Safety design of Experimental Plasma Generation Systems

Accepted Manuscript Safety design of experimental plasma generation systems Daniel Bondarenko, Hossam A.Gabbar, C.A. Barry Stoute PII:

S0950-4230(17)30190-0

DOI:

10.1016/j.jlp.2017.02.024

Reference:

JLPP 3433

To appear in:

Journal of Loss Prevention in the Process Industries

Received Date: 5 July 2015 Revised Date:

22 February 2017

Accepted Date: 22 February 2017

Please cite this article as: Bondarenko, D., A.Gabbar, H., Stoute, C.A.B., Safety design of experimental plasma generation systems, Journal of Loss Prevention in the Process Industries (2017), doi: 10.1016/ j.jlp.2017.02.024. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Safety Design of Experimental Plasma Generation Systems

Abstract

RI PT

Daniel Bondarenko(a), Hossam A.Gabbar(a,b)*, C.A. Barry Stoute (b) a) Faculty of Engineering and Applied Science, University of Ontario Institute of Technology, 2000 Simcoe Street North, Oshawa L1H7K4 ON, Canada b) Faculty of Energy systems and Nuclear Science, University of Ontario Institute of Technology, 2000 Simcoe Street North, Oshawa L1H7K4 ON, Canada Corresponding author: [email protected]

M AN U

SC

The Experimental Plasma Generation System (EPGS) is a system proposed for studying and investigating the plasma behavior in vacuum conditions and under the influence of the varying boundary conditions. Such device is intended to verify the computational models with the physical data and to gauge on possible sources of risk. The design of EPGS and the associated safety risks are purposed to encompass a variety of existing plasma devices and to act as a case study for researchers interested in preventing risks in their plasma devices. The safety design of EPGS takes into consideration the target safety system design, input and output parameters, diagnostics, as well as the fault propagation scenarios with respect to the Process Safety Management (PSM) via the IEC 61508 standard. The process to the design, the initial findings in the simulation and the experiment point to primary dangers of dealing with high voltages (HV) and vacuum devices, as well as the electromagnetic interference. The cumulative outcome of the current work is to provide groundwork for safety design of plasma devices that will have scalable properties in both the physical and the safety parameters.

Highlights:

TE D

Keywords—IEC 61508, PSM, plasma safety simulation and management, plasma device design, safety oriented design.

- Engineering design of proposed experiment of plasma generation system (EPGS) - Safety analysis of EPGS

- Apply IEC61508 on safety design of plasma devices

EP

- Simulation of EPGS safety system using Simulink

- Virtual Simulation and Experimental verification of plasma generation case study



 

Ma M 

Q qe

Definition Magnetic field Electric field Energy Planck constant (6.626·10-34 J·s)

AC C

Symbol
 E ℎ

Nomenclature Symbol R   

Current density

Boltzmann constant (1.380·10

-23

J/K)



Total mass

  e

Mach number Molecular Mass Pressure Rate of heat exchange between particles Electron charge (1.602·10-19 C)

1

Definition Universal Gas Constant time temperature Velocity of particle or fluid Permittivity of free space (8.854·10-12 F·m-1) Heat Flux Permeability of free space (1.256·10-6 H/m) Plasma conductivity Density Electron temperature

ACCEPTED MANUSCRIPT

1. Introduction and Background

AC C

EP

TE D

M AN U

SC

RI PT

Plasma safety and handling is one of the critical aspects for the modern manufacturing and energy industries. It is important to consider it by evaluating and analyzing the risks associated with plasma devices. This article focuses on plasma safety in the experimental environment and aims to provide a design and procedures for designing plasma devices. It is possible to reduce certain risks and to prevent hazards by making wise design decisions, following safety standards, and by the rigorous collection of experimental data. An excellent example of practical safety framework is the Process Safety Management (PSM) purposed for safe maintenance, integrity, and hazard prevention of the systems, materials, and energy related practices (Canadian Association of Petroleum Producers, 2014). Furthermore, the PSM is in close cohesion with the IEC 61508, an internationally recognized standard for the safe operation of the Equipment Under Control (EUC), where the control systems are based on the proper operation of Electrical/Electronic/Programmable Electronic Systems (E/E/PES) safety related systems and external risk reduction facilities. The EPGS will be largely based on the principles of PSM and the IEC61508. The reason for designing EPGS is to create an experimental setup for performing small scale plasma experiments and to study the plasma behavior under a variety of different conditions such as the plasma pressure, the electron density, plasma propagation velocity, plasma laser interactions, and plasma electron interactions. By studying these scenarios and creating case studies for plasma behavior in different situations it will be possible to make milestones in plasma safety and assign the limits to serve as guiding principles for the engineers and designers involved in plasma device design. The objective of the current work is to give the highlights of PSM and IEC 61508, and to apply them to the design of the EPGS. The purpose of conducting the PSM using the IEC 61508 is to minimize the risks and hazards of the designs that are their initial stages of conception, to make a set of rules for the safe operation, to create a list of possible risks and hazard, and to present a function that highlights the probability of risks associated with EPGS. Firstly, the overview of the EPGS design will be conducted, then the design of the target safety system for EPGS will be provided. The input and the output parameters, as well as the possible measurement techniques for the safe operation of EPGS will be highlighted so as to demonstrate the fault propagation scenarios for EPGS in accordance with the design requirements. Also, the safety limits, measurement, and protection barriers will be defined. The more complex a device, the more likely that some part of it will fail and the tracking of such failure will be made all more complex, and time consuming, unless there is a well-defined fault prevention system in place (Chiles, 2001). If the failures are expected as a result of intentional design, or as a feature leading to prevention of a bigger failure, then accounting for such feature makes it an important part of the design. However, there are failures that occur not by intention and lead to unexpected malfunction, and these are needed to be the focus when dealing with fault prevention in the design stage. In the case of the plasma devices the capacity for failure is in relationship with the main objective of plasma generation (T. J. Dolan, 2000). The fact that plasma devices incorporate the effect of electromagnetic systems and thermo-fluidic systems is critical during the design stage, especially in the field of plasma safety (Inan & Golkowski, 2011). Furthermore, this fact is generally accepted on the grounds that any new design can be evaluated as the sum of its composing parts (IEEE Standards Board, 1991). This approach is not wrong, though, it may lead to missing the net effects of plasma such as the generation of unwanted microwave frequencies, noise, residual charge carriers, and a spectrum of other effects that are entirely dependent on the specific purpose of the plasma device (Romanelli, 2011). The large portion of the plasma devices that exist in the market are oriented on for a well-defined objective use and have a limited scope of operation (M. F. da Silva, 2013). Nonetheless, in the nearby future it is very likely that the technological growth will incur complexity in the plasma devices and this may incur some underlying risks associated with plasma (Pasternak, 2000). It is worth-while to point out that in the present day the manufacturing of the digital devices is highly dependent of the plasma vapor deposition processes (Han, Lebchenko, & al.; M & I., 2013), the medical facilities utilize plasma disinfectant chambers for keeping sanitary environment (Plasmatreat, 2013), the success of the military operations may rely on the effectiveness of the communication networks, and radars, that operate via the plasma effects (J & G., 2009), and the energy production in the near future will be able to prevent the effects of pollution (Rosinski, 2011), possibly tap the nuclear fusion (S., 2011), and propel humanity to the exploration of the solar system (Bussard, 2006). Therefore, there is a need to seriously consider what can be done today to arm the future plasma engineers with the right tools and relatable case studies. The approach to the plasma technology design is only going to get more complex as it will involve a multitude of orchestrated multi-physics events actuated by systematically organized parts. The reciprocity of plasma with the design will inevitably lead to essential application of the safety principles and the enforcement for the risk accountability (Felton, 2011). This is not to say that the foundations of safety-oriented design currently implemented for the novel technologies is irrelevant, though it is reasonable to expect that alterations and the inclusions of new sections particular to specific technological developments in plasma will be taken into account.

