composition, structure and work function across a substrate. Even for a seemingly simple substrate such as Platinum, without contaminants or adsorbed layers, understanding of the surface structure, much less the electronic behavior across a surface, is poor. We present a systematic approach for computing surface structure and energies that resolves remaining limitations in density functional theory calculations of surface properties. The methods are established in validated Aluminum surface studies, and then detailed calculations for various reconstructed low-index surfaces for Platinum provide new explanations into observed surface reconstructions and seemingly anomalous growth behavior, and gives new insight into possible emission behavior interpreted in terms of a local work function. The implication of these results are interpreted in terms of models of electron emission and surface measurements with STM, as a foundation of a multiscale approach to understanding and modeling vacuum breakdown for grown Pt surfaces.

Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc. for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-NA0003525.

Speaker: Peter Schultz (Sandia)

MeVArc18_Schultz_v3-present.pdf

MeVArc18_Schultz_v3-present.pptx

Coffee Break

Field Emission

26 The Effects of Non-Uniform Work Functions on Field Emission

(abstract forthcoming)

Speaker: Dr Harold Hjalmarson (Sandia National Laboratories)

27 Imaging Dark Current at Microscale

Dark current (undesirable field emission electron current) is proven to be one of electric breakdown triggers. Breakdowns are accompanied with x-ray flashes and ion-electron plasma bursts. At the breakdown location the material is often damaged – it is caused to melt, flow and re-crystallize – which indirectly suggests exceedingly high temperature at those locations. Breakdowns happen at microscale, but compromise/interrupt the normal operation of bench-size high power vacuum electronics devices (e.g., TWT) or large scale accelerator facility (e.g., GeV or TeV linear colliders or light sources). Most importantly, dark current induced breakdown limits the figure of merit of an accelerator, the acceleration gradient. To push accelerator research frontier, the breakdown phenomenon is studied with highest attention at the accelerator research centers.

There are two basic questions. First – why and where is breakdown initiated? Second – what are consequences after a breakdown occurred? To directly address these questions and unveil processes associated with the dark current that take place at microscale, we conducted a series of combined dark current imaging experiments by making use of custom DC and RF field emission microscopes. For copper, we found that after multiple RF breakdown events occurred, most dark current emitters were breakdown damaged locations. By conducting DC field emission imaging of nanodiamond and carbon nanotube structures, we argue that breakdown is initiated by dark current via Nottingham heating channel. Material’s resistance to breakdown (stability) depends on its electronic band and crystal structure. Namely, the combination of electronic and lattice properties determine the level of temperature enhancement at distributed microscopic locations, and whether a solid state material stably performs or undergoes thermal runaway and evaporation followed by breakdown in the self-induced heating regime.

Speaker: Prof. Sergey Baryshev (Michigan State University)

MEVARC_Baryshev_upload.pptx

28 Dark current fluctuation measurements in RF structures and pulsed DC experiments

The dynamics of dislocation activity under intense electric fields has been proposed as a potential cause of breakdowns. In this talk, results of searches for high-frequency fluctuations in field-emitted current that could arise as a consequence of these dynamics are presented for high-gradient RF accelerating structures and pulsed DC experiments at CERN. Measurements of spatial profiles of field emission in RF structures and long-term variations will also be presented.

Speaker: Jan Paszkiewicz (University of Oxford (GB))

RF_Dark_Current_Fluctuations.pdf

RF_Dark_Current_Fluctuations.pptx

Conference Activity: Banquet

MeVARC 2019 site presentation.pdf

MeVARC 2019 site presentation.pptx

Thursday, 24 May

Modeling and Simulations

29 Nanotip-formation processes in electric fields

The vacuum arc processes are believed to be initiated by the formation of nanotips on the metallic surface under the influence of strong electric fields. However, the exact mechanism for how these kind of nanotips would form has so far not been identified. In this work, we will present results from Density Functional Theory (DFT) and Kinetic Monte Carlo (KMC) studies on how the migration of adatoms are influenced by electric fields and how a biased migration may result in growth of nanotips on metallic surfaces. The results are compared with experimental results in the literature. We will also present studies from recent Scanning Electron Microscope (SEM) studies where, among other things, carbon nanotips were observed to form by deposition on gold electrodes in an applied field.

