Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “LIME”

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.

At least 127 records · Page 7

ZeroCAL: Eliminating Carbon Dioxide Emissions from Limestone’s Decomposition to Decarbonize Cement Production

Limestone (calcite, CaCO 3 ) is an abundant and cost-effective source of calcium oxide (CaO) for cement and lime production. However, the thermochemical decomposition of limestone (~800 °C, 1 bar) to produce lime (CaO) results in substantial carbon dioxide (CO 2(g) ) emissions and energy use, i.e., ~1 tonne [t] of CO 2 and ~1.4 MWh per t of CaO produced. Here, we describe a new pathway to use CaCO 3 as a Ca source to make hydrated lime (portlandite, Ca(OH) 2 ) at ambient conditions (p, T) while nearly eliminating process CO 2(g) emissions (as low as 1.5 mol. % of the CO 2 in the precursor CaCO 3 , equivalent to 9 kg of CO 2(g) per t of Ca(OH) 2 ) within an aqueous flowelectrolysis/ pH-swing process that coproduces hydrogen (H 2(g) ) and oxygen (O 2(g) ). Because Ca(OH) 2 is a zero-carbon precursor for cement and lime production, this approach represents a significant advancement in the production of zero-carbon cement. The Zero CArbon Lime (ZeroCAL) process includes dissolution, separation/recovery, and electrolysis stages according to the following steps: (Step 1) chelator (e.g., ethylenediaminetetraacetic acid, EDTA)-promoted dissolution of CaCO 3 and complexation of Ca 2+ under basic (>pH 9) conditions, (Step 2a) Ca enrichment and separation using nanofiltration (NF), which allows separation of the Ca-EDTA complex from the accompanying bicarbonate (HCO 3 – ) species, (Step 2b) acidity-promoted decomplexation of Ca from EDTA, which allows near-complete chelator recovery and the formation of a Ca-enriched stream, and (Step 3) rapid precipitation of Ca(OH) 2 from the Ca-enriched stream using electrolytically produced alkalinity. These reactions can be conducted in a seawater matrix yielding coproducts including hydrochloric acid (HCl) and sodium bicarbonate (NaHCO 3 ), resulting from electrolysis and limestone dissolution, respectively. Careful analysis of the reaction stoichiometries and energy balances indicates that approximately 1.35 t of CaCO 3 , 1.09 t of water, 0.79 t of sodium chloride (NaCl), and ~2 MWh of electrical energy are required to produce 1 t of Ca(OH) 2 , with significant opportunity for process intensification. This approach has major implications for decarbonizing cement production within a paradigm that emphasizes the use of existing cement plants and electrification of industrial operations, while also creating approaches for alkalinity production that enable cost-effective and scalable CO 2 mineralization via Ca(OH) 2 carbonation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Initial Examination of Low Velocity Sphere Impact of Glass Ceramics

