Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “SiO2”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

713 records · Page 40

Tin-Essako 001: A Metal-Rich Ureilite?

Introduction: Metal-rich achondrites include a variety of types, and likely have a variety of origins. Models range from gravitational mixing at the core-mantle boundaries of differentatiated asteroids, to complex impact mixing scenarios. We describe a new type of metal-rich achondrite that might be the first metal-rich ureilite. Sample: Tin-Essako (TE) 001 (~4.3 g) was found in Mali in 2020 and purchased by Jay Piatek in 2021. It was classified as a metal-rich ungrouped achondrite, with olivine and oxygen isotope (d18O=8.2810‰, d17O=3.718‰, avg. 3) compositions suggesting affinity to ureilites [1]. We studied one polished section (~187.7 mm2) of TE 001. Petrography: TE 001 consists of ~60% metal and 40% silicates, heterogeneously distributed. The metal is largely fresh, but iron oxides (presumably terrestrial) occur along one edge and in some patches and veins in the interior. The silicates are dominantly olivine (≥90%), with melt-textured areas of plagioclase + Si-rich glass. Minor phases include chromite and carbon. Olivine occurs as rounded grains (up to ~2.5 mm) in metal, commonly with rims of melt-textured plagioclase + glass. Olivine also occurs in more massive areas having a “honeycomb” texture, with rounded olivine “cells” surrounded by an interstitial network of reduced olivine riddled with tiny metal grains, plus melt-textured plagioclase + Si-rich glass. Chromite occurs as subhedral to rounded grains; smaller grains (30-250 mm) are included in olivine and a few larger grains (400-500 mm) are isolated within metal. A carbon phase occurs as lacy-textured rims around olivine grains in metal, or small patches within metal. Mineral Compositions: The olivine (excluding interstitial areas) is Fo 73.9±0.5, with 0.25±0.02 wt.% CaO, 0.31±0.01 wt% Cr2O3, 0.01 wt.% NiO, and molar Fe/Mn=49.8±2.2 (53 analyses). Olivine in interstitial areas has Fo up to at least 91. Smaller chromite grains have Fe# (molar Fe/[Fe+Mg]) = 0.54±0.01, Cr# (molar Cr/[Cr+Al]) = 0.52±0.01, 0.52±0.02 wt.% V2O3 and 0.21±0.04 wt% ZnO (28 analyses). One larger grain is zoned from Fe# = 0.45, Cr# =0.52 to Fe# = 0.40, Cr# =0.57, and contains thin Al-rich lamellae not resolved by EMPA. One irregularly shaped patch of chromite included in olivine has Fe# =0.26 and contains no ZnO. Plagioclase laths are An ~53-60, with ≤0.01 wt.% K2O. Glass contains 75-76 wt% SiO2 and ~16 wt% Al2O3. The metal contains 5.2±0.2 wt% Ni, 0.46±0.02 wt.% Co, and 0.01±0.01 wt.% Cr, with Si and P below detection (168 analyses). Discussion: The olivine + chromite assemblage in TE 001 is similar to the most ferroan ureilites (Fo ~75-79 [2]), as are the oxygen isotopes [1]. The presence of a carbon phase supports this, although the identity of this phase (graphite as in ureilites?) remains to be determined. The “honeycomb” textured areas, in particular, the presence of olivine “reduction rims,” resemble shock-smelted olivine areas in ureilites [3], but interstitial melt-textured plagioclase laths + glass like those in TE 001 have not been reported in such (or any) ureilites. Olivine in TE 001 is marginally more ferroan than in the most FeO-rich ureilite, with Fe-Mg-Mn composition offset from the trend of olivine + low-Ca pyroxene ureilites similar to augite-bearing ureilites [4]. CaO and Cr2O3 contents are in the range of those in ureilite olivine [ ] , though Cr2O3 is at the extreme low end of the range [5]. Chromites (except the unusual one) have similar Fe# to the most ferroan primary chromites in ureilites [2], but distinctly lower Cr# (0.52 vs. 0.71). Metal compositions are with the range for metal in ureilites [7]. The absence of pyroxene and sulfide, and the high abundance of metal in TE 001, are unlike ureilites. One possibility is that a ferroan, chromite-bearing, pyroxene-poor ureilite was invaded (possibly due to impact) by a metallic liquid (low S content suggests very high temperature), resulting in complete melting of pyroxene and smelting of olivine, with rapid recrystallization of melted silicate as plagioclase + glass. Alternatively, a pre-existing metal-rich ureilite assemblage may have been impact melted. Additional types of data will be obtained to evaluate these hypotheses (i.e., is the metal indigenous?) and assess affinity to ureilites.