Speaker: Dr Ville Jansson (University of Helsinki)

jansson2018mevarc.pdf

30 Experiments and simulations of nanoscale surface-field interaction

The measured field enhancement factors of 50-100 in CLIC accelerating structures are associated with high aspect ratio surface irregularities. The formation of these surface structures is thought to occur due to subsurface dislocation activity and field-assisted surface atom diffusion. In the current work, we investigate these aspects using computer simulations and experimental approaches. Surface diffusion in the absence of field was studied for FCC metals using nanowires as model systems. Experimental observations and kinetic Monte Carlo simulations showed that nanowire junction disintegrate due to surface atom diffusion at temperatures much lower than the melting temperature. This was due to increased surface roughness at the junction spots. The stability of surfaces was further investigated with molecular dynamics and DFT calculations to estimate the change in work function due to surface roughness. It was found that surface tips and ridges can significantly reduce the work function. Further generalization of conducted studies resulted in in-situ SEM experiments with applied field between metal tips. Images were taken before and after voltage application. FEM simulations were used to reconstruct possible electric field distributions and surface behavior in observed situations.

Speaker: Simon Vigonski (University of Tartu)

MeVArc 2018 Vigonski Combined.pdf

MeVArc 2018 Vigonski Combined.pptx

31 Artificial neural networks for Cu surface migration barrier prediction

Diffusion is a slow process compared to atomic vibration. It is thus inefficient to simulate using methods that work in that short time scale, such as Molecular Dynamics. Substituting the chaotic motion of individual atoms with discrete jumps between potential energy minima yields an efficient and widely used method that is Atomic Kinetic Monte Carlo (AKMC). Each jump has a rate $\Gamma$ determined by its energy barrier $E_m$ and temperature $T$:
\begin{equation}
\Gamma \propto \exp\left(-\frac{E_m}{k_BT}\right)
\end{equation}

The weakness of AKMC is the requirement for the knowledge of the energy barriers associated with each jump. Accuracy of the simulation can be improved by allowing new types of jump events, and calculating their barriers, until the number of required barriers grows unfeasibly high.

An alternative way to approach this trade-off between accuracy and affordability is to cut corners with machine learning. A subset of jumps can be sampled from the entire configuration space, and an artificial intelligence can be taught to interpolate and extrapolate as necessary during the AKMC simulation.

In this work, Artificial Neural Networks (ANN) have been used to learn and predict Cu surface migration barriers. The ANN barrier predictor method has been implemented in an AKMC code Kimocs, developed earlier in our group. The results this far will be described in this presentation. The long-term goal of the project is to model the Cu surface under electric field more accurately than before, to study the phenomena that take place in Cu structures such as those of the Compact Linear Collider.

Speaker: Jyri Lahtinen

Lahtinen_Jyri_MeVArc_2018.odp

Lahtinen_Jyri_MeVArc_2018.pdf

Coffee Break

Applications

32 Mitigation of the breakdown induced effects on the 1 MeV – 16.5 MW Neutral Beam Injector experiment in Padova

The voltage breakdown in the accelerator of Neutral Beam Injectors for Plasma Heating and Current Drive (H&CD) has a very frequent occurrence, due to the harsh environment of the accelerator. The beams formed by accelerated defocused ions and electrons escaped from magnetic traps hit the electrodes (e.g. the acceleration grids and the vessel) melting the surfaces with massive production of gas, metal debris and intense Xray radiation: in such an environment, the breakdown occurrences increase dramatically. Moreover, due to the fact the p.d (pressure x distance) product inside the vessel locally can be very close to left branch of the Paschen curve, Townsend-like discharges have a great probability to occur.
For this reasons, the voltage breakdown in the NBI systems is regarded as an operating condition more than a fault one.
In MITICA - the prototype of the ITER NBI under construction in Padova - this situation is expected to be more and more severe, being its parameters magnitude larger than any other existing NBI device. Basically, the voltage breakdowns in MITICA (up to 1 MV in the case of the direct breakdown of the ion source to ground) produce huge and very fast transient currents. The detrimental effects can be grouped into three categories: i) permanent damages to the grids surface, ii) electrical stresses to the alumina insulators in vacuum, to the Power Supply components (Accelerator and Ion Source Power Supplies), to the Ion Source diagnostic and iii) safety/EMI effects due to current induced in the surrounding ground.
This presentation describes the solutions adopted to mitigate such effects, as well as the transient computer model of the overall system, which will be essential during the next commissioning phase and during the operation. Particular emphasis will be given to the presence of a huge current limiting device (40 tons magnetic core snubber - the key component to reduce the grid damage during breakdown) and to the grounding techniques adopted to keep under control the fast current flow.