This report summarizes US Army TARDEC sponsored work at Oak Ridge National Laboratory (ORNL) involving low velocity (< 30 m/s or < 65 mph) sphere impact testing of two materials from the lithium aluminosilicate family reinforced with different amounts of ceramic particulate, i.e., glass-ceramic materials, SCHOTT Resistan™-G1 and SCHOTT Resistan™-L. Both materials are provided by SCHOTT Glass (Duryea, PA). This work is a follow-up to similar sphere impact studies completed by the authors on PPG's Starphire® soda-lime silicate glass and SCHOTT BOROFLOAT® borosilicate glass. A gas gun or a sphere-drop test setup was used to produce controlled velocity delivery of silicon nitride (Si 3 N 4 ) spheres against the glass ceramic tile targets. Minimum impact velocities to initiate fracture in the glass-ceramics were measured and interpreted in context to the kinetic energy of impact and the elastic property mismatch between sphere and target material. Quasistatic spherical indentation was also performed on both glass ceramics and their contact damage responses were compared to those of soda-lime silicate and borosilicate glasses. Lastly, variability of contact damage response was assessed by performing spherical indentation testing across the area of an entire glass ceramic tile. The primary observations from this low velocity (< 30 m/s or < 65 mph) testing were: (1) Resistan™-L glass ceramic required the highest velocity of sphere impact for damage to initiate. Starphire® soda-lime silicate glass was second best, then Resistan™-G1 glass ceramic, and then BOROFLOAT® borosilicate glass. (2) Glass-ceramic Resistan™-L also required the largest force to initiate ring crack from quasi-static indentation. That ranking was followed, in descending order, by Starphire® soda-lime silicate glass, Resistan™-G1 glass ceramic, and BOROFLOAT® borosilicate glass. (3) Spheres with a lower elastic modulus require less force to initiate fracture in Resistan™-G1 from quasi-static spherical indentation. This indicates that friction is affecting ring crack initiation in Resistan™-G1. Friction also affected ring crack initiation in Starphire® soda-lime silicate and BOROFLOAT® borosilicate glasses. Among these three materials, friction was the most pronounced (largest slope in the RCIF-elastic modulus graph) in the Starphire® and least pronounced in the BOROFLOAT®. The reason for this is not understood, but differences in deformation behavior under high contact stresses could be a cause or contributor to this. (4) The force necessary to initiate contact-induced fracture is higher under dynamic conditions than it is under quasi-static conditions in Resistan™-L and Resistan™-G1 glass ceramics. This is a trend observed too in Starphire® and BOROFLOAT®. (5) There is a subtle indication there was intra-tile differences in spherical indentation-induced ring crack initiation forces. This is not a material property nor is it exclusive to glass-ceramic Resistan™-G1 glass ceramic, rather, it is a statistical mechanical response to an accumulated history of processing and handling of that specific tile.

36 MATERIALS SCIENCE↗

Contaminant Leach Testing of Hanford Tank 241-C-104 Residual Waste

Leach testing of Tank C-104 residual waste was completed using batch and column experiments. Tank C-104 residual waste contains exceptionally high concentrations of uranium (i.e., as high as 115 mg/g or 11.5 wt.%). This study was conducted to provide data to develop contaminant release models for Tank C-104 residual waste and Tank C-104 residual waste that has been treated with lime to transform uranium in the waste to a highly insoluble calcium uranate (CaUO4) or similar phase. Three column leaching cases were investigated. In the first case, C-104 residual waste was leached with deionized water. In the second case, crushed grout was added to the column so that deionized water contacted the grout prior to contacting the waste. In the third case, lime was mixed in with the grout. Results of the column experiments demonstrate that addition of lime dramatically reduces the leachability of uranium from Tank C-104 residual waste. Initial indications suggest that CaUO4 or a similar highly insoluble calcium rich uranium phase forms as a result of the lime addition. Additional work is needed to definitively identify the uranium phases that occur in the as received waste and the waste after the lime treatment.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Direct Causticizing for Black Liquor Gasification in a Circulating Fluidized Bed

