Materials Science & Engineering Theses and Dissertations

Permanent URI for this collectionhttp://hdl.handle.net/1903/2792

Browse

Search Results

Now showing 1 - 10 of 250
  • Thumbnail Image
    Item
    INVESTIGATION AND ENGINEERING OF HfZrO2 INTERFACES FOR FERROELECTRIC BASED NEUROMORPHIC DEVICES
    (2024) Pearson, Justin Seth; Takeuchi, Ichiro; Najmaei, Sina; Material Science and Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    This dissertation describes the study of ferroelectric hafnium zirconium oxide (HZO) and its integration into ferroelectric field effect transistors (FeFET). Ferroelectric HZO is uniquely situated for energy efficient, non-volatile memory applications such as FeFETs due to its CMOS compatibility and ferroelectricity at scaled thicknesses less than 10 nm[1]. This work covers material growth of HZO via atomic layer deposition (ALD), as well as electrode metallization (W and Pt) via sputtering and electron beam physical vapor deposition, to optimize ferroelectricity in capacitive structures. Preliminary results show Pt-based devices were sufficient in producing ferroelectric HZO, but had issues in electrode degradation at high thermal processing > 450 °C. In contrast, HZO capacitors in W devices showed drastic improvement in the ferroelectric response reaching remnant polarization values > 40 μC/cm2. To integrate into a FET structure, gate dielectrics (Al2O3 and HfO2) and the 2D semiconductor tungsten diselenide (WSe2) are introduced to the HZO stack. Material and electrical characterization was performed and gave indication of challenges such as: low remnant polarization (<10 μC/cm2), surface roughness (> 20 nm), and high trap characteristics in FeFET modulation. Electrical characterization was performed via variable pulsing, high frequency cycling, current vs voltage, capacitance vs voltage, and polarization vs voltage testing. Challenges such as low remnant polarization, leaky dielectrics, and surface roughness are identified through transmission electron microscopy, atomic force microscopy, and electrical characterization. These challenges were addressed by altering the growth conditions, scaling the thickness of each material, and thermally processing within the bounds of material stability. Upon integration of these various materials into FETs, the challenges of reliability, stochasticity, and consistency were evaluated on through various means of electrically testing such as, variable pulsing, high frequency cycling, current vs voltage, capacitance vs voltage, and polarization vs voltage. A greater depth of understanding of fundamental aspects of these device architectures is required to untangle the complex electrical characteristics of the fabricated devices. Characterization of material properties is performed by transmission electron microscopy (TEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and atomic force microscopy (AFM). Throughout the studies performed in this dissertation, the phase landscape of HZO was investigated on inert Ti/Pt electrodes. While the ferroelectric nature of the HZO was sufficiently explored at CMOS compatible temperatures, yielding remanent polarization values of 20 μC/cm2 and demonstrating multi state memory within Ferroelectric field effect devices (3.5 order of magnitude conductivity change), due to the phase landscape evolution under thermal processing. Higher temperatures were found to be incompatible with the electrode choice as the interdiffusion and breakdown resulted in poor device performance. W electrode HZO capacitors were then used to study the higher temperature ferroelectric devices as well as incorporate scaling of the ferroelectric films to better match the needs of modern device architectures. The optimal ferroelectric films were found to have remanent polarization values > 40 μC/cm2 and when implemented in a FEFET were able to demonstrate a memory window of 6.3 volts, allowing for a large range of modulation for neuromorphic devices.
