SEARCH · Engineering Papers

Results for “Mixtures”

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.

SourceResearch connectionFrom yearThrough year

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

At least 37 records · Page 2

All sources NASA NTRS DOE OSTI

DOE OSTI2024

Implementation of new mixture rules has a substantial impact on combustion predictions for H 2 and NH 3

Complex-forming reactions comprise a substantial fraction of all important combustion reactions and are central to combustion behavior. Despite being often called “pressure-dependent” reactions, their rate constants depend on not only the pressure but also the composition. While modern combustion codes allow arbitrarily high accuracy in treating pressure dependence, recent work has consistently demonstrated dramatic failures of essentially all available treatments of mixture dependence. In situations where mixture dependence is treated at all, it is inevitably treated through specification of pressure-dependent rate constants for a set of pure bath gases, which are then combined to estimate the rate constant in a mixture via a “mixture rule.” While there had been a generally unquestioning confidence in these mixture rules, they had, in reality, been scarcely tested until the last decade, when comparisons against master equation calculations revealed order-of-magnitude errors for important pressure-dependent reactions. New mixture rules, based on the reduced pressure, have recently been proposed and shown to reproduce master equation calculations for broad classes of complex-forming reactions very accurately. Here, in this work, we present an implementation of one such new mixture rule (“LMR-R”) in Cantera and then use it to enable simulations that use new high-accuracy ab initio data for individual bath gases (for the first time, since codes previously could not accommodate the complex bath gas dependence). Demonstrations focus on combustion of H 2 and NH 3 , where (1) high-accuracy ab initio data are available and (2) the impact is expected to be large due to the high fractions of efficient colliders (e.g., H 2 O and NH 3 ) in the burned and unburned gases. Indeed, we find the impact of this treatment to be substantial and may explain previous modeling difficulties for these important carbon-free fuels, particularly for NH 3 , whose extraordinarily high third-body efficiency (~20) is often omitted from kinetic models.

Ammonia↗

DOE OSTI2024

Bioactivity Profiling of Chemical Mixtures for Hazard Characterization

Abstract The assessment and regulation of chemical toxicity to protect human health and the environment are done one chemical at a time and seldom at environmentally relevant concentrations. However, chemicals are found in the environment as mixtures, and their toxicity is largely unknown. Understanding the hazard posed by chemicals within the mixture is critical to enforce protective measures. Here, we demonstrate the application of bioactivity profiling of environmental water samples using the sentinel and ecotoxicology model species Daphnia to reveal the biomolecular response induced by exposure to real-world mixtures. We exposed a Daphnia strain to 30 sampled waters of the Chaobai River and measured the gene expression response profiles. Using a multiblock correlation analysis, we establish correlations between chemical mixtures identified in 30 water samples with gene expression patterns induced by these chemical mixtures. We identified 80 metabolic pathways putatively activated by mixtures of inorganic ions, heavy metals, polycyclic aromatic hydrocarbons, industrial chemicals, and a set of biocides, pesticides, and pharmacologically active substances. Our data-driven approach discovered both known bioactivity signatures with previously described modes of action and new pathways linked to undiscovered potential hazards. This study demonstrates the feasibility of reducing the complexity of real-world mixture toxicity to characterize the biomolecular effects of a defined number of chemical components based on gene expression monitoring of the sentinel species Daphnia.

Engineering↗

NASA NTRS1986

Effect of silane concentration on the supersonic combustion of a silane/methane mixture

A series of direct connect combustor tests was conducted to determine the effect of silane concentration on the supersonic combustion characteristics of silane/methane mixtures. Shock tube ignition delay data indicated more than an order of magnitude reduction in ignition delay times for both 10 and 20 percent silane/methane mixtures as compared to methane. The ignition delay time of the 10 percent mixture was only a factor of 2.3 greater than that of the 20 percent mixture. Supersonic combustion tests were conducted with the fuel injected into a model scramjet combustor. The combustor was mounted at the exit of a Mach 2 nozzle and a hydrogen fired heater was used to provide a variation in test gas total temperature. Tests using the 20 percent silane/methane mixture indicated considerable combustion enhancement when compared to methane alone. This mixture had an autoignition total temperature of 1650 R. This autoignition temperature can be contrasted with 2330 R for hydrogen and 1350 R for a 20 percent silane/hydrogen mixture in similar hardware. Methane without the silane additive did not autoignite in this configuration at total temperatures as high as 3900 R, the maximum temperature at which tests were conducted. Supersonic combustion tests with the silane concentration reduced to 10 percent indicated little improvement in combustion performance over pure methane. The addition of 20 percent silane to methane resulted in a pyrophoric fuel with good supersonic combustion performance. Reducing the silane concentration below this level, however, yielded a less pyrophoric fuel that exhibited poor supersonic combustion performance.