Gasification of black liquor (BLG) has distinct advantages over direct combustion in Tomlinson recovery boilers. In this project we seek to resolve causticizing issues in order to make pressurized BLG even more efficient and cost-effective. One advantage of BLG is that the inherent partial separation of sulfur and sodium during gasification lends itself to the use of proven high yield variants to conventional kraft pulping which require just such a separation. Processes such as polysulfide, split sulfidity, ASAQ, and MSSAQ can increase pulp yield from 1% to 10% over conventional kraft but require varying degrees of sulfur/sodium separation, which requires additional [and costly] processing in a conventional Tomlinson recovery process. However during gasification, the sulfur is partitioned between the gas and smelt phases, while the sodium all leaves in the smelt; thus creating the opportunity to produce sulfur-rich and sulfur-lean white liquors for specialty pulping processes. A second major incentive of BLG is the production of a combustible product gas, rich in H2 and CO. This product gas (a.k.a. “syngas”) can be used in gas turbines for combined cycle power generation (which is twice as efficient as the steam cycle alone), or it can be used as a precursor to form liquid fuels, such as dimethyl ether or Fischer Tropsh diesel. There is drawback to BLG, which has the potential to become a third major incentive if this work is successful. The causticizing load is greater for gasification of black liquor than for combustion in a Tomlinson boiler. So implementing BLG in an existing mill would require costly increases to the causticizing capacity. In situ causticizing [within the gasifier] would handle the entire causticizing load and therefore eliminate the lime cycle entirely. Previous work by the author and others has shown that titanate direct causticizing (i.e. in situ) works quite well for high-temperature BLG (950°C), but was limited to pressures below about 5 bar. It is desirable however to operate BLG at 20-30 bar for efficiency reasons related to either firing the syngas in a turbine, or catalytically forming liquid fuels. This work focused on achieving high direct causticizing yields at 20 bars pressure. The titanate direct causticizing reactions are inhibited by CO2. Previous work has shown that the partial pressure of CO2 should be kept below about 0.5 bar in order for the process to work. This translates to a total reactor pressure limit of about 5 bar for airblown BLG, and only 2 bar for O2-blown BLG. In this work a process was developed in which the CO2 partial pressure could be manipulated to a level under 0.5 bar with the total system pressure at 10 bar during O2-blown BLG. This fell short of our 20 bar goal but still represents a substantial increase in the pressure limit. A material and energy balance was performed, as well as first-pass economics based on capital and utilities costs. Compared to a reference case of using BLG with a conventional lime cycle [Larson, 2003], the IRR and NVP were estimated for further replacing the lime kiln with direct causticizing. The economics are strongly dependent on the price of lime kiln fuel. At $6/mmBTU the lime cycle is the clear choice. At $8/mmBTU the NPV is $10M with IRR of 17%. At $12/mmBTU the NPV is $45M with IRR of 36%. To further increase the total allowable pressure, the CO2 could be further decreased by further decreasing the temperature. Testing should be done at 750C. Also a small pilot should be built.

09 BIOMASS FUELS↗

IMPROVED BAR IMPACT TESTS USING A PHOTONIC DOPPLER VELOCIMETER

Bar impact tests, using the techniques described elsewhere in this symposium, were used to measure compressive and tensile strengths of borosilicate glass, soda lime glass, and a glass ceramic. The glass ceramic was 25% crystalline spinel, furnished by Corning Inc. There are two measures of compressive strength: the peak stress that can be transmitted in unconfined compression, and the 'steady state' strength. For borosilicate glass and soda lime glass, these values were similar, being about 1.8 and 1.5 GPa, respectively. The glass ceramic (25% spinel) was almost 50% stronger. Tensile failure in the glass and glass ceramic takes places via surface flaws, and thus tensile strength is an extrinsic, as opposed to intrinsic property.

36 MATERIALS SCIENCE↗

Causticizing for Black Liquor Gasifiers

The cost-benefit outlook of black liquor gasification (BLG) could be greatly improved if the smelt causticization step could be achieved in situ during the gasification step. Or, at a minimum, the increase in causticizing load associated with BLG could be mitigated. A number of chemistries have been proven successful during black liquor combustion. In this project, three in situ causticizing processes (titanate, manganate, and borate) were evaluated under conditions suitable for high temperature entrained flow BLG, and low temperature steam reforming of black liquor. The evaluation included both thermodynamic modeling and lab experimentation. Titanate and manganate were tested for complete direct causticizing (to thus eliminate the lime cycle), and borates were evaluated for partial causticizing (to mitigate the load increase associated with BLG). Criteria included high carbonate conversion, corresponding hydroxide recovery upon hydrolysis, non process element (NPE) removal, and economics. Of the six cases (three chemistries at two BLG conditions), only two were found to be industrially viable: titanates for complete causticizing during high temperature BLG, and borates for partial causticizing during high temperature BLG. These two cases were evaluated for integration into a gasification-based recovery island. The Larsen [28] BLG cost-benefit study was used as a reference case for economic forecasting (i.e. a 1500 tpd pulp mill using BLG and upgrading the lime cycle). By comparison, using the titanate direct causticizing process yielded a net present value (NPV) of $25M over the NPV of BLG with conventional lime cycle. Using the existing lime cycle plus borate autocausticizing for extra capacity yielded a NPV of $16M.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Understanding Calibration and Error Propagation in Longitudinal and Lateral Manganin Gauge Shock Experiments

Soda-lime glass remains elastic under shock pressures up to approximately 6.5 GPa. In this paper, soda lime glass shock experiments are used to understand the calibration and error propagation in longitudinal and lateral manganin gauge experiments. The errors in the calibration process for both longitudinal and lateral manganin gauge experiments are identified and discussed. Five soda-lime glass experiments up to 3 GPa are analyzed using this refined technique. The results show good agreement between the experimental data and the published calibration by Rosenberg and Moshel. Errors are propagated from the experimental data measurements and the assumptions that are made in the calibration, where the bulk of the error seems to come from uncertainty in the dynamic yield strength of manganin. Finally, the method presented in this paper allows for analysis of dynamic shock experiments with longitudinal and lateral manganin gauges including error propagation.