  • Thumbnail Image
    Item
    Insulating Materials for an Extreme Environment in a Supersonically Rotating Fusion Plasma
    (2024) Schwartz, Nick Raoul; Koeth, Timothy W; Material Science and Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Fusion energy has long been sought as the “holy grail” of energy sources. One of the most critical remaining challenges in fusion is that of plasma-facing materials, even denoted by the National Academies of Science. The materials challenge is particularly acute for centrifugal mirrors, an alternative concept to the industry-standard tokamak that may offer a more efficient scheme with a faster path to development. The centrifugal mirror incorporates supersonic rotation into a conventional magnetic mirror scheme, providing three primary benefits: (1) increased confinement, (2) suppression of instabilities, and (3) plasma heating through shear flow. However, this rotation, which is driven by an axial magnetic field and a radial electric field, requires the magnetic field lines to terminate on electrically insulating surfaces to avoid “shorting” the plasma. This unique requirement presents a novel materials challenge: the insulator must not only resist irradiation and thermal damage, but also be an excellent electrical insulator and thermal conductor that can be actively cooled. To address this materials challenge, the Centrifugal Mirror Fusion Experiment (CMFX) was developed at the University of Maryland. CMFX serves as a test bed for electrically insulating materials in a fusion environment, as well as a proof-of-concept for the centrifugal mirror scheme. To guide the design of future power plants and better understand the neutronand ion flux on the insulators, a zero-dimensional (0-D) scoping tool, called MCTrans++, was developed. This software, discussed in Chapter 2, demonstrates the ability to rapidly model experimental parameter sets in CMFX and predict the scaling to larger devices, informing material selection and design. Assuming the engineering challenges have been met, the centrifugal mirror has been demonstrated as a promising scheme for electricity production via fusion energy. One of the key aspects to the operation of CMFX is the high voltage system. This system, discussed in Chapter 3, was developed in incremental stages, beginning with a 20 kV, then 50 kV pulsed power configuration, and finally culminating in a 100 kV direct current power supply to drive rotation at much higher voltages, creating an extreme environment for materials testing. This work identified hexagonal boron nitride (hBN) as a promising insulator material. Computational modeling (Chapter 4) demonstrated hBN’s superior resistance to ion-irradiation damage compared to other plasma-facing materials. Additionally, fusion neutrons are crucial for assessing both material damage and power output. Chapter 5 details the neutronics for CMFX, including 3He proportional counters, which have been installed on CMFX to measure neutron production. In parallel, Monte Carlo computational methods were used to predict neutron transport and material damage in the experiment. Ultimately, a materials test stand was installed on CMFX to expose electrically insulating materials to high energy fusion plasmas (Chapter 6). Comparative analysis of hBN and silicon carbide after exposure revealed superior performance of hBN as a plasma-facing material. Two primary erosion mechanisms were identified by surface morphology and roughness measurements: grain ejection and sputtering, both more pronounced in silicon carbide. This work advances our understanding of insulating material behavior in fusion environments and paves the way for the development of the next-generation centrifugal mirror fusion reactors. Chapter 7 discusses conclusions and proposes future work. In particular this section suggests some changes that may allow CMFX to operate at much higher voltages, unlocking higher plasma density and temperature regimes for further material testing.
  • Thumbnail Image
    Item
    HIGH-THROUGHPUT COMBINATORIAL EXPLORATION OF QUANTUM MATERIALS AND DEVICES FOR SPINTRONIC AND TOPOLOGICAL COMPUTING APPLICATIONS
    (2024) Park, Jihun; Takeuchi, Ichiro; Material Science and Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    This doctoral dissertation aims to explore via high-throughput methodologies heavy-element-based quantum materials and devices for spintronic and topological computing applications. It is organized into three parts: (1) the development of spin wave devices based on magnetic insulators for magnon spintronics, (2) the search for spin-triplet superconductors based on Bi alloys (Bi–Ni and Bi–Pd) for superconducting spintronics, and (3) fabricating Josephson junctions based on topological insulators for topological quantum computing.The first part of this dissertation is to develop spin wave devices based on acoustically driven ferromagnetic resonance (ADFMR) using magnetic materials, including yttrium iron garnet (YIG). Spintronic devices based on ferromagnetic metals entail Joule heating and energy loss due to the moving of charge carriers. On the other hand, spin waves can be used without resistive losses. ADFMR is an efficient platform for generating and detecting spin waves via magneto-elastic coupling. While numerous ADFMR studies in ferromagnetic metals have been reported, there is no such report on magnetic insulators. This is due to (1) thermal degradation of piezoelectric substrates (e.g., LiNbO3) during the film crystallization (T > 800°C for YIG), (2) reaction between substrate and film materials, and (3) low ADFMR signals due to intrinsically low magnetostriction. The first part of this thesis attempts to address these issues