olivine↗

Machiite, Al2Ti3O9, A New Oxide Mineral from the Murchison Carbonaceous Chondrite: A New Ultra-refractory Phase From the Solar Nebula

Machiite (IMA 2016-067), Al2Ti3O9, is a new mineral that occurs as a single euhedral crystal, 4.4 mm in size, in contact with an euhedral corundum grain, 12 mm in size, in a matrix of the Murchison CM2 carbonaceous chondrite. The mean chemical composition of holotype machiite by electron probe microanalysis is (wt%) TiO2 59.75, Al2O3 15.97, Sc2O3 10.29, ZrO2 9.18, Y2O3 2.86, FeO 1.09, CaO 0.44, SiO2 0.20, MgO 0.10, total 99.87, giving rise to an empirical formula (based on 9 oxygen atoms pfu) of (Al1.17Sc0.56Y0.10Ti4+0.08Fe0.06Ca0.03Mg0.01)(Ti4+2.71Zr0.28Si0.01)O9. The general formula is (Al,Sc)2(Ti4+,Zr)3O9. The end-member formula is Al2Ti3O9. Machiite has the C2/c schreyerite-type structure with a = 17.10 Å, b = 5.03 Å, c = 7.06 Å, b = 107°, V = 581 Å3, and Z = 4, as revealed by electron backscatter diffraction. The calculated density using the measured composition is 4.27 g/cm3. The machiite crystal is highly 16O-depleted relative to the coexisting corundum grain (D17O = –0.2 ± 2.4‰ and –24.1 ± 2.6‰, respectively; where D17O = d17O – 0.52 × d18O). Machiite is a new member of the schreyerite (V2Ti3O9) group and a new Sc,Zr-rich ultrarefractory phase formed in the solar nebula, either by gas-solid condensation or as a result of crystallization from a Ca,Al-rich melt having solar-like oxygen isotopic composition (D17O ~ –25‰) under high-temperature (~1400–1500 °C) and low-pressure (~10-4–10-5 bar) conditions in the CAI-forming region near the protosun. The currently observed disequilibrium oxygen isotopic composition between machiite and corundum may indicate that machiite subsequently experienced oxygen isotopic exchange with a planetary-like 16O-poor gaseous reservoir either in the solar nebula or on the CM chondrite parent body. The name machiite is in honor of Chi Ma, mineralogist at California Institute of Technology, for his contributions to meteorite mineralogy and discovery of many new minerals representing extreme conditions of formation.

Alexander N Krot↗

Adrianite, Ca12(Al4Mg3Si7)O32Cl6, a New Cl-rich Silicate Mineral From the Allende Meteorite: An Alteration Phase in a Ca-Al-rich Inclusion

Adrianite (IMA 2014-028), Ca12(Al4Mg3Si7)O32Cl6, is a new Cl-rich silicate mineral and the Si,Mg analog of wadalite. It occurs with monticellite, grossular, wadalite, and hutcheonite in altered areas along some veins between primary melilite, spinel, and Ti,Al-diopside in a Type B1 FUN (Fractionation and Unidentified Nuclear effects) Ca-Al-rich inclusion (CAI), Egg-3, from the Allende CV3 carbona-ceous chondrite. The mean chemical composition of type adrianite by electron probe microanalysis is (wt%) CaO 41.5, SiO2 27.5, Al2O3 12.4, MgO 7.3, Na2O 0.41, Cl 13.0, O=Cl –2.94, total 99.2, giving rise to an empirical formula of (Ca11.69Na0.21)(Al3.85Mg2.88Si7.23)O32Cl5.80. The end-member formula is Ca12(Mg5Si9)O32Cl6. Adrianite has the I43d wadalite structure with a = 11.981 Å, V = 1719.8 Å3, and Z = 2, as revealed by electron backscatter diffraction. The calculated density using the measured composition is 3.03 g/cm3. Adrianite is a new secondary mineral in Allende, apparently formed by alkali-halogen metasomatic alteration of primary CAI minerals such as melilite, anorthite, perovskite, and Ti,Al-diopside on the CV chondrite parent asteroid. Formation of secondary Cl-rich minerals sodalite, adrianite, and wadalite during metasomatic alteration of the Allende CAIs suggests that the metasomatic fluids had Cl-rich compositions. The mineral name is in honor of Adrian J. Brearley, mineralogist at the University of New Mexico, U.S.A., in recognition of his many contributions to the understanding of secondary mineralization in chondritic meteorites.