Northam, G. B.↗

NASA NTRS1991

New approach in direct-simulation of gas mixtures

Results are reported for an investigation of a new direct-simulation Monte Carlo method by which energy transfer and chemical reactions are calculated. The new method, which reduces to the variable cross-section hard sphere model as a special case, allows different viscosity-temperature exponents for each species in a gas mixture when combined with a modified Larsen-Borgnakke phenomenological model. This removes the most serious limitation of the usefulness of the model for engineering simulations. The necessary kinetic theory for the application of the new method to mixtures of monatomic or polyatomic gases is presented, including gas mixtures involving chemical reactions. Calculations are made for the relaxation of a diatomic gas mixture, a plane shock wave in a gas mixture, and a chemically reacting gas flow along the stagnation streamline in front of a hypersonic vehicle. Calculated results show that the introduction of different molecular interactions for each species in a gas mixture produces significant differences in comparison with a common molecular interaction for all species in the mixture. This effect should not be neglected for accurate DSMC simulations in an engineering context.

Chung, Chan-Hong↗

NASA NTRS1950

Foaming of mixtures of pure hydrocarbons

Mixtures of pure liquid hydrocarbons are capable of foaming. Nine hydrocarbons were mixed in pairs, in all possible combinations, and four proportions of each combination. These mixtures were sealed in glass tubes, and the foaming was tested by shaking. Mixtures of aliphatic with other aliphatic hydrocarbons, or of alkyl benzenes with other alkyl benzenes, did not foam. Mixtures of aliphatic hydrocarbons with alkyl benzenes did foam. The proportions of the mixtures greatly affected the foaming, the maximum foaming of 12 of 20 pairs being at the composition 20 percent aliphatic hydrocarbon, 80 percent alkyl benzene. Six seconds was the maximum foam lifetime of any of these mixtures. Aeroshell 120 lubricating oil was fractionated into 52 fractions and a residue by extraction with acetone in a fractionating extractor. The index of refraction, foam lifetime, color, and viscosity of these fractions were measured. Low viscosity and high index fractions were extracted first. The viscosity of the fractions extracted rose and the index decreased as fractionation proceeded. Foam lifetimes and color were lowest in the middle fractions. Significance is attached to the observation that none of the foam lifetimes of the fractions or residue is as high as the foam lifetime of the original Aeroshell, indicating that the foaming is not due to a particular foaming constituent, but rather to the entire mixture.

LUBRICATING OILS - DISTILLATION↗

NASA NTRS1992

Dielectric constant of liquid alkanes and hydrocarbon mixtures

The complex dielectric constants of n-alkanes with two to seven carbon atoms have been measured. The measurements were conducted using a slotted-line technique at 1.2 GHz and at atmospheric pressure. The temperature was varied from the melting point to the boiling point of the respective alkanes. The real part of the dielectric constant was found to decrease with increasing temperature and correlate with the change in the molar volume. An upper limit to all the loss tangents was established at 0.001. The complex dielectric constants of a few mixtures of liquid alkanes were also measured at room temperature. For a pentane-octane mixture the real part of the dielectric constant could be explained by the Clausius-Mosotti theory. For the mixtures of n-hexane-ethylacetate and n-hexane-acetone the real part of the dielectric constants could be explained by the Onsager theory extended to mixtures. The dielectric constant of the n-hexane-acetone mixture displayed deviations from the Onsager theory at the highest fractions of acetone. The dipole moments of ethylacetate and acetone were determined for dilute mixtures using the Onsager theory and were found to be in agreement with their accepted gas-phase values. The loss tangents of the mixtures exhibited a linear relationship with the volume fraction for low concentrations of the polar liquids.