2

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

The process safety needs to be implemented in the very early stages of the engineering design. The PSM is achieved by outlining clear instructions for operation and maintenance in order to verify the asset integrity and to prevent any major incidents. Leadership and culture are the center of safety management for the reason that active leadership for the safe practices exemplifies the importance of maintaining a safe environment (Canadian Society for Chemical Engineering, 2012). Upon establishing the safety leadership, the appropriate information needs to be provided for the engaged parties to have competence in PSM. The demonstration of full competence in the process and its risks enables the involved parties to take timely actions in order to prevent hazardous events. The common aspects of PSM are: 1.Hazard Identification (noting the possible risks of equipment operation); 2. Equipment Integrity (ensuring that there are no broken components in the system); 3. Chemical Hazard Data (providing the documentation and labeling for the chemicals involved); 4. Inspections and Audits (regular checking of the system performance); 5. Documentation Systems (a maintained and regular track record of the system operation with specified times and operation procedures). Furthermore, there are four fundamental pillars of PSM. The first pillar is a commitment to process safety through process safety culture, accountability for objectives and goals, compliance with standards, codes, and regulations, active involvements of engaged parties, as well as the stakeholder outreach. The second pillar encompasses the understanding of hazards and risks by familiarization with process safety information and documentation, the ability to identify hazards, and analysis of possible process risks. The third pillar is the risk management through follow-through with operating procedures, safe work practices, design quality and performance assurance, safe design implementation at the design stage of the processes, as well as checking the asset integrity and reliability and providing regular maintenance. The risk management is also achieved through contractor management, training of personnel, pre-start-up safety review and the emergency planning and response. The knowledge and the associated trade secrets are the foundations of the fourth pillar of PSM because it focuses on learning from experience by conducting incident investigations, measuring the existing safety parameters, performing periodic safety audits, and reviewing the management procedures for process safety enhancement. Having the knowledge of PSM is crucial when approaching a novel design, and albeit its evident and straightforward structure it is important to follow it thoroughly. It should be naturally obvious that zero risk is impossible even when all the safety precautions are in place. The safety must be implemented from the very start of the design process because it helps the designer to plan for the end-user safety and to minimize the non-tolerable risks. IEC 61508 is intended for safety engineering and management from the very conception and to the decommission by considering the aspects of the system itself (i.e. the human factors and the safety cases/proofs are not part of the standard) (MTL Instruments Group, 2002). IEC 61508 helps in implementing the safe design and it is composed of 7 sequential and integrated parts: 1. Development of the Overall Safety Requirements (General); 2. Realization phase for E/E/PES safety-related systems requirements; 3. Realization phase for safety-related software requirements; 4. Definitions and abbreviations; 5. Examples of methods for the determination of safety integrity levels (SILS); 6. Guidelines on the application of Parts 2 and 3; 7. Overview of techniques and measures. Parts 1 through 3 are critical for the safety design of new E/E/PES and EUCs. Part 1 (General Requirements) defines the activities to be carried out at each stage of the overall safety lifecycle, as well as the requirements for documentation, conformance to the standard, management and safety assessment. Part 2 (Requirements for (E/E/PES) Safety-Related Systems) and Part 3 (Software Requirements) interpret the general requirements of Part 1 in the context of hardware and software respectively. Parts 4 through 7 are guidelines and examples for development and are primarily suited for evaluating the potential designs (EXIDA, 2006). It is important to note that IEC 61508 is not just a technical guideline, it is a fully implementable version of PSM intended for the E/E/PES and corresponding EUCs. The risk reduction is achieved via the “Safety Function”, which are based on the understanding of the risks associated with the system (Smith & Simpson, 2004). In order to develop the safety functions, the following steps need to be taken: 1. Identify and analyze the risks; 2. Determine the tolerability of each risk; 3. Determine the risk reduction necessary for each intolerable risk; 4. Specify the safety requirements for each risk reduction, including their SILs; 5. Design safety functions to meet the safety requirements; 6. Implement the safety functions; 7. Validate the safety functions. According to IEC 61508, the purpose of the overall safety life-cycle is to force safety to be addressed independently of functional issues, thus overcoming the assumption that functional reliability will automatically produce safety. Phases 1 and 2 indicate the need to consider the safety implications of the EUC and its control system, at the system level, when first they are conceived. In Phase 3, the risks are identified, analyzed, and assessed against tolerability criteria. In Phase 4, safety requirements for risk-reduction measures are specified. In Phase 5 these are translated into the design of safety functions, which are implemented in safety-related systems, depending on the selected manner of implementation. Phases 6, 7, and 8 provide the overall planning of Operation and Maintenance, Safety Validation and Installation and Commissioning, respectively. Phases 9, 10, and 11 need to showcase the realization of Safety Related E/E/PES, Other technology safety related systems, and the external risk