Chi Ma↗

Machiite, Al2Ti3O9, a New Oxide Mineral from the Murchison Carbonaceous Chondrite: A New Ultra-refractory Phase from the Solar Nebula

Machiite (IMA 2016-067), Al2Ti3O9, is a new mineral that occurs as a single euhedral crystal, 4.4 mm in size, in contact with an euhedral corundum grain, 12 mm in size, in a matrix of the Murchison CM2 carbonaceous chondrite. The mean chemical composition of holotype machiite by electron probe microanalysis is (wt%) TiO2 59.75, Al2O3 15.97, Sc2O3 10.29, ZrO2 9.18, Y2O3 2.86, FeO 1.09, CaO 0.44, SiO2 0.20, MgO 0.10, total 99.87, giving rise to an empirical formula (based on 9 oxygen atoms pfu) of (Al1.17Sc0.56Y0.10Ti4+0.08Fe0.06Ca0.03Mg0.01)(Ti4+2.71Zr0.28Si0.01)O9. The general formula is (Al,Sc)2(Ti4+,Zr)3O9. The end-member formula is Al2Ti3O9. Machiite has the C2/c schreyerite-type structure with a = 17.10 Å, b = 5.03 Å, c = 7.06 Å, b = 107°, V = 581 Å3, and Z = 4, as revealed by electron backscatter diffraction. The calculated density using the measured composition is 4.27 g/cm3. The machiite crystal is highly 16O-depleted relative to the coexisting corundum grain (D17O = –0.2 ±2.4‰ and –24.1 ± 2.6‰, respectively; where D17O = d17O – 0.52 ×d18O). Machiite is a new member of the schreyerite (V2Ti3O9) group and a new Sc,Zr-rich ultrarefractory phase formed in the solar nebula, either by gas-solid condensation or as a result of crystallization from a Ca, Al-rich melt having solar-like oxygen isotopic composition (D17O ~ –25‰) under high-temperature (~1400–1500 °C) and low-pressure (~10-4–10-5 bar) conditions in the CAI-forming region near the protosun. The currently observed disequilibrium oxygen isotopic composition between machiite and corundum may indicate that machiite subsequently experienced oxygen isotopic exchange with a planetary-like 16O-poor gaseous reservoir either in the solar nebula or on the CM chondrite parent body. The name machiite is in honor of Chi Ma, mineralogist at California Institute of Technology, for his contributions to meteorite mineralogy and discovery of many new minerals representing extreme conditions of formation.

Machiite↗

Adrianite, Ca12(Al4Mg3Si7)O32Cl6, a New Cl-rich Silicate Mineral from the Allende Meteorite: An Alteration Phase in a Ca-Al-rich Inclusion

Adrianite (IMA 2014-028), Ca12 (Al4Mg3Si7)O32Cl6, is a new Cl-rich silicate mineral and the Si, Mg analog of wadalite. It occurs with monticellite, grossular, wadalite, and hutcheonite in altered areas along some veins between primary melilite, spinel, and Ti, Al-diopside in a Type B1 FUN (Fractionation and Unidentified Nuclear effects) Ca-Al-rich inclusion (CAI), Egg-3, from the Allende CV3 carbonaceous chondrite. The mean chemical composition of type adrianite by electron probe microanalysis is (wt%) CaO 41.5, SiO2 27.5, Al2O3 12.4, MgO 7.3, Na2O 0.41, Cl 13.0, O=Cl –2.94, total 99.2, giving rise to an empirical formula of (Ca11.69Na0.21) (Al3.85Mg2.88Si7.23) O32Cl5.80. The end-member formula is Ca12 (Mg5Si9) O32Cl6. Adrianite has the I43d wadalite structure with a = 11.981 Å, V = 1719.8 Å3, and Z = 2, as revealed by electron backscatter diffraction. The calculated density using the measured composition is 3.03 g/cm3. Adrianite is a new secondary mineral in Allende, apparently formed by alkali-halogen metasomatic alteration of primary CAI minerals such as melilite, anorthite, perovskite, and Ti,Al-diopside on the CV chondrite parent asteroid. Formation of secondary Cl-rich minerals sodalite, adrianite, and wadalite during metasomatic alteration of the Allende CAIs suggests that the metasomatic fluids had Cl-rich compositions. The mineral name is in honor of Adrian J. Brearley, mineralogist at the University of New Mexico, U.S.A., in recognition of his many contributions to the understanding of secondary mineralization in chondritic meteorites.