NASA Program Exobiology↗

NASA NTRSDate not supplied

Demonstration of Capability to Simulate Particle Irregular Shape and Poly-Disperse Mixtures Within Lunar Lander Plume-Surface Interaction

Plume-Surface Interaction (PSI) between lander engine plumes and regolith soil creates hazards in obscuration and contamination by particle clouds, high-energy ejecta streams, and landing area cratering damage. The MSFC Fluid Dynamics Branch is developing simulation tools to offer a predictive PSI capability to NASA customers such as the Human Lander System (HLS) and Commercial Lunar Payload Services (CLPS). The Gas-Granular Flow Solver (GGFS) is the main application tool for coupled gas-particle two-phase flow simulations to predict the range of PSI effects from onset of surface erosion to deep crater formation. GGFS features an Eulerian-Eulerian modeling approach, treating both gas and granular material as interacting continuum phases. Modeling the lunar regolith granular material fluidic characteristics poses special challenges due to complex particle shapes and mixture composition. The lunar regolith is poorly sorted with broad particle size distributions and large fines content. It has significant cohesion, due to interlocking jagged particle shapes. Eulerian granular material flow modeling requires closure formulations for the granular material constitutive models (stress, friction, collisional and kinetic energy dissipation, drag, etc.). While closure models for spherical particles are available from particle kinetic theory, closure models for realistic non-spherical particles must be extracted from unit physics Discrete Element Model (DEM) particle interaction simulations and provided in the form of tabular datasets. The effects of particle irregular shape (non-spherical shape factors, angular particle surface roughness, and interlocking features) are simulated by approximating the particle features in the form of grouped elemental spheres to form composite particles in the DEM simulations. The effects of the wide range of regolith mixture particle sizes and the strong effects of the presence of the small particle sizes results in high cohesion and low porosity of the regolith mixture. The range of particle sizes is simulated by binning the particle sizes into an appropriate finite number of particle-size species and solving the problem as a species mixture. Combining these two modeling approaches enables simulations to capture both, the contributions of the irregular particle shape and the particle size distribution. The integration and maturation of the DEM-based constitutive model database generation process and poly-disperse mixture binning approach into the GGFS simulation framework are proceeding under funding by the NASA Game Changing Development program. The status of current capabilities will be presented in comparisons of crater characteristics resulting for spherical and irregular shape particles, and for mono-, bi-, and tri-disperse mixture simulations of Apollo LM plume-surface interaction. The computational results confirm the significance of including the particle shape and mixture effects. Going forward plans for the full implementation of the general poly-disperse regolith modeling capability and maturation towards NASA project application readiness under the GCD program will be presented.

Peter A Liever↗

NASA NTRSDate not supplied

Demonstration of Capability to Simulate Particle Irregular Shape and Poly-Disperse Mixtures Within Lunar Lander Plume-Surface Interaction

Peter A. Liever↗

NASA NTRS1981

Laboratory measurements of selected optical, physical, chemical, and remote-sensing properties of five water mixtures containing Calvert clay and a nonfluorescing dye

Total suspended solids concentrations ranged from 6.1 ppm to 24.3 ppm and sizes ranged between 1.5 micrometers and 10 micrometers with the most frequently occurring size less than 2 micrometers. Iron concentration was less than 1 percent of the total suspended solids. Nonfluorescing dye concentrations of the two mixtures were 20 ppm and 40 ppm. Attenuation coefficient for the five mixtures ranged from 4.8/m to 21.3/m. Variations in volume scattering function with phase angle were typical. Variations in attenuation and absorption coefficient with wavelength were similar for the mixtures without the dye. Attenuation coefficient of the mixtures with the dye increased for wavelengths less than 600 nm due to the dye's strong absorption peak near 500 nm. Reflectance increased as the concentration of Calvert clay increased and peaked near 600 nm. The nonfluorescent dye decreased the magnitude of the peak, but had practically no effect on the variation for wavelengths greater than 640 nm. At wavelengths less than 600 nm, the spectral variations of the mixtures with the dye were significantly different from those mixtures without the dye.

Usry, J. W.↗

NASA NTRS1987

RL10A-3-3B high mixture ratio qualification program

The results of the high mixture ratio qualification testing of the RL10 engine for the Shuttle/Centaur program are presented. The objective of the engine qualification test was to demonstrate the suitability of the RL10A-3-3B engine for space vehicle flight by subjecting it to the testing specified in RL10A-3-3B Model Specification Number 2295 dated February 1986. The applicable section of the specification is presented. Due to payload volume advantages which can be achieved by increasing the operating mixture ratio of the RL10, a decision was made to qualify the engine to run at a higher mixture ratio. A program was created to qualify the RL10 engine for operation at 15,000 pounds thrust and a nominal 6.0 to 1 mixture ratio. This model of the engine was designated the RL10A-3-3B. The qualification program included three test series as follows: (1) hardware durability and limits test in which the engine completed 23 firings and 4605.7 seconds with 1588.7 seconds at less than 6.6 mixture ratio; (2) preliminary qualification test in which the engine completed 26 firings and 5750 seconds; and (3) qualification test in which the engine completed 26 hot firings and 5693.4 seconds with 905.9 seconds at 6.7 mixture ratio. Several changes in engine hardware were required for operation of the RL10A-3-3B engine in the Space Shuttle which include a duel pressure switch ignition, an oxidizer flow control, and helium plumbing changes.