3

ACCEPTED MANUSCRIPT

2.Proposed Approach

RI PT

reduction facilities, respectively. In Phases 12, 13, and 14 outline the process of carrying out the functions of installation and commissioning, safety validation, and operation and maintenance, to be on the overall systems, regardless of the technologies of the safety-related systems. Phases 15 and 16 cover later modification and retrofit of the system and decommissioning, respectively. Given that EPGS will be primarily based off the existing technologies it will be evaluated on the principles of the components. The installation of EPGS is similar to the pressure vessel and has the perceived hazards of implosion, electric shock, EM radiation, X-ray radiation, gas dissipation, explosion, heat, audible noise, and laser light.

AC C

EP

TE D

M AN U

SC

To have safety oriented design of EPGS it is critical to work with the key defining parameters of EPGS and to use them in order to set the safety measurement limits, to define the safety protection barriers, and to make a deterministic representation for the real time safety verification. The probabilistic representation for plasma instabilities can be made by consideration of the design features and the associated component properties and by this manner can be expressed as a fault propagation tree-diagram. The design safety can be achieved for the plasma devices by following through with a design approach outlined in the following flow chart. This chart provides a process for the design approach with intrinsic focus on safety, where the design features are considered to be approached with an engineering design grounded in physical and safety foundations. This approach makes sense when the key tool for the design validation is tested. In present work, the first column of the chart is highlighted in the method of approach, and the second and third columns are used in the section on results and discussions respectively. To proceed with the implementation of the safety protection layers in EPGS it is worthwhile to outline the design features of the proposed design. EPGS is an enclosed chamber that contains two plasma generators facing each other, and includes a vacuum pump, three electron beam generators, laser and optics systems, HV generator, high current generator, and a set of diagnostic tools for the analysis of the plasma and its dynamics. The objectives of EPGS include the studies of the plasma modes, plasma generation options, stability, methodologies for control, novel diagnostic methods, interaction of plasma with materials, plasma mixing, and the plasma wall interaction. It is worthwhile to highlight some of the key components of EPGS in order to understand the how EPGS meets its objectives. The figures below will point to the specific components of EPGS as well as highlight the points where a risk of safety breach are possible. The components of EPGS in the figure are indicated by the arrows: blue is electron gun; orange is a hysteresis magnetic coil; yellow is a control magnetic coil; red is vacuum pump connection valve; violet a quartz view-port; and green is a port for high power laser light input. The potential safety breach risk points of EPGS are indicated by stars: (a) Hot, HV area, and electron generation; (b) Magnetic field for electron trajectory control; (c) View-port dislodging/cracking, possible leakage area, explosion/implosion hazard (EIH); (d) Possible Leakage Area (PLA), EIH; (e) PLA, EIH; (f) Chamber Strain; (g) View-port dislodging/cracking, PLA, EIH. Figure 3 is a close-up view of the plasma generator in cross-section. The plasma generators of this configuration are located on the opposite end of each other (top and bottom inside the actual EPGS chamber). The components are presented in figure 4 by the color-coded arrows: blue is a theta-pinch flat coil; orange and yellow are electrodes for alternate polarities; red is an isolator and a puff control mount for the conductors; green points to a gas supply prepuff chamber; violet is a working gas supply tube; grey is quartz window for laser beam source (green arrow in figure 2); white indicates the additional coils for plasma circulation control. The possible safety breach risk points of the plasma generator are indicated by stars: (a) PLA and quartz cracking points, EIH; (b) Current supply for the magnetic and theta coil control;(c) Laser beam input area (plasma laser coupling); (d) Plasma discharge area, hot zone; (e) Puff pre-mixing zone (possible cause for instabilities). The overall design of EPGS is therefore summarized in the pictures above, though these designs act more as sketches to get the idea of what a potential, final design may look like and what safety-related issues need to be considered. Figure 4 is a detailed view of the electron gun assembly components: blue arrow indicates the copper coupler (e-gun is rigidly attached to this portion); orange arrow point to a precision magnetic coil for electron trajectory control; yellow arrow indicates the hysteresis magnetic loop for stray electron control; red shows a connection point of the electron generation and static control; green points to the thermo-ionic electron generation; and violet arrow show the HV electron trajectory selector. The possible safety breach risk points for electron gun assembly include: (a) PLA and quartz cracking points, EIH; (b) Current supply for the magnetic coil control; (c) HV area for charge accumulation; (d) Thermo-ionic electron generation, hot area and x-ray source.