Speaker: Antonio De Lorenzi (Consorzio RFX - )

BD EFFECTS- MEVARC_18 PRESENTATION_FINAL.pptx

33 Mitigation of High Field Corona Losses by Surface Treatments

We are applying our understanding of vacuum arc mechanisms by attempting to mitigate coronal losses on high voltage transmission line conductors. These losses seem to involve field emission and occur at surface fields around 20 MV/m. The results of high-voltage tests on the corona discharge in the conditions of simulating the rain of samples of aluminum wire of grade AC 300/39 coated with high-temperature alumina α-Al2O3 modified with graphene oxide and carbon nanotubes are presented. A significant effect is shown of up to 40% of the decrease in the power loss to the corona due to the hydrophilic properties of the coating.

Speaker: Prof. Zeke Insepov (Purdue University)

Lunch Break

Experiments and Diagnostics

34 High-field measurements with kHz repetition rate, microsecond dc pulses

Measurement of different aspects of high-field behaviour including pre-breakdown processes, breakdown events and parameters for conditioning of materials ongoing in pulsed dc systems at CERN are described. The system uses parallel plate electrodes with 40 and 62 mm diameter together with a well-controlled high-voltage high repetition rate generator up to 6 kHz. The dc pulsed experiments are complemented by breakdown localization technique and post-mortem microscopic analysis.
The dependences of several factors (as surface electric field, pulse width, repetition rate and others) to BDR were investigated. The data for different electrodes material and dependencies are collected to the wide database. The results from the different electrodes, along with selected radio frequency results, and ideas for next tests are presented.

Speaker: Iaroslava Profatilova (CERN)

Profatilova_MeVArc_2018.pdf

Profatilova_MeVArc_2018.pptx

35 Characteristics and statistics of breakdowns on copper electrodes

Understanding the microscopical phenomena behind vacuum arc generation is crucial for being able to control the breakdown rate thus improving the efficiency of many applications where breakdown generation is a limiting factor.

Statistical properties of breakdowns, such as pulses between breakdowns, breakdown waveforms and changes in vacuum pressure, are studied using a Large Electrode DC Spark System in Helsinki. In the system, copper electrodes separated by a 60 $\mu$m gap, are placed in near ultra high vacuum. Electric field up to 80 MV/m is pulsed across the gap, resulting in electric discharges - breakdowns.

Statistics are collected over thousands of breakdown events using electric field, pulse length, pulsing frequency or electrode treatment as variables. The statistical analysis is compared to the to the breakdown-induced features on the electrode surfaces which are imaged using various imaging and surface analysis techniques, such as white optical microscopy, white light interferometry and immersion ultrasonics.

The resulting statistics are used to classify the breakdown events in order to understand the underlying processes leading to them.

Speaker: Anton Saressalo (Helsinki Institute of Physics (FI))

2018_05_MeVArc_Anton.pdf

2018_05_MeVArc_Anton_Saressalo.pptx

36 Sputtering in the heat spike and high-temperature regimes

After initiation of an arc, the plasma forming it must be fed by sputtering. The basic mechanisms of sputtering in the linear cascade regime are well understood from decades of theory and computer simulation development. However, under an electrical arc, the surface is either hot or molten, the plasma bombardment intensity very high, and the surface may have nanosize protrusions.
The mechanisms of sputtering under such extreme conditions have started to become clear only recently.

In this presentation, we review the knowledge of sputtering from heat spikes, nanostructures and the highly heated surface regimes, and present results of a recent systematic study of high-temperature sputtering of Cu.

Speaker: Kai Nordlund (University of Helsinki)

nordlund_mevarc18.pdf

Conference Activity: Closing and next MeVArc announcement

MeVARC 2019 site presentation.pdf

MeVARC 2019 site presentation.pptx

Coffee Break