Vogel, T.↗

NASA NTRS1991

Experimental thermal conductivity, thermal diffusivity, and specific heat values for mixtures of nitrogen, oxygen, and argon

Experimental measurements of thermal conductivity and thermal diffusivity obtained with a transient hot-wire apparatus are reported for three mixtures of nitrogen, oxygen, and argon. Values of the specific heat, Cp, are calculated from these measured values and the density calculated with an equation of state. The measurements were made at temperatures between 65 and 303 K with pressures between 0.1 and 70 MPa. The data cover the vapor, liquid, and supercritical gas phases for the three mixtures. The total reported points are 1066 for the air mixture (78.11 percent nitrogen, 20.97 percent oxygen, and 0.92 percent argon), 1058 for the 50 percent nitrogen, 50 percent oxygen mixture, and 864 for the 25 percent nitrogen, 75 oxygen mixture. Empirical thermal conductivity correlations are provided for the three mixtures.

Perkins, R. A.↗

NASA NTRS1991

Soot formation during combustion of unsupported methanol/toluene mixture droplets in microgravity

Results are reported of an experimental study tracing the influence of liquid composition on soot formation and the burning rate of a droplet composed of a binary miscible mixture of liquids. The mixture components represented a highly sooting fuel, toluene, and a nonsooting fuel, methanol. The toluene concentration in methanol was shown to dramatically influence flame luminosity and soot production. Neither burning rates nor a propensity for flame extinction appeared to be significantly affected by toluene mixture fractions. Five-percent toluene mixture droplets behaved like pure methanol droplets in terms of burning rate, lack of flame luminosity, and extinction. Increasing the toluene concentration in the droplets to 25 percent increased flame luminosity, yet no visible soot agglomerates were observed. The 50-percent-mixture droplets burned with highly luminous flames and large amounts of soot agglomerates collecting inside the flame. All the mixture droplets showed burning rates similar to those of pure methanol and likewise exhibited flame extinction before complete droplet vaporization.

Jackson, G. S.↗

NASA NTRS1996

Isopycnic Phases and Structures in H2O/CO2/Ethoxylated Alcohol Surfactant Mixtures

Ternary mixtures of H2O and CO2 with ethoxylated alcohol (C(i)E(j)) surfactants can form three coexisting liquid phases at conditions where two of the phases have the same density (isopycnic phases). Isopycnic phase behavior has been observed for mixtures containing the surfactants C8E5, C10E6, and C12E6, but not for those mixtures containing either C4E1 or CgE3. Pressure-temperature (PT) projections for this isopycnic three-phase equilibrium were determined for H2O/CO2/C8E5 and H2O/CO2/C10E6 mixtures at temperatures from approximately 25 to 33 C and pressures between 90 and 350 bar. As a preliminary to measuring the microstructure in isopycnic three component mixtures, phase behavior and small angle neutron scattering (SANS) experiments were performed on mixtures of D2O/CO2/ n-hexaethyleneglycol monododecyl ether (C12E6) as a function of temperature (25-31 C), pressure (63.1-90.7 bar), and CO2 composition (0-3.9 wt%). Parameters extracted from model fits of the SANS spectra indicate that, while micellar structure remains essentially unchanged, critical concentration fluctuations increase as the phase boundary and plait point are approached.

Paulaitis, Michael E.↗

NASA NTRS1999

Phase Separation Kinetics in Isopycnic Mixtures of H2O/CO2/Ethoxylated Alcohol Surfactants

Ternary mixtures of H2O and CO2 with ethoxylated alcohol (C(sub i)E(sub j)) surfactants form three coexisting liquid phases at conditions where two of the phases have equal densities (isopycnic phases). Isopycnic phase behavior has been observed for mixtures containing C8E5, C10E6, and C12E6 surfactants, but not for those mixtures containing either C4E1 or C8E3 surfactants. Pressure-temperature (PT) projections for this three-phase equilibrium were determined for H2O/CO2/C8E5 and H2O/CO2/C10E6 mixtures at temperatures from approximately 25 to 33 C and pressures between 90 and 350 bar. Measurements of the microstructure in H2O/CO2/C12E6 mixtures as a function of temperature (25-31 C), pressure (63.1-90.7 bar), and CO2 composition (0-3.9 wt%) have also been carried out to show that while micellar structure remains essentially un-changed, critical concentration fluctuations increase as the phase boundary and plait point are approached. In this report, we present our first measurements of the kinetics of isopycnic phase separation for ternary mixtures of H2O/CO2/C8E5.

