A deep learning perspective into the figure-of-merit of thermoelectric materials

Jaafreh, Russlan; Seong, Kang Yoo; Kim, Jung-Gu; Hamad, Kotiba

doi:10.1016/j.matlet.2022.132299

Cited by 14 publications

(9 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Machine Learning Proceduresmentioning

confidence: 99%

“…Feature generation for materials can be done through numerous ways depending on the scale of the intended description that is needed in a specific research. [26,16] In this work, the feature generation process is conducted according to the method developed by Ward et al [27,28] and using their proprietary software (Magpie). [29] In this method, a compound with a specific crystal structure can be described based on features that consist of two categories, the crystal and chemical composition features.…”

Section: Feature Generation and Processingmentioning

confidence: 99%

See 1 more Smart Citation

A Machine Learning‐Assisted Approach to a Rapid and Reliable Screening for Mechanically Stable Perovskite‐Based Materials

Jaafreh

Sharan

Sajjad

et al. 2022

Adv Funct Materials

Self Cite

View full text Add to dashboard Cite

The present work is designed to discover new perovskite-based materials, which are expected to show high mechanical stability during their applications, using machine learning (ML) techniques, and based on the Pugh's criterion for distinguishing brittle and ductile behaviors. For this purpose, ML models to predict the moduli of materials, bulk (B) and shear (G), are built using their crystal structure and composition information. The ML process is initiated with the information of 5663 compounds, including composition, crystal structure and moduli, as listed in AFLOW database. Following a procedure of data characteristics, feature generation, feature processing, training, and testing, the ML models are constructed with acceptable accuracy (tenfold cross-validation R 2 score of 0.90 and 0.89 for B and G, respectively). The validation process of the models, which is conducted using the corresponding density functional theory calculations, reveals that these models are reliable to be employed in a largescale screening process. Indeed, the B-and G-based ML models are incorporated in a screening process, and this is also conjugated with other screening criterions, to find out thermodynamically stable and formable perovskite-based materials with improved mechanical performance.

show abstract

Section: Machine Learning Proceduresmentioning

confidence: 99%

Section: Feature Generation and Processingmentioning

confidence: 99%

A Machine Learning‐Assisted Approach to a Rapid and Reliable Screening for Mechanically Stable Perovskite‐Based Materials

Jaafreh

Sharan

Sajjad

et al. 2022

Adv Funct Materials

Self Cite

View full text Add to dashboard Cite

show abstract

“…Using ML to learn models that predict the output of ab initio calculations is sensible, since invoking an ML model is much faster (and less computationally expensive) than carrying out an ab initio calculation. ML models of various thermoelectric properties, such as the Seebeck coefficient [52][53][54][55], electrical conductivity [56,57], power factor [58][59][60][61], lattice thermal conductivity [62][63][64][65][66][67][68][69][70][71][72][73][74], and even zT [75][76][77][78][79][80], have been developed.…”

Section: Introductionmentioning

confidence: 99%

Predicting thermoelectric transport properties from composition with attention-based deep learning

Antunes

Butler

Grau‐Crespo

2023

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

Thermoelectric materials can be used to construct devices which recycle waste heat into electricity. However, the best known thermoelectrics are based on rare, expensive or even toxic elements, which limits their widespread adoption. To enable deployment on global scales, new classes of effective thermoelectrics are thus required. Ab initio models of transport properties can help in the design of new thermoelectrics, but they are still too computationally expensive to be solely relied upon for high-throughput screening in the vast chemical space of all possible candidates. Here, we use models constructed with modern machine learning techniques to scan very large areas of inorganic materials space for novel thermoelectrics, using composition as an input. We employ an attention-based deep learning model, trained on data derived from ab initio calculations, to predict a material’s Seebeck coefficient, electrical conductivity, and power factor over a range of temperatures and n- or p-type doping levels, with surprisingly good performance given the simplicity of the input, and with significantly lower computational cost. The results of applying the model to a space of known and hypothetical binary and ternary selenides reveal several materials that may represent promising thermoelectrics. Our study establishes a protocol for composition-based prediction of thermoelectric behaviour that can be easily enhanced as more accurate theoretical or experimental databases become available.