Adrianite↗

An Improved Hf Vapor Etching Apparatus for Stardust Particle Extraction

Introduction: The NASA Stardust mission captured thousands of particles from the Jupiter-family comet 81P/Wild 2 in a collector composed of aluminum foil and blocks of silica aerogel [1]. To date, most Wild 2 particles available for study are relatively large and coherent particles extracted individually from the ends of hollow, carrot-shaped impact tracks produced during impact into aerogel. However, >65% of the impacting mass can be found in the ‘bulb’ of the track, including nearly all of the <1 μm size fraction [2]. This fraction contains organic-rich material and is likely to include presolar grains, representing a critical opportunity to constrain the organic and presolar inventory of primitive outer solar system materials. However, the small size and susceptibility of this fraction to melting or alteration during capture poses significant analytical challenges. Previous attempts to extract and concentrate fine-grained material from the bulb of Stardust tracks have attempted to develop techniques that efficiently destroy aerogel and leave impactor particles relatively unharmed. The low density and high porosity of silica aerogel makes it more susceptible to attack by etching with hydrofluoric acid (HF) [3] or CF4 plasma ashing [4] than collected cometary silicates. Previous studies of HF vapor etching used HF solutions varying between 5 to 49% and noted that at high concentrations and etch rates, a liquid droplet was produced according to the etching reaction 4HF + SiO2 → SiF4 (g) + 2H2O [3]. SiF4 readily decomposes into HF and silicic acid in water, which has the potential to alter the freed cometary silicates. The plasma ashing technique avoids production of a liquid droplet and minimizes damage to embedded silicates but requires specialized equipment [4]. In this abstract, we present recent improvements to the HF etching procedure with the aim of constructing an etching chamber capable of slowly etching silica aerogel using small quantities of dilute HF. The etching chamber can be assembled using readily available materials.

Stardust↗

Thermodynamic Modeling of the Vapor in Equilibrium With Apollo 17 Basalts.

Lunar mare basalts sampled during the Apollo 17 mission provide insight to the chemical evolution of mare basalt magmas and their associated vapor as these samples erupted and cooled rapidly at the lunar surface. We focus on minerals present on vesicle and vug surfaces in Apollo 17 basalts, such as 71036, which was opened as part of the Apollo Next Generation Sample Analysis (ANGSA) program. The mineral assemblage observed includes native Fe, SiO2, and a Mg-P mineral, possibly merrillite [1, 2]. The pet-rographic context of these phases suggests that they may have been deposited directly from the vapor phase after eruption. Here we use thermodynamic modeling to assess the plausibility of the vapor-phase deposition hypothesis.