4

ACCEPTED MANUSCRIPT

Breakdown design requirements into engineering tasks

Use the safety data to limit the options of the design and to prevent dangers

Use plasma physics foundations that meet the objectives of the design

Create concepts and models of plasma device in CAD as a virtual prototype (STP file)

SC

Outside the limits

Use multi-physics tools for plasma device simulation

Assign respective equations to the bodies of interest

TE D

Follow through with proper meshing of the device and respective components (UNV format)

M AN U

Use PSM and IEC61508 to isolate the scenarios and cases in the plasma device that may pose possible risks

Define the safety layers and protection boundaries by using a fault propagation tree diagram

RI PT

Obtain Plasma Device Design Requirements

Assign materials to the bodies of interest

Run plasma device simulation

Compare the simulation results with the safety limits

AC C

EP

Indicate the body forces

Set the initial conditions

Set boundary conditions

Within the limits Design safe for physical prototype

Figure 1. Plasma device design flow-chart process using PSM standard and a multi-physics tool.

5

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Figure 2: An over-all view of EPGS

Figure 3: Potential-plasma generator cross-section

6

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Figure 4: Electron-gun cross-section The key input parameters into EPGS include: 1. Electricity for the electron guns, plasma generators, lasers, vacuum pumps, control systems, and diagnostics; 2. Working gases such as Helium or Argon are preferable, since, the use of gases like hydrogen and oxygen have a higher risk of explosion; 3. Laser light may be introduced into the plasma to yield the wake-field acceleration effects as well as for the diagnostic purposes (Benedetti, Schroeder, Esarey, & Leemans, 2014); 4. Accelerated electrons will provide a unique approach to controlling the plasma instabilities in the plasma (Budker, 1967); and 5. RF signals and RF heating are utilized for energizing the plasma as well as the diagnostics purposes. On the other hand, the key outputs of EPGS are: 1. Diagnostic readings of plasma temperature, pressure, polarization, signal absorptivity, propagation velocity, as well as the auxiliary data such as the vacuum pressure, the puff feed rate, and the x-ray count (for safe operation the sensor will detect when to shut the system down); 2. EM radiation may be produced and create some interference; 3. X-rays are possible, although minute, their alleviation is preferable for the lab operation; 4. Heat dissipation will occur at certain points on the equipment and it is best to keep track of the high heat areas and cool them; and 5. Auxiliary charge accumulation may occur, so the system needs to be grounded after the conduction of each experiment in order to prevent charge interferences. The majority of outputs from EPGS are measurements of the plasma parameters because they allow to track plasma properties. These parameters include plasma temperature, pressure, polarization, signal absorptivity, propagation velocity, the vacuum pressure, the puff feed rate, and the x-ray count. The measurement of the plasma temperature, pressure, and polarization will be conducted by the use of the CCD sensors, polarization filters, and the diode IR lasers. The signal absorptivity will be conducted by placing a small antenna array to supply the signals at one corner of the setup and receive them at the other end (on either side of the two opposing chambers). The propagation velocity can be estimated using high speed cameras in combination with the signal absorptivity antennas and the CCD sensors. The vacuum pressure will be read using the pressure gauge for achieving medium vacuum, as well as a Penning cell, in order to measure the high vacuum. The puff feed rate will be determined by the operation of the precision solenoid valves, based on the opening and closing times. Lastly, the x-ray count will be determined using miniature isolated PV cells connected to signal amplification circuit. In order to monitor the plasma behavior these devices will mostly act as active components, and hence relying on these components for the safety maintenance will not be essential but will greatly assist in making deductions about the state of the experimental setup. The key parameters for the safety measurement are the chamber pressure, temperature, the strain on the chamber walls, and the x-rays generated in the process of plasma activity. The pressure inside the chamber

7

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

determines whether the chamber will collapse, or explode based on the amount of available gas, measured via the pressure gauge and a Penning cell. The temperature of the chamber outside surface determines the amount of heat released in the process of plasma creation and will be measured via a thermo-couple. The heat generated can be dangerous for operation of auxiliary systems and it may be a cause the un-necessary strain due to the metal expansion. In the lieu of the pressure and temperature measurement, the strain at the chamber walls depends on the strain of the chamber walls it can be estimated for how long can the chamber be utilized; after a certain number of operational runs the fatigue may cause the chamber failure. The chamber wall strain is simply measured via the strain gauges. Lastly, the x-rays generated are measured via a photo-voltaic detector system. The amount of x-rays generated can be dangerous if it is above the medical maximum, so only a certain amount of runs can be performed at a time depending on the amount of x-rays produced. To manage the hazards and to maintain the equipment integrity the safety system has to take care of the possible risks of EPGS. Certain risks can only be reduced by considering the specific design features, while other risks can be prevented by placing warning systems. Since most of the hazards/risks have been identified above, it is worthwhile to focus on them. The identified risks can be approached with the following possible risk alleviation techniques: 1. Hot areas need to be labeled and cooled; 2. HV areas, in case they are non-isolatable, need to be shielded and labeled; 3. The magnetic field interference also requires proper shielding from outside; 4. View-port dislodging and cracking, possible leaks, as well as explosion/implosion hazards can only be prevented by employment of quality seals and flanges; 5. The chamber strain is eased by using rigid confinement vessel; 6. The charge accumulation and X-ray areas need to have shielding and labeling; 7. Plasma gas puff instabilities, irregular generation, mismatched timing of plasma discharge/gas release can prevented only by the proper selection of quality components and good control systems; 8. Laser beam input area (plasma laser coupling) has to be labeled and handled with care. The identification of the risk areas, their labelling and isolation are essential to a safe design of a plasma device, although they are meant to handle the device with care they may not alleviate all the risks. The above techniques are passive methods for assuring the safe operation of EPGS and it should be evident that the use of active safety components would add an unnecessary layer of risk, since the active components would require a source of energy that can fail and, consequentially compromise the whole system. Nonetheless, in the next section the full scope of EPGS safety representation will be presented to highlight the critical design features that allow for full implementation of PSM and require the attention for design safety assurance.

