An official website of the United States government
Limestone calcined clay cement (LC3) is a sustainable alternative to ordinary Portland cement, capable of reducing the binder’s carbon footprint by 40% while satisfying all key performance metrics. The inherent compositional heterogeneity in select components of LC3, combined with their convoluted chemical interactions, poses challenges to conventional analytical models when predicting mechanical properties. Although some studies have employed machine learning (ML) to predict the mechanical properties of LC3, many have overlooked the pivotal role of feature selection. Proper feature selection not only refines and simplifies the structure of ML models but also enhances these models’ prediction performance and interpretability. This research harnesses the power of the random forest (RF) model to predict the compressive strength of LC3. Three feature reduction methods—Pearson correlation, SHapley Additive exPlanations, and variable importance—are employed to analyze the influence of LC3 components and mixture design on compressive strength. Practical guidelines for utilizing these methods on cementitious materials are elucidated. Through the rigorous screening of insignificant variables from the database, the RF model conserves computational resources while also producing high-fidelity predictions. Additionally, a feature enhancement method is utilized, consolidating numerous input variables into a singular feature while feeding the RF model with richer information, resulting in a substantial improvement in prediction accuracy. Overall, this study provides a novel pathway to apply ML to LC3, emphasizing the need to tailor ML models to cement chemistry rather than employing them generically.
more »
« less
Predicting mechanical properties of ultrahigh temperature ceramics using machine learning
Abstract Ultrahigh temperature ceramics (UHTCs) have melting points above 3000°C and outstanding strength at high temperatures, thus making them apposite structural materials for high‐temperature applications. Di‐borides, nitride, and carbide compounds—processed via various techniques—have been extensively studied and used in the manufacture of UHTCs. Current analytical models, based on our current but incomplete understanding of the theory, are unable to produce a priori predictions of mechanical properties of UHTCs based on their mixture designs and processing parameters. As a result, researchers have to rely on experiments—which are often costly and time‐consuming—to understand composition–structure–performance links in UHTCs. This study employs machine learning (ML) models (i.e., random forest and artificial neural network models) to predict Young's modulus, flexural strength, and fracture toughness of UHTCs in relation to a wide range of mixture designs, processing parameters, and testing conditions. Outcomes demonstrate that adequately trained ML models can yield reliable predictions, a priori, of the three aforesaid mechanical properties. The prediction performance on Young's modulus is superior to flexural strength and fracture toughness. Next, the ML model with the best prediction performance is utilized to evaluate and rank the impacts of input variables on Young's modulus. Finally, on the basis of such classification of consequential and inconsequential input variables, this study develops an easy‐to‐use, closed‐form analytical model to predict Young's modulus of UHTCs. Overall, this study highlights the ability of data‐driven numerical models to complement, or even replace, time‐consuming experiments, thereby accelerating the development of UHTCs.
more »
« less
- PAR ID:
- 10370372
- Publisher / Repository:
- Wiley-Blackwell
- Date Published:
- Journal Name:
- Journal of the American Ceramic Society
- Volume:
- 105
- Issue:
- 11
- ISSN:
- 0002-7820
- Format(s):
- Medium: X Size: p. 6851-6863
- Size(s):
- p. 6851-6863
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Zirconium carbide (ZrC) powder, batched to a ratio of 0.98 C/Zr, was prepared by carbothermal reduction of ZrO2with carbon black. Nominally phase‐pure ZrC powder had a mean particle size of 2.4 μm. The synthesized powder was hot‐pressed at 2150°C to a relative density of > 95%. The mean grain size was 2.7 ± 1.4 μm with a maximum observed grain size of 17.5 μm. The final hot‐pressed billets had a C/Zr ratio of 0.92, and oxygen content of 0.5 wt%, as determined by gas fusion analysis. The mechanical properties of ZrC0.92O0.03were measured at room temperature. Vickers’ hardness decreased from 19.5 GPa at a load of 0.5 kgf to 17.0 GPa at a load of 1 kgf. Flexural strength was 362.3 ± 46 MPa, Young's modulus was 397 ± 13 MPa, and fracture toughness was 2.9 ± 0.1 MPa•m1/2. Analysis of mechanical behavior revealed that the largest ZrC grains were the strength‐limiting flaw in these ceramics.more » « less
-