D. L. Thompson↗

Additional Trials & Tribulations in Synthesizing a Sulfide Standard

Well-characterized sulfide reference materials that can serve as matrix-matched calibrants for in-situ trace element analyses via laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) remain elusive. Here, we describe the creation an in-house sulfide standard at NASA JSC that will be used to measure siderophile and chalcophile trace elements in low pressure experimental products, specifically pentlandite ([FeNi]9S8) and pyrrhotite ([FeNi]1-xS), as well as natural sulfides in terrestrial and meteoritic mineral assemblages. Recent methods in creating a homogenous sulfide standard include pressed pellets, synthesized chips, or fused glass [e.g. 1-3]. Other studies, however, have had success in forming homogenous FeS standards via synthesization methods [e.g. 4-6]. We present a variation of the method described in [4], in which we create a pyrrhotite standard doped with a variety of trace elements (Zn, As, Se, Mo, Ru, Rh, Pd, Sn, Sb, Te, W, Os, Ir, Pt, Au) at ~10-40 ppm and Cu at ~200 ppm. Sulfide compositions were constructed using Fe and Ni metal and elemental sulfur powders. Trace elements were added to elemental sulfur from Atomic Absorption Spectroscopy (AAS) elemental standards as either nitrate or chloride solutions to prevent oxidation of the metal powders. The dried sulfur mixture was mixed with Fe and Ni powders and mechanically mixed before two aliquots were placed in separate SiO2 tubes. Each tube was held under vacuum for ~30 minutes, sealed under vacuum, and then heated at 800C for 48 hours. Like [4], the synthesis products were composed of pourous sulfide crystals. Major element analyses of both experimental aliquots, collected using an Electron Probe Microanalyser (EPMA), yield indistinguishable major element compositions (uncertainties in 2SE & 2RSE[%]), with an average of 57.90  0.09 (0.16 %), 4.95  0.03 (0.55 %), and 39.70  0.14 (0.34 %), for Fe, Ni, and S, respectively. Trace element data were measured using a Photon Machines 193nm laser ablation system coupled to a Thermo-Scientific Element-XR ICP-MS. Spot sizes were limited to 50 μm due to the porous nature of the sulfide target material. Trace element abundances, normalized to Fe as an internal standard, were also found to be homogenous between the two aliquots, with weighted mean 2RSE (%) values of <3.0 for all trace elements. Synthesized products were re-powdered and absolute concentrations measured via solution ICP-MS. Although the sulfide appears to be homogenous, sintering experiments will be performed to more closely match the standard density to natural sulfides and minimize differences in ablation behavior. Improved density also allows for higher sensitivity (i.e. more compact target material) and larger spot sizes or traverses, as void space is eliminated. Additional major and trace element analyses on the products of the sintering experiments will be undertaken.

Jacob B Setera↗

Fabrication and Performance Characterization of Electrospun Nanofibers for Terrestrial and Space Applications

Passive thermal control is vital in space, especially for extended missions involving cryogen storage needing protection from sunlight. This presentation covers the development of a thermal coating with wavelength-selective characteristics using polymeric materials that scatter sunlight with minimal absorption, resulting in high solar reflectance. Furthermore, these coatings allow emission in infrared wavelengths, making them suitable for deep-space missions requiring extended storage. The study chose polymer and surface morphology suitable as solar reflectors that are lightweight, compact, and easily manufacturable, focusing on different multilayered nanofibers made using polytetrafluoroethylene and polyethylene-oxide (PTFE-PEO), polyvinylidene fluoride-hexafluoropropylene (PVDF-HFP), and silica (SiO2). To understand how these materials interact with sunlight, we analyze light propagation through several layers of electrospun nanofibers to determine spectral reflectance, absorptance, and transmittance by solving the Maxwell equations. The model is also compared with measurements of spectral reflectance and transmittance of the various electrospun nanofibers using spectrophotometers interfaced with integrating spheres in the ultraviolet, visible, near-infrared, and mid-infrared wavelengths. The presentation will also describe the design and analysis, fabrication of the nanofibers using electrospinning, and several experimental characterization techniques to determine suitability for space applications. The nanofibers are tested using various tools to determine their suitability for cryogenic applications in space. Lastly, the samples were tested under low vacuum conditions and exposing them to sunlight inside a vacuum chamber maintained at 300 K. Relative to the incoming solar flux (1000 W/m2), only 1% of the energy was absorbed under these operating conditions. With further optimization and testing under high vacuum cryogenic environments, the samples could demonstrate much lower absorption or higher reflection of incoming solar radiation, which can translate to low temperatures suitable for cryogen storage.