TE D

3. Fundamental Theory and Calculations

AC C

EP

In the methodology section EPGS was broken down into four primary sub-systems: main chamber structure, plasma generator, the electron gun, and diagnostics systems. There are also the core-support systems that are part of EPGS that include the vacuum pump, electron gun power and plasma generator supplies, and laser systems. The core-support systems are pre-made, unlike the three primary sub-systems of EPGS which need to be custom made. By considering the components of EPGS it is possible to estimate the probabilities of failure for all the inclusive subsystems and classify them in order to keep an account of known risks. Although the probabilistic approach to the formation of a fault propagation tree diagram is based on the experimental data it is fairly deterministic, compared to possible and unknown faults that may occur as a result of particular modes of plasma activity; the probabilistic approach allows to track down faults that occur as a result of design features and not the anomalies of plasma. The protection and safety barriers guard the user from the possible mishaps that may occur as a result of lack of awareness either when around the device or during its use. First and foremost, the frame structure of EPGS has to be solid enough to maintain the mechanical integrity of the structure during its various stages of operation. In addition to the safety measures there is a need to make simple and obvious indicators that light up when the chamber passes through the specific stages of operation: Green: vacuum pump activated for chamber cleansing; Yellow: the auxiliary systems such as the magnetic field generators, the electrostatic systems , the lasers, and the diagnostics are on stand-by; Blue: the plasma puff is released, the laser beam is on, the electrons are projected onto the plasma collision region, and the diagnostic systems are fully engaged; Red: the system is idle and there is an internal error due to the breached safety measurement settings; operation is unsafe either for the experiment or the staff; and, the system is open for maintenance without procedure of proper shut-down. There also needs to be a physical barrier between the people in the lab and the experiment, possibly, a thick plexiglass barrier with a secure lock, to avoid any possible malpractice. The computer connected to the experiment needs to have an access code available only to the staff specifically devoted to the operation of EPGS. Following through with implementation of these barriers and encouraging the safety leadership is practice that will reduce the chances of risk and prolong the

8

ACCEPTED MANUSCRIPT

operational usefulness of EPGS. By considering the above it is possible to organize fault sub categories that are specific to each portion of EPGS as represented in Table 1.

Table 1. Fault categories for EPGS sub-systems Fault Type

Chamber Structure

1.

Number of locations

Dislodging of seals

5 near observation windows

RI PT

EPGS Sub-system

3 near the electron generators 2 near main chamber seals Plasma Generator x2

2.

Chamber cracks and damage

3 chambers

3.

Conduit damage

2 electrodes

2 theta pinch wires

SC

2 confinement wires Gas supply line

Conduits around plasma generator

5.

Flow instabilities

Pre-puff chamber

6.

Over-current

Magnetic confinement coil

7.

Charge accumulation

7 control surfaces

8.

Hot spots

7 control surfaces and thermionic emitter

9.

x-ray source

7 control surfaces

10. CCD sensor failure

3 locations around the chamber

11. IR lasers failure

3 locations around the chamber

12. Polarization filter dimming

3 locations around the chamber

13. Antenna damage

2 at the opposite end of EPGS

14. Pressure gauge failure

2 locations around the chamber

15. Solenoid valve failure

4 locations

16. PV cell failure

7 locations

TE D

Measurement Systems

Hot spot formation

M AN U

Electron Gun x3

4.

9 locations

EP

3 locations

18. Strain gauge failure

19. Vacuum pump failure

1 location

20. Electron gun power failure

3 locations

21. Plasma failure

2 locations

generator

power

AC C

Core-Support Systems

17. Temperature sensor failure

Safety Barriers

22. Laser systems failure

1 location

23. Plexiglass cracks

5 locations

24. Safety lights failure

4 locations

25. Lock failure

1 location

26. Control computer failure

1 location

The accounting for the parts and components can be related to the safety structure of EPGS using a fault tree diagram as represented in figure 5. The initiating events are as follows: A. Loose tightening of flanges; B. Seal breakage; C. Physical damage to structure by strain or impacts; D. Improper installation or manufacturing; E. Plasma anomaly aftereffects; F. Breach of limitations. The consequences of the initiating events are fault types in table 1. In the figure the initiating events can lead to a failure in EPGS and considering how they can impact the performance it is possible to create a strategy that can reduce the amount of risk to the researchers. Multiple fault scenarios can lead to fault

9

ACCEPTED MANUSCRIPT

RI PT

escalation scenarios where one particular failure can lead to an array of faults that can occur at the later stages of EPGS operation. Some of the expected fault escalation scenarios, for instance: 1. Cracked port-hole window leads to breach of vacuum system that causes the contamination of the chamber systems through oxidation and can result in malfunctioning of plasma generator and thermal ionization components that can cause possible overheating and explosion; 2. Magnetic field interference from outside sources can interfere with electron path and cause plasma instabilities that can lead to damage electronic components and diagnostic systems; 3. Charge accumulation can create plasma gas puff instabilities due to irregular generation and a mismatched timing of plasma discharge which likewise can cause damage to electronic components and diagnostic systems