36 MATERIALS SCIENCE↗

Explainable AI (XAI)-driven vibration sensing scheme for surface quality monitoring in a smart surface grinding process

Local Interpretable and Model-agnostic Explanation (LIME), an explainable artificial intelligence (XAI) approach is adapted to identify the globally important time-frequency bands for predicting average surface roughness (Ra) in a smart grinding process. The smart grinding setup consisted of a Supertech CNC precision surface grinding machine, instrumented with a Dytran piezoelectric accelerometer attached to the tailstock quill along the tangential direction (Y-axis). For every grinding pass, vibration signatures were captured, and the ground truth surface roughness values were recorded using a Mahr Marsurf M300C portable surface roughness profilometer. The roughness values ranged from 0.06 to 0.14 microns over the complete set of experiments. Time-frequency domain spectrogram frames were extracted for each of the vibration signals collected during the grinding process. Convolutional Neural Networks (CNNs) were modeled to predict the surface roughness based on these spectrogram frames and their image augmentations. The best CNN model was able to predict the roughness values with an overall R2-score of 0.95, training R2-score of 0.99, and testing R2-score of 0.81 with only 80 sets of vibration signals corresponding to 4 experiments with 20 trials each. Although the data size is not large enough to guarantee such performance metrics in real-world scenarios, one can extract statistically consistent explanations underlying the relationships these complex deep learning models capture. Further, the LIME methodology was implemented on the developed surface roughness CNN model to identify the important time-frequency bands (i.e., the superpixels of a spectrogram) influencing the predictions. Based on the identified important regions on the spectrogram frames, the corresponding frequency characteristics were determined that influence the surface roughness predictions. The important frequency range based on LIME results was approximately 11.7 to 19.1 kHz. The power of XAI was demonstrated by cutting down the sampling rate from 160 kHz to 30, 20, 10, and 5 kHz based on the important frequency range and considering Nyquist criteria. Separate CNN models were developed for these ranges by only extracting time-frequency contents below their corresponding Nyquist cut-offs. A proper data acquisition strategy is proposed by comparing the model performances to argue the selection of a sufficient sampling rate to capture the grinding process successfully and robustly.

42 ENGINEERING↗

Reprint of: Explainable AI (XAI)-driven vibration sensing scheme for surface quality monitoring in a smart surface grinding process

Local Interpretable and Model-agnostic Explanation (LIME), an explainable artificial intelligence (XAI) approach is adapted to identify the globally important time-frequency bands for predicting average surface roughness (Ra) in a smart grinding process. The smart grinding setup consisted of a Supertech CNC precision surface grinding machine, instrumented with a Dytran piezoelectric accelerometer attached to the tailstock quill along the tangential direction (Y-axis). For every grinding pass, vibration signatures were captured, and the ground truth surface roughness values were recorded using a Mahr Marsurf M300C portable surface roughness profilometer. The roughness values ranged from 0.06 to 0.14 microns over the complete set of experiments. Time-frequency domain spectrogram frames were extracted for each of the vibration signals collected during the grinding process. Convolutional Neural Networks (CNNs) were modeled to predict the surface roughness based on these spectrogram frames and their image augmentations. The best CNN model was able to predict the roughness values with an overall R2-score of 0.95, training R2-score of 0.99, and testing R2-score of 0.81 with only 80 sets of vibration signals corresponding to 4 experiments with 20 trials each. Although the data size is not large enough to guarantee such performance metrics in real-world scenarios, one can extract statistically consistent explanations underlying the relationships these complex deep learning models capture. Further, the LIME methodology was implemented on the developed surface roughness CNN model to identify the important time-frequency bands (i.e., the superpixels of a spectrogram) influencing the predictions. Based on the identified important regions on the spectrogram frames, the corresponding frequency characteristics were determined that influence the surface roughness predictions. The important frequency range based on LIME results was approximately 11.7 to 19.1 kHz. The power of XAI was demonstrated by cutting down the sampling rate from 160 kHz to 30, 20, 10, and 5 kHz based on the important frequency range and considering Nyquist criteria. Separate CNN models were developed for these ranges by only extracting time-frequency contents below their corresponding Nyquist cut-offs. A proper data acquisition strategy is proposed by comparing the model performances to argue the selection of a sufficient sampling rate to capture the grinding process successfully and robustly.