to achieve YIG ADFMR devices by utilizing rapid thermal annealing to minimize thermal damage, a SiO2 buffer layer to avoid unwanted chemical reactions during crystallization, and a time-gating method for enhanced signal-to-noise ratio. YIG thin films deposited via pulsed laser deposition and crystallized by rapid thermal annealing show decent ferromagnetic behavior. YIG devices show exotic angle- and field-dependent absorption features, indicative of ADFMR. The observed ADFMR pattern is consistent with simulations. This result indicates the first demonstration of ADFMR in magnetic insulators. The second part of this work performs combinatorial synthesis of Bi–Ni and Bi–Pd alloys, which possibly show spin-triplet superconductivity. Such spin-triplet Cooper pairing would allow field-controllable spin polarization in superconductors, enabling superconducting spintronic applications. Furthermore, this type of device possibly provides evidence of superconducting pairing symmetries. In Bi–Ni spread study, Bi3Ni acts as a superconducting host material, where the superconductivity is identified to be varied according to two competing mechanisms: carrier doping and impurity scattering. These results can provide useful guidance in studying superconducting materials with stoichiometric defects. In the Bi–Pd spread films, two superconducting phases are identified with maximum Tc of 3.1 and 3.7 K, corresponding to BiPd and Bi2Pd phases, respectively. With Bi2Pd thin films, spin injection devices are fabricated and characterized. The Bi2Pd spin injection device showed unusual pair-breaking behavior where the superconductivity of Bi2Pd is destroyed significantly by unpolarized current injection. These superconducting spintronic studies demonstrate prompt device exploration via combinatorial methods, efficiently providing insight into spin-triplet superconductivity and its applications. Lastly, this dissertation aims to fabricate topological Josephson junctions based on Yb6/SmB6/Yb6 trilayers. SmB6 is a topological insulator characterized by a robust insulating bulk state and topological surface states. Superconducting proximity effects on the topological surface states can generate topological superconductivity, which can be utilized for fault-tolerant topological quantum computing. This dissertation addresses challenges in fabricating topological Josephson devices. With statistical analysis, device failure mechanisms are identified and addressed, allowing for improved design and fabrication. The improved devices showed Josephson junction-like behavior. The junction characterization revealed that 100% of measured samples showed Josephson features with prominent statistical reproducibility, possibly induced by the Klein effect. The dependence of SmB6 dimensions on the junction behavior is also investigated, along with possible proposed scenarios. These results demonstrate that the combinatorial approaches allow for efficient and prompt investigation of novel quantum materials and devices, facilitating phase diagram studies, materials screening, and stoichiometric controls.
  • Thumbnail Image
    Item
    Scalable Rapid Fabrication of Low-Cost, High-Performance, Sustainable Thermal Insulation Foam for Building Energy Efficiency
    (2024) Siciliano, Amanda Pia; Hu, Liangbing; Material Science and Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Bio-based thermal insulation materials offer a promising path towards energy savings in the buildings sector. However, these materials face competitiveness challenges against conventional petroleum-based alternatives due to issues with inferior insulation performance, poor compressive strength, and limited manufacturing scalability. Various fabrication methods such as freeze drying, thermal bonding, and chemical treatment have been proposed to enhance the material’s internal structure by introducing additional pores, creating a more complex path for heat transfer, and improving insulation efficiency. Despite advancements, the manufacturing scalability of these methods and their integration into industrial production remain unachieved.This thesis aims to bridge the gap between laboratory experiments and large-scale production by developing low-cost, sustainable cellulose-based thermal insulation. By investigating both aqueous and non-aqueous-based processing strategies, this work proposes several different fabrication techniques, leading to significant savings in energy, time, and cost. Establishing a comprehensive understanding of the interactions among the fabrication process, insulation foam, manufacturing scalability, and intended product application is imperative. This understanding accounts for variations in processing parameters (e.g., pretreatments, binders, temperature, time) and their impact on the insulation foam’s internal structure and overall performance. By examining the relationship between processing parameters and material structure, this thesis not only advances the fundamental understanding necessary for optimizing fabrication but also provides strategic guidance for selecting and designing scalable bio-based thermal insulation foams. Studying and characterizing commercially viable methods that seamlessly integrate with current industrial infrastructures is crucial for facilitating the transition from small-scale laboratory experimentation to large-scale industrial production. Through various technical strategies, this work illustrates how our understanding can be utilized to offer direction for fabrication method selection, design, and processing, ultimately optimizing the scalable rapid fabrication of low-cost, high-performance sustainable thermal insulation materials for building energy efficiency.