Lesemann, Markus↗

NASA NTRS2000

Marangoni Effects on Near-Bubble Microscale Transport During Boiling of Binary Fluid Mixtures

In earlier investigations, Marangoni effects were observed to be the dominant mechanism of boiling transport in 2-propanol/water mixtures under reduced gravity conditions. In this investigation we have examined the mechanisms of binary mixture boiling by exploring the transport near a single bubble generated in a binary mixture between a heated surface and cold surface. The temperature field created in the liquid around the bubble produces vaporization over the portion of its interface near the heated surface and condensation over portions of its interface near the cold surface. Experiments were conducted using different mixtures of water and 2-propanol under 1g conditions and under reduced gravity conditions aboard the KC135 aircraft. Since 2-propanol is more volatile than water, there is a lower concentration of 2-propanol near the hot surface and a higher concentration of 2-propanol near the cold plate relative to the bulk quantity. This difference in interface concentration gives rise to strong Marangoni effects that move liquid toward the hot plate in the near bubble region for 2-propanol and water mixtures. In the experiments in this study, the pressure of the test system was maintained at about 5 kPa to achieve the full spectrum of boiling behavior (nucleate boiling, critical heat flux and film boiling) at low temperature and heat flux levels. Heat transfer data and visual documentation of the bubble shape were extracted from the experimental results. In the 1-g experiments at moderate to high heat flux levels, the bubble was observed to grow into a mushroom shape with a larger top portion near the cold plate due to the buoyancy effect. The shape of the bubble was somewhat affected by the cold plate subcooling and the superheat of the heated surface. At low superheat levels for the heated surface, several active nucleation sites were observed, and the vapor stems from them merged to form a larger bubble. The generation rate of vapor is moderate in this regime and the bubble shape is cylindrical in appearance. In some instances, the bubble interface appeared to oscillate. At higher applied heat flux levels, the top of the bubble became larger, apparently to provide more condensing interface area adjacent to the cold plate. Increasing the applied heat flux ultimately led to dry-out of the heated surface, with conditions just prior to dryout corresponding to the maximum heat flux (CHF). A more stable bubble was observed when the system attained the minimum heat flux (for film boiling). In this regime, most of the surface under the bottom of the bubble was dry with nucleate boiling sometimes occuring around the contact perimeter of the bubble at heated surface. Different variations (e.g. gap between two plates, molar concentration of the liquid mixture) of the experiments were examined to determine parametric effects on the boiling process and to determine the best conditions for the KC135 reduced gravity tests. Variation of the gap was found to have a minor impact on the CHF. However, reducing the gap between the hot and cold surface was observed to significantly reduce the minimum heat flux for fixed molar concentration of 2-propanol. In the reduced gravity experiments aboard the KC135 aircraft, the bubble formed in the 6.4 mm gap was generally cylindrical or barrel shaped and it increased its extent laterally as the surface superheat increased. In reduced gravity experiments, dryout of the heated surface under the bubble was observed to occur at a lower superheated temperature than for 1g conditions. Observed features of the boiling process and heat transfer data under reduced gravity will be discussed in detail. The results of the reduced gravity experiments will also be compared to those obtained in comparable 1g experiments. In tandem with the experiments we are also developing a computational model of the transport in the liquid surrounding the bubble during the boiling process. The computational model uses a level set method to model motion of the interface. It will incorporate a macroscale treatment of the transport in the liquid gap between the surfaces and a microscale treatment of transport in the regions between the bubble interface and the solid surfaces. The features of the model will be described in detail. Future research directions suggested by the results to date will also be discussed.