SC

EPGS Failure

6 1

2

3

23

25

M AN U

4 7

8

9

12

13

E

5

A

B C

10

11

15

D

19

20

21

16

14

17

18

24

26

F

22

AC C

EP

TE D

Figure 5. Fault tree representation of EPGS. Thus far, the faults expected from EPGS are purely based off the design features and they do not cover the anomalies that could possibly happen as a result of plasma anomalies. This matter is dangerous because even though the design of EPGS on its own may be safe, during the actual operation for plasma testing the plasma activity may lead to some effects in electromagnetic interference or possibly produce energetic electrons that can create a shower of x-rays onto un-suspecting researchers. To add a layer of safety for EPGS it is necessary to run computer simulations that will showcase the plasma behavior and allow to set the test limits for the halt on the experiment, Elmer FEM was used to model some plasma case studies by the research group of current work. The importance of familiarizing engineers with the plasma phenomena and its risks is not only in the matter of plasma being a hot ionized gas that can violently react depending on the surrounding chemicals, it is also able to interfere with sensitive electronics and lead to irreparable damage. For modelling of plasma devices, the use of Elmer FEM requires some fundamental understanding of plasma physics, MHD, and the property interactions of between the plasma and the boundaries. The full set of a steady state MHD flow equations are the following:

10

ACCEPTED MANUSCRIPT

(1a)

 2 +  ∙ ∇ = −∇ + ∇&  − ∇ ' ∇ ∙ * + + (, +  × .),  3

(1b)

E 2 +  ∙ ∇E = ∇ ∙ 0 + 1∇ ∙  − ' ∇ ∙ *2 ∙  + + , ∙ ,  3

(1c)

. . 

(1d)

SC

∇×,= −

RI PT

 +  ∙ ∇ = 0, 

AC C

EP

TE D

M AN U

Equation set 1 is the equations of flow continuity, flow momentum, energy, and Faraday’s law, respectively. The explanation of the equations is to be understood by the reader (Howard, 2002). The reductive form of the turbulent flow under the simplified conditions does not make it easy to compute it along with the equations for the plasma behavior. Hence, although the methodology taken in modelling the plasma behavior implements the theoretical foundations presented, solving the Navier-Stokes equations with the MHD coupling terms even for the steady state flows would become an intensive mathematical challenge of solving complex partial differential equations. In order to avoid this cumbersome path the application of FEM for computing the fluid-dynamic changes at very fine meshes will be implemented through Elmer FEM as it can closely produce a solution via the use of a Large Eddy Simulation (LES) technique, which has the capacity to compete with the direct numerical simulation (DNS), provided it is done correctly (Shepherd, 1965; White, 2008). The LES technique makes approximations, unlike the DNS method, though these approximations are based on the rigorous data sampling of actual flows. By and large, the LES method takes into account the physical phenomena and implements the ODEs with approximations based off the measurement of eddies in the experiment, thereby making the equation value models work together with the real-world data. The coupling to electric and magnetic fields can be expressed through a variable for external force acting on the fluid (Shepherd, 1965). In electro-kinetics the fluid may have charges that are coupled to external electric fields and lead to an external force. The charge density may be a variable, depending on how quick the ions are generated and removed, although in steady state it can be approximated to be constant. Also, if the fluid has ionized species it will also couple with the magnetic field (Goedbloed & Poedts, 2004). The diffusion-advection-convection effect in flow is the transport of scalar quantity of ionized species by diffusion, advection, and convection. The difference in the nomenclature usually indicates that an advected quantity does not have an effect on the velocity field of the total fluid flow, and a convected quantity has (Raback, Malinen, Ruokolainen, Pursula, & Zwinger, 2015). The diffusion-advectionconvection effect is derived from the principle of mass conservation of each species in the fluid mixture and may have sources or sinks either at the boundary conditions or within the body of plasma. Hence, multiple advectiondiffusion equations may be coupled together to analyze plasma interaction in mixtures and convective flows at different temperatures. If the velocity field is zero, then the advection-diffusion equation reduces to the diffusion equation, which is applicable only in solids (Dinklage, Klinger, Marx, & Schweikhard, 2008). For a feature of the plasma generator, the modelling of the direct discharge plasma generation uses the multi-physics Elmer FEM toolboxes based on the estimations of the plasma properties. The key property to modelling the plasma discharge is the plasma conductivity. Provided all the initial and boundary conditions were set correctly the limits of the electron temperature, the current density, and the electron density will be determined by this property. For the sake of modelling with available computing power, it was assumed that the plasma is quasineutral, and that the resistive modelling of plasma is applicable. The conductivity computation for the plasma can be roughly estimated as a ratio of current passing between the surfaces at ionizing potential. The simulation critical variables known prior to the configuration of the experiment and the simulation include the current in the arc discharge (I=0.152 A), the potential (V=9000 V), the reference pressure (Pref=7180.8 Pa =53.8604 Torr), and the 6 9 RMS mass transfer coefficient between the electrons and ions: 7 ∙ = 1.36 ∙ 10<= (Dinklage, Klinger, Marx, & 68

√&

Schweikhard, 2008). Based on the current and the potential it is possible to attain an approximate ion-electron

11

ACCEPTED MANUSCRIPT

generation rate, using the Stoletov constants (ηopt-Air = 81 eV/ion-electron pair)) and the following relation, Where, qe is the fundamental charge unit 1.602·10-19 C (Roth, 1995): @∙A >? = (2) ηCDE
RI PT

In Elmer FEM, the concentration at the boundary is expressed as a ratio to the number of available neutral particles. In the current case, the rate of neutral particles entering the chamber corresponds to the flow rate found to be 2.47·1020 particles/s. The ratio of these rates yields a value that is interpreted by Elmer for computing the ion behaviors in the plasma 0.426415 ion/particles. The concentration flux is the other parameter necessary to compute the plasma behavior, it is approximated at the emitter to roughly correspond the number of ions passing through the area confined by the glass (ceramic) chamber Flux Area = 1.2566·10-5 m2. The equation for concentration flux is the mass flow of ions, JK , through an area : JK UJK ∙ VK = (4) LMNO_QRST LMNO_QRST