  • Thumbnail Image
    Item
    INTERFACES IN THIN-FILM SOLID-STATE BATTERIES
    (2024) Castagna Ferrari, Victoria; Rubloff, Gary GWR; Material Science and Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    The lack of a diagnostics approach to monitor interface kinetics in solid-state batteries (SSBs) results in an incomplete knowledge of the mechanisms affecting device performance. In this study, a new protocol for process control of SSB interface formation and their evolution during operation is presented. Thin-film SSBs and diagnostic test devices that are composed by a permutation of isolated layers were simultaneously fabricated using sequential sputtering deposition and in-situ patterning using shadow masks. Physics-based electric circuit models were designed for deconvolution of impedance profiles, which enabled an evaluation of bulk properties and space-charge layers at interfaces individually and during operation under different states-of-charge. Relative permittivity values of fundamental battery components (cathode, electrolyte and anode) were calculated as a function of the frequency and the applied voltage. Interfacial impedances, as well as space-charge layers formed at heterojunctions during charge and discharge processes, were successfully deconvoluted using the diagnostic test devices and electric circuit modeling. The cathode-electrolyte interphase was kinetically stable under a voltage window of 0 – 3.6 V vs Cu, and it had an estimated ionic conductivity of the order of 10-9 S/cm, hence it is a localized limiting factor for Li+ transfer. The anode-electrolyte interphase was thermodynamically stable upon completion of the fabrication process, but it became kinetically unstable during charge and discharge cycles. Hence, the proposed diagnostics protocol enlightened the necessity of implementing interfacial engineering on these interphases in the future for improvement of cyclability and stability of SSBs and ionic devices.
  • Thumbnail Image
    Item
    DIRECT INK WRITING SOLID-STATE LI+ CONDUCTING CERAMICS FOR NEXT GENERATION LITHIUM METAL BATTERIES
    (2024) Godbey, Griffin Luh; Wachsman, Eric D; Material Science and Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    The global pursuit of safer and higher-capacity energy storage devices emphasizes the crucial link between the microstructures of electrochemically active materials and overall battery performance. The emergence of solid-state electrolytes featuring multi-layered, variable porosity microstructures presents fresh opportunities for developing the next generation of rechargeable batteries. However, this advancement also brings forth novel challenges in terms of device integration and operation. In this dissertation, solid-state Li-ion conducting electrolytes were 3D printed to enhance the performance of porous electrolyte layers within porous-dense-porous trilayer systems.LLZO-based garnet electrolyte scaffolds were fabricated via 3D printing using direct ink writing (DIW), facilitating the generation of scaffolds with minimal tortuosity and constriction in comparison to previous porous scaffolds manufactured through tape casting. Rheological techniques, including stress and time sweep tests, were employed to characterize the DIW inks and discern their conformal and self-supporting properties. The analysis focused on ink characteristics critical for Direct Ink Writing (DIW), emphasizing properties essential for achieving high aspect ratio printing and minimal constriction in 3D structures. Drawing from this ink research, two distinct 3D architectures, columns and grids, were fabricated. Column structures were utilized in assembling Li-NMC622 and Li-SPAN cells, with detailed discussions highlighting improvements in printing and sintering outcomes. Notably, NMC622, characterized by larger particle sizes, demonstrated complete infiltration within 3D printed porous networks, yielding a promising specific capacity of 169.9 mAh/g with minimal capacity fade. Further optimization involved integrating a porous 3D scaffold to facilitate SPAN infiltration in Li-SPAN cells, resulting in a specific capacity of 1594 mAh/g, albeit with significant capacity fade. The Li-S was implemented into a grid structure achieving 763 mAh/gS with less than 0.25% capacity loss over 16 cycles. Lastly, comprehensive morphology analysis was conducted to evaluate the effectiveness of our optimal DIW structures and to inform future enhancements of such cells.