Calcium aluminate cement (CAC) has been explored as a sustainable alternative to Portland cement, the most widely used type of cement. However, the hydration reaction and mechanical properties of CAC can be influenced by various factors such as water content, Li2CO3 content, and age. Due to the complex interactions between the precursors in CAC, traditional analytical models have struggled to predict CAC binders’ compressive strength and porosity accurately. To overcome this limitation, this study utilizes machine learning (ML) to predict the properties of CAC. The study begins by using thermodynamic simulations to determine the phase assemblages of CAC at different ages. The XGBoost model is then used to predict the compressive strength, porosity, and hydration products of CAC based on the mixture design and age. The XGBoost model is also used to evaluate the influence of input parameters on the compressive strength and porosity of CAC. Based on the results of this analysis, a closed-form analytical model is developed to predict the compressive strength and porosity of CAC accurately. Overall, the study demonstrates that ML can be effectively used to predict the properties of CAC binders, providing a valuable tool for researchers and practitioners in the field of cement science.more » « less
-
Thermoplastic elastomers based on ABA triblock copolymers are typically limited in modulus and strength due to crack propagation within the brittle regions when the hard end-block composition favors morphologies that exhibit connected domains. Increasing the threshold end-block composition to achieve enhanced mechanical performance is possible by increasing the number of junctions or bridging points per chain, but these copolymer characteristics also tend to increase the complexity of the synthesis. Here, we report an in situ polymerization method to successfully increase the number of effective junctions per chain through grafting of poly(styrene) (PS) to a commercial thermoplastic elastomer, poly(styrene)–poly(butadiene)–poly(styrene) (SBS). The strategy described here transforms a linear SBS triblock copolymer–styrene mixture into a linear-comb-linear architecture in which poly(styrene) (PS) grafts from the mid-poly(butadiene) (PBD) block during the polymerization of styrene. Through systematic variation in the initial SBS/styrene content, nanostructural transitions from disordered spheres to lamellar through reaction-induced phase transitions (RIPT) were identified as the styrene content increased. Surprisingly, maximum mechanical performance (Young's modulus, tensile strength, and elongation at break) was obtained with samples exhibiting lamellar nanostructures, corresponding to overall PS contents of 61–77 wt% PS (including the original PS in SBS). The PS grafting from the PBD block increases the modulus and the strength of the thermoplastic elastomer while preventing brittle fracture due to the greater number of junctions afforded by the PS grafts. The work presented here demonstrates the use of RIPT to transform standard SBS materials into polymer systems with enhanced mechanical properties.more » « less
-
Strong, tough, and lightweight composites are increasingly needed for diverse applications, from wind turbines to cars and aircraft. These composites typically contain sheets of strong and high-modulus fibers in a matrix that are joined with other materials to resist fracture. Coupling these dissimilar materials together is challenging to enhance delamination properties at their interface. We herein investigate using a trace amount of carbon nanotube sheets to improve the coupling between composite skins and core in a composite sandwich. Ultra-thin (~100 nm) forest-drawn multi-walled carbon nanotube (MWNT) sheets are interleaved within the skin/core interphase, with MWNTs aligned in the longitudinal direction. The mechanical behavior is characterized by end notched flexural testing (ENF). With two MWNT sheets placed in the skin/core interphase, the following performance enhancements are achieved: 36.8 % increase in flexural strength; 127.3 % and 125.7 % increases in mode I & II fracture toughness values, respectively; and 152.8 % increase in interfacial shear strength (IFSS). These are achieved with negligible weight gain of the composite sandwich (0.084 wt% increase over the baseline sandwich without MWNT sheets). The finite element simulation results show that MWNT sheets enhance the skin/core coupling by reducing stress concentration, enabling the transition from catastrophic brittle failure to a stable ductile failure mode. The MWNT sheets interleaved sandwich composites are thus demonstrated to be stronger and tougher while providing electrical conductivity (4.3 × 104 S/m) at the skin/core interface for potential de-icing, electromagnetic interference shielding, and structural health monitoring.more » « less