Chieloka Ibekwe↗

Foreign Object Damage in a Yb2Si2O7 Environmental Barrier Coating

Foreign object damage (FOD) is one of the key damage/failure modes in environmental barrier coatings (EBCs) developed for gas turbine engines. A limiting factor of EBC development is the growth of a SiO2 thermally grown oxide (TGO) layer which exacerbates spallation. While there has been extensive research on FOD in metals, ceramics, and ceramic matrix composites (CMCs), there is a limited understanding of the effects of FOD in EBC systems. Previous work studied the effects of TGO growth on FOD in NASA’s Generation II Ytterbium Disilicate EBC. Recently, it has been shown that the addition of oxide modifiers to the Ytterbium Disilicate EBC resulted in significant reduction in TGO growth. The present work will compare FOD in a modified Ytterbium Disilicate EBC to that of the baseline Ytterbium Disilicate EBC. FOD testing was conducted on both as-processed and steam cycled samples at room temperature using a 1.59mm steel ball projectile with particle velocities ranging from 50–300 m/s at a normal incidence angle. The impact damage was characterized by optical profilometry and scanning electron microscopy (SEM) of the cross-sections.

Leland Hoffman↗

System Modeling of a Lunar Molten Regolith Electrolysis Plant

Introduction: In-Situ Resource Utilization (ISRU) is the process of extracting local resources to produce commodities for propulsion, life support systems, and off-planet construction rather than transporting consumables from Earth. Molten Regolith Electrolysis (MRE) is a novel ISRU method of extracting oxygen gas and metal alloy from lunar regolith. The MRE process involves placing lunar regolith between two electrodes, through which current is passed, to melt the regolith and reduce the metal oxide constituents by direct electrolysis (e.g. FeO, SiO2, MgO, Al2O3) into oxygen gas and metal alloys. The oxygen is liquefied and used as propellant for landers, while the metals (e.g. Ferro-alloys) are further processed and used in structural building materials and parts manufacturing. A system model was developed that accounted for the major processes of an MRE plant (from excavation of raw materials to storage of products) to assess the feasibility of a lunar MRE plant. The System Engineering and Integration (SE&I) ISRU Modeling and Analysis (SIMA) team utilized its previously documented system sizing model, the Mission Analysis and Integration Tool (MAIT) [1] as framework of the system model. MAIT uses MATLAB/Simulink to integrate subsystem models into a complete system model of the MRE plant. Total mass, volume, and power requirements were computed for numerous iterations of a MRE plant. System Model: Figure 1: MRE Plant Block Diagram The regolith excavation model determines the mass and power needed to excavate sufficient regolith. The preheating auger initiates the regolith heating process before regolith enters the MRE re-actor to reduce the energy required to turn the solid into a molten liquid. The MRE reactor is modeled in COMSOL Multiphysics and based on the research by Dominguez, Sibille, and Schreiner [2, 3, 4]. This preliminary reactor model provides an accurate calculation of thermal equilibrium during electrochemical operation of the reactor system to assess the optimal mass and power required to process the inlet flow of regolith. The model also computes the outlet flowrates of oxygen and molten products. For this analysis, the primary components of the metal alloy considered were iron and silicon. The oxygen is then purified using an Yttrium Stabilized Zirconia (YSZ) electrode, followed by liquefaction using a 90K cryocooler to be stored as liquid oxygen in insulated cylindrical tanks. In future iterations of the system model, the molten metal tapped from the MRE reactor will undergo additional processing or refinement. However, downstream handling of metals is currently a technology gap that is missing a high TRL subsystem model. Therefore, for this analysis, the accumulated metal alloy stream terminates after leaving the MRE reactor. Study Goals: This analysis investigates multiple input variables to the system to determine the sensitivity of a (near) complete plant at full-scale. This preliminary investigation ran parametric sweeps on the MRE reactor geometry, electrical current supply, layers of multi-layer insulation (MLI) on the reactor, size of the electrodes in the oxygen purification model, and regolith composition (based on landing site location). Three production targets of oxygen (1,000, 10,000, and 50,000 kg/yr) were investigated for this analysis. The parametric sweeps conducted in this analysis provide valuable insight into the expected impact of the various model inputs on plant size. This information can be used to identify the most critical components of the plant and guide future decisions on allocating funding for research and development, providing subsystem developers with appropriate interfaces with downstream and upstream processes, and assessing the overall feasibility of MRE when compared to other ISRU plants. References: [1] Carlson, A. et al. (2024) ICES, ICES-2024-53. [2] Dominguez, D.A., and Sibille, L. (2011) AIAA, AIAA-2011-700. [3] Schreiner, S.S. (2015) MIT, Dissertation. [4] Schreiner, S.S. et al. (2016) ASR, 57(7), pp.1585-1603.

ISRU↗