  • Thumbnail Image
    Item
    ELECTROLYTE AND INTERFACE DESIGNATION FOR HIGH-PERFORMANCE SOLID-STATE LITHIUM METAL BATTERIES
    (2024) Zhang, Weiran; Wang, Chunsheng; Material Science and Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    The demand for advanced battery technology is intensifying as electric energy becomes the foundation of modern technologies, such as smart devices, transportation, and artificial intelligence. Batteries play a crucial role in meeting our increasing energy demands and transitioning towards cleaner and more sustainable energy sources. However, range anxiety and safety concerns still hinder the widespread application of battery technology.Current Li-ion batteries, based on graphite anode, have revolutionized battery technology but are nearing the energy density limits. This necessitates the development of metal batteries, employing lithium metal as anode which eliminates host materials that do not contribute to capacity, thereby offering 10 times higher specific capacity. Recent research on lithium metal batteries has seen a significant surge, with growing knowledge transitioning from Li+ intercalation chemistry (graphite) to Li metal plating/stripping. The electrolyte, which was previously regarded as an inert material and acting as a Li+ ion transportation mediator, has gradually attracted researchers’ attention due to its significant impact on the solid electrolyte interphase (SEI) and the Li metal plating/stripping behaviors. Compared to the traditional liquid electrolytes, solid-state lithium metal batteries (SSLMB) have been regarded as the holy grail, the future of electric vehicles (EVs), due to their high safety and potential for higher energy density. However, there are notable knowledge gaps between liquid electrolytes and solid-state electrolytes (SSEs). The transition from liquid-solid contact to solid-solid contact poses new challenges to the SSLMB. As a result, the development of SSLMB is strongly hindered by interface challenges, including not only the Li/SSE interfaces and SSE/cathode interfaces but also SSE/SSE interfaces. In this dissertation, I detailed our efforts to highlight the role of electrolytes and interfaces and establish our understanding and fundamental criteria for them. Building on this understanding, we propose effective and facile engineering solutions that significantly enhance batterie metrics to meet real-world application demand. Rather than simply introducing new compositions or new designations, we are dedicated to introducing our understanding and mechanism behind it, we hope the scientific understanding, the practical solution, and the applicability to various systems can further guide and inspire the electrolyte and interface designation for next-generation battery technology.
  • Thumbnail Image
    Item
    INTERFACE AND STRUCTURES IN LITHIUM-GARNET QUASI-SOLID-STATE BATTERIES
    (2024) Gritton, Jack Evans; Wachsman, Eric D; Material Science and Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    A confluence of adoption of the internet of things, mobile electronics, electric vehicles, and shift towards adoption of intermittent green energy sources has led to a need for rapid improvement in battery technology in metrics ranging from rate capability and energy density to safety. While significant strides have been made through traditional liquid-based lithium-ion batteries, these oft-conflicting demands require fundamental shifts in battery chemistry, especially enabling safe incorporation of lithium metal anodes. Given their high conductivity, non-flammability, wide electrochemical stability window, and stability to lithium metal, lithium-stuffed garnets of the family LLZO provide one of the most promising alternative electrolytes to replace traditional flammable electrolytes. Two of the largest factors holding back these ceramic electrolytes are interfacial compatibilities and the interplay between processing and electrolyte mass. While drastic improvements have been made in the interface between garnet and lithium metal to improve rate capability, similar jumps in full cells have not been observed for rate and capacity. Using a varied cathode loading and a combination of EIS and DRT, we showed that garnet-catholyte interface was the main contributor to resistance in quasi-solid-state batteries of reasonable cathode loadings utilizing Pyr14TFSI based catholyte. Two methods were then used to improve this interface: modification of the garnet structure interfacing with catholyte, and modification of catholyte composition. Through the use of these methods, rate capabilities and capacity were drastically improved from the baseline system, both at elevated and room temperature. In addition to reducing interfacial resistance, cell polarization can be reduced through using thinner electrolytes. Given its higher mass density and lower conductivity in comparison to liquid electrolytes, garnet has historically had to rely on its greater stability to higher energy density electrodes to maintain competitive energy densities or utilize thin-film procedures that reduce mass but result in orders of magnitude lower conductivity than bulk produced garnets. To balance conductivity, ease of processing, and cell mass, a new combination of bulk-derived processing has been developed that allows for thin free-standing cubic garnet and thin, flexible, porous garnet. Cells using these new thin garnets achieved high cycling rates, and significant capacities.