M AN U

SC

Where, UJK is the count of propagating ions and VK is the mass of each individual atom, in a case where the ion speeds are far below that of the speed of light. For the direct discharge mode, the concentration flux is computed to be 0.3971kg/m2 as required to set the boundary conditions in Elmer FEM. If plasma is unaffected by the flow the theoretical plasma temperature can be estimated using the Saha equation (Inan & Golkowski, 2011) to yield 23365 K or 2.014 eV. This value showcases an upper limit to the ideal quasi-neutral plasma state provided that the species have indeed ionized fully and neither any external flow nor any heat dissipative sinks are present. The computation of the expected current density reading range and the expected Joule heating parameter are based on the Elmer FEM models manual (Raback, Malinen, Ruokolainen, Pursula, & Zwinger, 2015). Although, the plasma conductivity can be obtained from the experiment for the direct discharge configuration, the attainment of this parameter for the single emitter condition is made significantly easier by approximating the conductivity based on the potential available at the emitters and the emission current.

AC C

EP

TE D

4. Results and Discussions

Figure 6: Schematic layout for direct discharge plasma experiment.

As a case study leading to the creation of EPGS consider the above device schematic for a plasma generation experiment. The most notable experiment risks in this configuration include the use of high voltage and high current electrical equipment under vacuum conditions. The corresponding risk components needed to be handled in the experiment include a 9000 V transformer and 3.8 litre vacuum chamber. As in case of EPGS it is possible to break down this experiment according to the fault tree representation as presented in figure 7, which is a simpler form of figure 5 due to the removal of certain components.

12

ACCEPTED MANUSCRIPT

4 1

2

3

23

17

SC

14

RI PT

Experiment Failure

M AN U

F

5

A

B C

19

21

D

Figure 7. Fault tree representation of direct discharge plasma experiment.

AC C

EP

TE D

The simulation of the device was conducted using Elmer FEM with the following figures indicating the performance of the plasma generator in the direct discharge mode. The figures below are representative of the results for the direct discharge and include the properties of plasma such as temperature, the ion-to-neutrals concentration ratio, current, and the magnetic strength.

Figure 8. Temperature profile in Kelvins.

13

SC

RI PT

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

Figure 9. Ion concentration ratio profile from 0% to 100%

AC C

Figure 10. Current per unit length of arc profile in A/m.

Based on the experimental findings and the computational models it is possible to draw some conclusions to the matters of using Elmer FEM for simulations of the plasma generator. It appears to be possible to implement the computer models and produce working concepts that are able to closely match the results of an experiment. However, several pitfalls of using equations prior to setting of an experimental model need to be addressed as it may become evident that there are several discrepancies between what has been computed, and what has been found via an experiment and simulations.

14

SC

RI PT

ACCEPTED MANUSCRIPT

M AN U

Figure 11. Magnetic strength profile in mT.

AC C

EP

TE D

Figure 12 presents the plasma generating end of the experiment. This nozzle and probe configuration has to be preassembled onto the vacuum chamber lid prior to the conduction of any experiments. Respectively, there is a safety procedure for working with this configuration.

Figure 12: Plasma generator assembly

It is worthwhile to consider what does the experimental data implies with respect to the limitations of the triple Langmuir probe. The maximum error in the case of the triple Langmuir probe can reach 15% of the measured value. Therefore, there is a need to consider the average value in terms of the maximums and minimums, as well as the results obtained from Elmer FEM, as indicated in table 2. Table 2. The plasma properties for the direct discharge configuration measured using TLP. Method Te[eV] Jplas[A/m2] ne [m-3] Direct Discharge Min. 0.03642 1793 0.929*1020 Direct Discharge Ave. 0.04285 2109.4 1.093*1020 Direct Discharge Max. 0.04928 2425.8 1.257*1020 Simulation Results 0.04327 12095.8 1.3*1020

15

ACCEPTED MANUSCRIPT

RI PT

Provided that the measurement using the triple Langmuir probe do not yield the exact values, it is useful nonetheless to have a range of values at minimal and maximal errors. The reason is that the use of the simulation model may not yield the exact solutions either, so having a range of approximate results certainly helps in tuning of the simulation model to attain reliable engineering design. Furthermore, although the equipment setting that was used for the experimentation is not of the level of a high budget lab, it does the job it was set out to do. The reasons for the possible deviations between the experimental and simulation results can include: 1. Secondary emission as a result of AC discharge; more electrons are generated as a result of alternating electric fields. 2. Spread of the electron flow onto the Langmuir probe; the electrons disperse and create a differentiating ow of ions. 3. Splattering effect from the walls; the electrons bombarding the back wall reflect back at the Langmuir probe due to the proximity of the wall and the secondary emission effects. 4. General flow effects; flow simply reflects back at the Langmuir probe due to the proximity of the wall.

M AN U

SC

It is worth noting that the differences between the results obtained by experiment and from the simulation are fairly small. The area of interest is near the outlet of the generator and in all cases it is relatively uniform. Nonetheless, the current density computed using the Elmer FEM far exceeds that of the current density obtained from the experiment. The reason for this particular deviation is in the fact that the Langmuir probe was picking up the current in the zone carried by the flow and outside the actual arc discharge. Also, the current in Elmer FEM is representative of the electron flow; so even though the values for the temperature and the ion concentrations were in the range reasonable with the experiment, the current densities do not match because Elmer FEM indicated the electron flow, unlike the ion flow detected by the TLP. By keeping in mind this distinction between the experiment and the simulation it is understandable that Elmer FEM can be used to model the plasma phenomena well. The highlights of how the process presented is representative of EPGS only on a smaller scale is understood by considering the similarities to EPGS in terms of components involved and their relation to the wider scope of PSM that went into devising of EPGS. Furthermore, the layer of simulation of the plasma effects provided a deeper insight into the anomalies due to plasma rather than the rigid representation by the features of the design alone.