47 OTHER INSTRUMENTATION↗

Achieving Energy Efficiency In Small Businesses In Low-Income Areas

Lime Energy, with funding from the Department of Energy’s Office of Energy Efficiency and Renewable Energy (EERE), recently concluded a project demonstrating a new business model for executing energy efficiency for small businesses, in which we aspired to focus on low-income urban areas. As initially conceived, the Project would advance beyond the current best practices used to foster energy efficiency action in small and midsized commercial buildings (under 25,000 square feet), which represent 95% of all commercial customers in the United States. As the project proceeded, Lime Energy discovered that the business model needed to be further refined in areas where Lime already had a presence in the community and the benefit of a relationship with the known utility. The need for this intermediate step – of deploying the business model in an established region – has since been validated.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Influence of the composition of cement kiln dust on its interaction with fly ash and slag

Cement kiln dust (CKD), a by-product of the cement industry, contains significant amounts of alkali, free lime, chloride and sulfate. Wide variation reported in the chemical composition of CKDs limits their potential application as a sustainable binder component in concrete. In the current study, the performance of two different CKDs as components in a novel binder is evaluated. Several binders are developed by blending CKDs with fly ash or slag. Binders with 70% CKD were prepared at a water-to-binder ratio of 0.4, and heat-cured at 75 °C to accelerate the strength development. The hydration progress was monitored using X-ray diffraction, and morphological examination was performed using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Ettringite and calcium aluminosilicate hydrate (C-A-S-H) were identified as the main hydration products in the hardened binder system. Strength development of CKD-based binder was found to be significantly influenced by its free lime and sulfate contents. -- Highlights: •Interaction of cement kiln dust with fly ash and slag was explored. •CKD with higher free lime and sulfate content increased the strength of binder. •C-S-H like reaction gel with fibrillar morphology is observed in CKD-based binders.

36 MATERIALS SCIENCE↗

Pozzolanic reactivity of natural pozzolans, ground glasses and coal bottom ashes and implication of their incorporation on the chloride permeability of concrete

It is well understood that the partial replacement of portland cement with pozzolans in sufficient proportions results in improved long-term performance of concrete. However, dwindling availability of the most widely used pozzolan in the construction industry, fly ash, is forecast due to the termination of coal-fired electricity power stations around the world, necessitating the exploration of alternative pozzolans. Experimental investigations are conducted to study the performance of traditional pozzolans such as fly ash, silica fume and natural pozzolans (metakaolin, pumice, perlite and lassenite) together with novel pozzolans, including ground glasses (high-alkali and low-alkali), and ground bottom ash. Twenty-one concrete mixtures are examined for compressive strength, electrical conductivity, electrical resistivity, chloride permeability, and chloride migration coefficient. The reactivity of pozzolans studied is also determined in terms of compressive strength in lime-pozzolan mortar using a modified lime-reactivity test method. Many of the materials tested demonstrate the potential to be used in place of fly ash. However, a wide variation in the performance of these materials is evident which highlights the need for a reliable test to determine the level of reactivity of a pozzolan. The results indicate that the 7-day strength in the modified lime-reactivity test provides a good indication of the pozzolanic reactivity of the material and of how the material can be expected to contribute to the strength and permeability of concrete. The use of electrical resistivity as an indicator of the performance of pozzolans in terms of strength development and chloride penetration resistance when used to partially replace portland cement in concrete is also discussed.