  • Thumbnail Image
    Item
    CATALYST DEVELOPMENT FOR NON-OXIDATIVE METHANE UPGRADING TOWARDS HYDROCARBONS AND HYDROGEN PRODUCTION
    (2024) LIU, ZIXIAO; Liu, Dongxia; Material Science and Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Methane, the main constituent of natural gas and biogas is deemed to be an alternative source to replace crude oil in the production of chemicals and fuel. Non-oxidative methane conversion enables methane coupling or splitting to produce hydrogen and more significant hydrocarbons, but catalyst deactivation has been a challenge in past research. This dissertation addresses catalyst deactivation issues in non-oxidative methane conversion by inventing novel catalyst systems. For direct non-oxidative methane coupling, a pathway for methane upgrading into hydrogen, olefin, and aromatic products, the silica-supported catalysts were synthesized by flame fusion of a mixture of quartz silica and metal silicate precursors. Compared to the cristobalite silica-supported catalysts reported previously, vitreous silica-supported catalysts have disordered Si-O bonds and structural defects, enabling better metal dispersion and more vital metal-support interaction. The as-prepared vitreous silica-supported iron catalyst had a shorter induction period in methane activation and lower coke yield. The increase in iron concentration elongated the catalyst induction period and promoted aromatics and coke formation. Among different transition metal catalysts, the cobalt supported by vitreous silica had the best methane conversion, hydrocarbon product yield, and catalyst stability. For catalytic methane pyrolysis, a pathway producing COx-free hydrogen and valuable carbon product, a siliceous zeolite-supported cobalt catalyst was invented. In comparison to the methane pyrolysis catalysts in literature, the siliceous zeolite support in the invented catalyst has limited Brönsted acidity and increased mesoporosity, which limited the acid-catalyzed deactivation mechanism and facilitated the mass transport, and thus significantly increased the catalyst lifetime. The cobalt sites change the cluster sizes and coordination structures with the loading concentrations in the zeolite support, which leads to carbon products with different properties.
  • Thumbnail Image
    Item
    LOW TEMPERATURE PLASMA-METAL INTERACTIONS: PLASMA-CATALYSIS AND ELECTRON BEAM-INDUCED METAL ETCHING
    (2024) Li, Yudong; Oehrlein, Gottlieb G; Material Science and Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Low-temperature plasma can generate different types of chemically reactive species at gas temperatures far below what is required to form such species from thermal excitation. Interactions between these reactive plasma-generated species and material surfaces have great potential for various applications, such as semiconductor etching or gas conversion. Synergistic effects, where the production rate with two inputs is greater than the sum of the consequences of each individually, have been demonstrated by combining the plasma with other energy inputs such as heat or kinetic energy from ions or electrons. Understanding the mechanisms by which these species interact with relevant surfaces is vital for the future development of plasma processing, chemistry and physics. In this work, we focus on the interaction of long-lived plasma species, particularly neutrals, with metal. A remote plasma-surface configuration was applied, where the plasma itself does not directly contact the surface. Two examples of plasma-metal interactions will be discussed, one taking place at atmospheric and the other at low pressure. The first case is plasma-assisted catalytic oxidation of methane (CH4) using a nickel (Ni) catalyst at atmospheric pressure, implemented by combining a remote plasma jet. The interrelation of real-time measurements of reaction products and surface adsorbates and plasma diagnostics allowed the identification of atomic oxygen as the key plasma-generated species that drives the synergistic plasma-catalytic reaction. The in-situ characterizations of the surface and gas phase reactions reveal the possible key reaction pathways for the plasma-catalysis reactions. We also observed the activation of the catalyst resulting from long-lasting catalyst surface modification induced by plasma species interaction. The second case is the damage-free etching of refractory metals, ruthenium (Ru) and tantalum (Ta), at low pressure. This was implemented by combining a remote plasma source (RPS) with an electron beam (EB) source. We investigated the effects of CF4 and Cl2 additions to Ar/O2 RPS effluents and we find that Ar/O2 with Cl2 addition induces the highest Ru etch rate (ER) and best removal selectivity over Ta. The surface chemistry characterization by spatially-resolved XPS reveals the possible mechanism of the electrons and neutrals induced materials etching. We also proposed a model that considers the fundamental aspects of the etching reaction and successfully predicts the major features of the electron and neutral induced etching reactions.