TE D

5. Conclusions and Future Work

AC C

EP

The nature of EPGS is to be an experimental unit for observing the plasma behavior under controlled conditions. In order to maintain the controlled conditions, however, a high degree of prudence needs to be maintained in order to have quality data from experimentation. If one of the system components is not able to function properly then the system has to shut-down and re-calibrated before further operation. The implementation of various operational conditions on plasma can lead to certain plasma instabilities, if the experiment is intended for such a purpose, and the system on its own must not fail both structurally and performance-wise. In case of the EPGS there is no current method to reduce the dynamic failures. It is best to keep the experiment error free and prevent the failures before they have a chance of occurring by pre-checking the system operation before each run by having a monitoring system that checks the system status and indicates where a possible failure may occur prior to the experimentation, especially through the process of simulation. If any of the experimental scenarios include the unstable behaviors that may occur due to the plasma nature it is best to take the precautions by adjusting the plasma controlling parameters to the safe operation range for sustaining the plasma instabilities but without going beyond the extent of the EPGS operational range.

Acknowledgement

Authors would like to thank members of the Energy Safety and Control Lab (ESCL) – UOIT, and GPROSYS Corp. for their support to this research.

References Benedetti, C., Schroeder, C. B., Esarey, E., & Leemans, W. P. (2014). Plasma wakefields driven by an incoherent combination of laser pulses: A path towards high-average power laser-plasma accelerators. Physics of Plasmas, 21(056706).

16

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Bondarenko, D., & Gabbar, H. A. (2015). Risk Assessment of High Density Plasma Experimentation. International Journal of Latest Research in Science and Technology, 4(5), 126-135. Budker, G. (1967). An Effective Method of Damping Particle Oscillations in Proton and Antiproton Storage Rings. Atomnaya Energyiya, 22(5), 346-348. Bussard, R. W. (2006). The advent of Clean Nuclear Fusion: Superperformance Space Power and Propulsion. 57th International Astronautical Congress. Valencia. Canadian Association of Petroleum Producers. (2014). Process Safety Management: Regulatory Scan. Ottawa: CAPP. Canadian Society for Chemical Engineering. (2012). Process Safety Management Standard. Ottawa: Canadian Society for Chemical Engineering. Chiles, J. R. (2001). Inviting Disaster. New York: Harper Business. Dinklage, A., Klinger, T., Marx, G., & Schweikhard, L. (2008). Plasma Physics. Berlin Heidelberg: Springer. EXIDA. (2006). IEC 61508 Overview Report. Sellersville: Exida. Felton, B. (2011). Safety study IDs leading causes of accident. Winchester: InTech: Morn Hill. Goedbloed, H., & Poedts, S. (2004). Principles of Magnetohydrodynamics With Applications to Laboratory and Astrophysical Plasmas. Cambridge: Cambridge University Press. Han, Z. J., Lebchenko, I., & al., e. (n.d.). Plasma nanofabrication and nano materials safety. Journal of Physics and Applied Physics, 44, 1-8. Howard, J. (2002). Introduction to Plasma Physics. Canberra: Australian National University. IEEE Standards Board. (1991). IEEE Recomended Practice for the Design of Reliable Industrial and Commercial Power Systems. New York: IEEE Inc. Inan, U., & Golkowski, M. (2011). Principles of Plasma Physics for Engineers and Scientists. Cambridge: Cambridge University Press. J, P. S., & G., E. J. (2009). Microplasm a Lighting: Microcavity Plasma Arrays for Future Lighting Applications. IEEE 36th International Conference on Plasma Science (ICOPS). San Diego. M, K., & I., B. I. (2013). Plasma Engineering: Applications from Aerospace to Bio and Nanotechnology. London: Academic Press. M. F. da Silva, e. a. (2013). Electric Equivalent Model for Induction Electrode-less Fluorescent Lamps. IEEE Transactions on Power ELectronics, 28(7). MTL Instruments Group. (2002). An Introduction to Functional Safety and IEC 61508. Brampton: MTL Instruments Group. Pasternak, A. (2000). Global Energy Futures and Human Development: A Framework for Analysis, US Department of Energy Report UCRL-ID-140773. Livermore: Lawrence Livermore National Laboratory. Plasmatreat. (2013). Plasmatreat expands its R&D department, Ingredients & Processing Chain. Ancaster: Plasmatreat. Raback, P., Malinen, M., Ruokolainen, J., Pursula, A., & Zwinger, T. (2015). Elmer Models Manual. Espoo, Finland: CSC IT Center for Science. Romanelli, F. (2011). Nuclear Energy Encyclopedia. New Jersey: John Wiley & Sons Inc. Rosinski, S. (2011). Waste-to-Energy Technology. Electric Power Research Institute. Roth, J. R. (1995). Industrial Plasma Engineering. Bristol: Institute of Physics Publishing. S., D. (2011). Nuclear Energy Encyclopedia: Historical Origins and Development of Fusion Research. Hoboken, New Jersey: John Wiley & Sons, Inc. Shepherd, D. G. (1965). Elements of Fluid Mechanics. New York: Harcord, Brace and World. Smith, D. J., & Simpson, K. G. (2004). Functional Safety, A straightforward guide to applying IEC 61508 and related standards. Oxford: Elsvier Butterworth-Heinemann. T. J. Dolan. (2000). Fusion Research Principles Experiments and Technology. Pergamon. White, F. (2008). Fluid mechanics. New York: McGraw-Hill.

17