36 MATERIALS SCIENCE↗

Femtosecond laser ablation dynamics of fused silica extracted from oscillation of time-resolved reflectivity

Femtosecond laser ablation dynamics of fused silica is examined via time-resolved reflectivity measurements. After optical breakdown was caused by irradiation of a pump pulse with fluence F{sub pump} = 3.3–14.9 J/cm{sup 2}, the reflectivity oscillated with a period of 63 ± 2 ps for a wavelength λ = 795 nm. The period was reduced by half for λ = 398 nm. We ascribe the oscillation to the interference between the probe pulses reflected from the front and rear surfaces of the photo-excited molten fused silica layer. The time-resolved reflectivity agrees closely with a model comprising a photo-excited layer which expands due to the formation of voids, and then separates into two parts, one of which is left on the sample surface and the other separated as a molten thin layer from the surface by the spallation mechanism. Such oscillations were not observed in the reflectivity of soda-lime glass. Whether the reflectivity oscillates or not probably depends on the layer viscosity while in a molten state. Since viscosity of the molten fused silica is several orders of magnitude higher than that of the soda-lime glass at the same temperature, fused silica forms a molten thin layer that reflects the probe pulse, whereas the soda-lime glass is fragmented into clusters.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Mechanical resilience and cementitious processes in Imperial Roman architectural mortar

The pyroclastic aggregate concrete of Trajan’s Markets (110 CE), now Museo Fori Imperiali in Rome, has absorbed energy from seismic ground shaking and long-term foundation settlement for nearly two millenia while remaining largely intact at the structural scale. The scientific basis of this exceptional service record is explored through computed tomography of fracture surfaces and synchroton X-ray microdiffraction analyses of a reproduction of the standardized hydrated lime–volcanic ash mortar that binds decimeter-sized tuff and brick aggregate in the conglomeratic concrete. The mortar reproduction gains fracture toughness over 180 d through progressive coalescence of calcium–aluminum-silicate–hydrate (C-A-S-H) cementing binder with Ca/(Si+Al) ≈ 0.8–0.9 and crystallization of strätlingite and siliceous hydrogarnet (katoite) at ≥90 d, after pozzolanic consumption of hydrated lime was complete. Platey strätlingite crystals toughen interfacial zones along scoria perimeters and impede macroscale propagation of crack segments. In the 1,900 year old mortar, C-A-S-H has low Ca/(Si+Al) ≈ 0.45–0.75. Dense clusters of 2- to 30-µm strätlingite plates further reinforce interfacial zones, the weakest link of modern cement-based concrete, and the cementitious matrix. These crystals formed during long-term autogeneous reaction of dissolved calcite from lime and the alkali-rich scoriae groundmass, clay mineral (halloysite), and zeolite (phillipsite and chabazite) surface textures from the Pozzolane Rosse pyroclastic flow, erupted from the nearby Alban Hills volcano. The clast-supported conglomeratic fabric of the concrete presents further resistance to fracture propagation at the structural scale.

36 MATERIALS SCIENCE↗

Mesoscale Phase Field Modeling of Glass Strengthening Under Triaxial Compression

Recent hydraulic bomb and confined sleeve tests on transparent armor glass materials such as borosilicate glass and soda-lime glass showed that the glass strength was a function of confinement pressure. The measured stress-strain relation is not a straight line as most brittle materials behave under little or no confinement. Moreover, borosilicate glass exhibited a stronger compressive strength when compared to soda-lime glass, even though soda-lime has higher bulk and shear moduli as well as apparent yield strength. To better understand these experimental findings, a mesoscale phase field model is developed to simulate the nonlinear stress versus strain behaviors under confinement by considering heterogeneity formation under triaxial compression and the energy barrier of a micro shear banding event (referred to as pseudo-slip hereafter) in the amorphous glass. With calibrated modeling parameters, the simulation results demonstrate that the developed phase field model can quantitatively predict the pressure-dependent strength, and it can also explain the difference between the two types of glasses from the perspective of energy barrier associated with a pseudo-slip event.