V. Carey↗

NASA NTRS1996

Near azeotropic mixture substitute for dichlorodifluoromethane

The present invention comprises a refrigerant mixture having two halocarbon components. The first component is present in a mole fraction of about 0.7 to less than 1.0 while the second component is present in a mole fraction of more than 0.0 to about 0.3. The first component is CH.sub.2 FCF.sub.3. The second component can be CHClFCF.sub.3, CH.sub.3 CClF.sub.2, a mixture of CHClFCF.sub.3 and CH.sub.3 CClF.sub.2, a mixture of CHF.sub.2 CH.sub.3 and CHClFCF.sub.3, a mixture of CHF.sub.2 CH.sub.3 and CH.sub.3 CClF.sub.2, or a mixture of CHClFCF.sub.3, CH.sub.3 CClF.sub.2 and CHF.sub.2 CH.sub.3. The preferred embodiment of this invention comprises about 0.7 to less than 1.0 mole fraction CH.sub.2 FCF.sub.3, and more than 0.0 to about 0.3 mole fraction of a mixture of CHClFCF.sub.3 and CH.sub.3 CClF.sub.2. The most preferred embodiment of this invention comprises about 0.7 to less than 1.0 mole fraction CH.sub.2 FCF.sub.3 and more than 0.0 to about 0.3 mole fraction CH.sub.3 CClF.sub.2. The resulting refrigerant has a vapor pressure close to-that of CF.sub.2 Cl.sub.2, a nearly constant vapor pressure with evaporation, and is substantially less damaging to the Earth's ozone layer than CF.sub.2 Cl.sub.2.

Jones, Jack A.↗

NASA NTRS2011

Exploring the Fate of Nitrogen Heterocycles in Complex Prebiotic Mixtures

A long standing question in the field of prebiotic chemistry is the origin of the genetic macromolecules DNA and RNA. DNA and RNA have very complex structures with repeating subunits of nucleotides, which are composed of nucleobases (nitrogen heterocycles) connected to sugar-phosphate. Due to the instability of some nucleobases (e.g. cytosine), difficulty of synthesis and instability of D-ribose, and the likely scarcity of polyphosphates necessary for the modern nucleotides, alternative nucleotides have been proposed for constructing the first genetic material. Thus, we have begun to investigate the chemistry of nitrogen heterocycles in plausible, complex prebiotic mixtures in an effort to identify robust reactions and potential alternative nucleotides. We have taken a complex prebiotic mixture produced by a spark discharge acting on a gas mixture of N2, CO2, CH4, and H2, and reacted it with four nitrogen heterocycles: uracil, 5-hydroxymethyluracil, guanine, and isoxanthopterin (2-amino-4,7-dihydroxypteridine). The products of the reaction between the spark mixture and each nitrogen heterocycle were characterized by liquid chromatography coupled to UV spectroscopy and Orbitrap mass spectrometry. We found that the reaction between the spark mixtUl'e and isoxanthopterin formed one major product, which was a cyanide adduct. 5-hydroxymethyluracil also reacted with the spark mixture to form a cyanide adduct, uracil-5-acetonitrile, which has been synthesized previously by reacting HCN with S-hydroxymethyluracil. Unlike isoxanthopterin, the chromatogram of the 5-hydroxymethyluracil reaction was much more complex with multiple products including spark-modified dimers. Additionally, we observed that HMU readily self-polymerizes in solution to a variety of oligomers consistent with those suggested by Cleaves. Guanine and uracil, the biological nucleobases, did not react with the spark mixture, even at high temperature (100 C). This suggests that there are alternative nucleobases which are more reactive under prebiotic conditions and may have been involved in producing precursor nucleotides.

Smith, Karen E.↗

DOE OSTI2025

Using Neural Networks to Identify Mixture Components in Hyperspectral Reflectance Data

Neural networks have been employed to identify materials of interest from hyperspectral data (generally imagery) based on their unique spectral signatures. This approach assumes that there is a single material that is standing out from the rest of the spectrum to be identified. However, pixels often contain more than one material, or a material of interest may itself be a mixture of multiple materials. Neural networks are only as good as the data used to train them, and it takes a great deal of work in the laboratory to identify, make, and measure all potential mixtures of interest. Thus, researchers often calculate synthetic spectra using algorithms with varying degrees of fidelity to the physics that govern the interactions between light and multiple materials. In this work, we have (1) adapted a neural network designed to identify mixture components from Raman spectroscopy to work with visible to near‐infrared reflectance data and (2) tested three common mixture algorithms to determine the most accurate and least computationally expensive method to build synthetic training datasets. With our initial test dataset, we have achieved accuracies of > 90% and found that the synthetic training dataset produced using the Hapke mixture model provides the best results.

99 GENERAL AND MISCELLANEOUS↗