show abstract

“…ZT value largely determines the efficiency of power generation and thermoelectric refrigeration [5–8] . Therefore, increasing ZT is a key in thermoelectric materials research [3,9,10] …”

Section: Introductionmentioning

confidence: 99%

“…[5][6][7][8] Therefore, increasing ZT is a key in thermoelectric materials research. [3,9,10] Among the popular thermoelectric materials, the Zintlphase Mg 3 Sb 2 compound [11][12][13][14][15] stands out with the merits of high Seebeck coefficient, low lattice thermal conductivity, and consisting of inexpensive and eco-friendly constituents, making it promising for future applications. [16] However, the key challenge of this material lies in the poor electrical conductivity, mainly due to its low carrier concentration and mobility.…”

Section: Introductionmentioning

confidence: 99%

Effects of Bi Alloying on the Thermoelectric Properties of p‐Type Mg₃Sb_2‐xBi_x Solid Solutions

Zhu

et al. 2022

ChemNanoMat

View full text Add to dashboard Cite

Bismuth is the most common alloying element for high-performance Mg 3 Sb 2 -based thermoelectric compounds. However, the electrical and thermal transport data is lacking for undoped Mg 3 (Sb, Bi) 2 solid solutions. In this study, theoretical calculation results show that Bi alloying can reduce the band gap of Mg 3 Sb 2 , favoring the regulation of carrier concentration. Both Mg 3 Sb 2 and Mg 3 SbBi compounds present a single band structure on the valence band maximum at Γ of the Brillouin zone. Incorporating the Bi element can slightly increase the density of states (DOS) on the valence band maximum, which is beneficial for the Seebeck coefficient. Mg 3 Sb 2-x Bi x (0 � x � 0.8) compounds were experimentally prepared via ball milling (BM) combined with the spark plasma sintering (SPS) technique. The carrier concentration, electrical conductivity, and power factor of the samples tend to increase with the increasing Bi content. Bismuth alloying significantly reduces the total thermal conductivity in the low-temperature range, with the lowest value of 0.65 Wm À 1 K À 1 obtained at ~400 K for the x = 0.2 sample. The peak ZT value of 0.077 at 800 K and the average ZT of 0.051 in the temperature range of 450-800 K are obtained for the x = 0.2 sample, which is 93% and 219% higher than those of the unalloyed sample, respectively. This work reveals the Bi-alloying induced transport mechanism and provides fundamental thermoelectric data regarding the undoped Mg 3 Sb 2-x Bi x (0 � x � 0.8) solid solutions.

show abstract

A deep learning perspective into the figure-of-merit of thermoelectric materials

Cited by 14 publications

References 10 publications

A Machine Learning‐Assisted Approach to a Rapid and Reliable Screening for Mechanically Stable Perovskite‐Based Materials

A Machine Learning‐Assisted Approach to a Rapid and Reliable Screening for Mechanically Stable Perovskite‐Based Materials

Predicting thermoelectric transport properties from composition with attention-based deep learning

Effects of Bi Alloying on the Thermoelectric Properties of p‐Type Mg₃Sb_2‐xBi_x Solid Solutions

Contact Info

Product

Resources

About

A deep learning perspective into the figure-of-merit of thermoelectric materials

Cited by 14 publications

References 10 publications

A Machine Learning‐Assisted Approach to a Rapid and Reliable Screening for Mechanically Stable Perovskite‐Based Materials

A Machine Learning‐Assisted Approach to a Rapid and Reliable Screening for Mechanically Stable Perovskite‐Based Materials

Predicting thermoelectric transport properties from composition with attention-based deep learning

Effects of Bi Alloying on the Thermoelectric Properties of p‐Type Mg3Sb2‐xBix Solid Solutions

Contact Info

Product

Resources

About

Effects of Bi Alloying on the Thermoelectric Properties of p‐Type Mg₃Sb_2‐xBi_x Solid Solutions