Confined Borosilicate↗

Logic in Memory Emulator

Logic in Memory Emulator (LiME) is a hardware/software tool specially designed for memory system evaluation and experiment. Emerging memories display a wide range of bandwidths, latencies, and capacities, making it challenging for the computer architect to navigate the design space of potential memory configurations, and for the application developer to assess performance implications of using such memories. With the LiME framework, architectural ideas can be prototyped in great detail yet with sufficient performance to support realistic evaluation on long running applications. LiME consists of two fundamental components: 1) the hardware and OS infrastructure for the emulator, and 2) a suite of benchmark applications to assist in characterizing the performance of current and future computer architectures. Some of the applications have been collected from other open source projects. Uses: Logging, replay and analysis of an application's memory behavior Evaluate impact of emerging memory technology on application performance. Emulate complex memory interactions in whole applications orders of magnitude faster than software simulation. Emulate acceleration hardware co-located with the memory subsystem. Features: Capture and log external memory accesses to a separate off-chip memory device without affecting application execution. Memory traces include the address, length, timestamp, and optionally the data for each transaction. Captured trace data can be saved to an SD card for off-line analysis. Configure a wide range of memory latencies in sub-nanosecond increments that encompass highbandwidth and storage class memories. Specify regions of interest (ROI) in applications to reduce the amount of trace data captured for analysis. Currently supports execution on Xilinx Zynq SoC which integrates an ARM processor with FPGA logic on a single device. Applications can be run under Linux or in bare metal mode on the ARM cores.

Jain, AbhishekK↗

Explaining machine-learning models for gamma-ray detection and identification

As more complex predictive models are used for gamma-ray spectral analysis, methods are needed to probe and understand their predictions and behavior. Recent work has begun to bring the latest techniques from the field of Explainable Artificial Intelligence (XAI) into the applications of gamma-ray spectroscopy, including the introduction of gradient-based methods like saliency mapping and Gradient-weighted Class Activation Mapping (Grad-CAM), and black box methods like Local Interpretable Model-agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP). In addition, new sources of synthetic radiological data are becoming available, and these new data sets present opportunities to train models using more data than ever before. In this work, we use a neural network model trained on synthetic NaI(Tl) urban search data to compare some of these explanation methods and identify modifications that need to be applied to adapt the methods to gamma-ray spectral data. We find that the black box methods LIME and SHAP are especially accurate in their results, and recommend SHAP since it requires little hyperparameter tuning. We also propose and demonstrate a technique for generating counterfactual explanations using orthogonal projections of LIME and SHAP explanations.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Reduction of Glass Surface Reflectance by Ion Beam Surface Modification

This is the final report for DOE contract DE-EE0000590. The purpose of this work was to determine the feasibility of the reduction of the reflection from the front of solar photovoltaic modules. Reflection accounts for a power loss of approximately 4%. A solar module having an area of one square meter with an energy conversion efficiency of 18% generates approximately 180 watts. If reflection loss can be eliminated, the power output can be increased to 187 watts. Since conventional thin-film anti-reflection coatings do not have sufficient environmental stability, we investigated the feasibility of ion beam modification of the glass surface to obtain reduction of reflectance. Our findings are generally applicable to all solar modules that use glass encapsulation, as well as commercial float glass used in windows and other applications. Ion implantation of argon, fluorine, and xenon into commercial low-iron soda lime float glass, standard float glass, and borosilicate glass was studied by implantation, annealing, and measurement of reflectance. The three ions all affected reflectance. The most significant change was obtained by argon implantation into both low-iron and standard soda-lime glass. In this way samples were formed with reflectance lower than can be obtained with a single-layer coatings of magnesium fluoride. Integrated reflectance was reduced from 4% to 1% in low-iron soda lime glass typical of the glass used in solar modules. The reduction of reflectance of borosilicate glass was not as large; however borosilicate glass is not typically used in flat plate solar modules. Unlike conventional semiconductor ion implantation doping, glass reflectance reduction was found to be tolerant to large variations in implant dose, meaning that the process does not require high dopant uniformity. Additionally, glass implantation does not require mass analysis. Simple, high current ion implantation equipment can be developed for this process; however, before the process can be employed on full scale solar modules, equipment must be developed for ion implanting large sheets of glass. A cost analysis shows that the process can be economical. Our finding is that the reduction of reflectance by ion beam surface modification is technically and economically feasible. The public will benefit directly from this work by the improvement of photovoltaic module efficiency, and indirectly by the greater understanding of the modification of glass surfaces by ion beams.

14 SOLAR ENERGY↗