Page 135 - Read Online

Page 12 of 21 Chen et al. Energy Mater. 2025, 5, 500120 https://dx.doi.org/10.20517/energymater.2024.311

the elements tailored to reduce material waste, cost, and time consumption is also crucial. He et al. reported
[56]
the prediction of superior thermoelectric performance in unexplored doped-BiCuSeO via ML . The
experimental data of Bi M CuSeO (M represents Ca, Mg, Sr, Ba, Na, La, Sm, Er, Ho, Cd, Sb, Pb, Ag, Al, Fe,
x
1-x
K, Rb, Co, Cs, Mn, Nd, Sn, Yb, Zn, and Ni) was collected from the literature. The descriptors, a set of input
parameters such as temperature, the content of the doping element, the relative molecular mass of
Bi M CuSeO, the Mendeleev number of dopant, the Pauling electronegativity of dopant, the first ionization
1-x
x
energy of dopant, the ionic radius of Bi (r) and the Pauling electronegativity of Bi , where x ranges from 0
1-x
1-x
to 0.2 referred the doping content, used to generate a ML model. The workflow from data collection to ML
prediction with the application of ML is shown in Figure 6A. Totally, six ML approaches were developed to
solve this problem. Among these, GBR has an R of 0.96, indicating better fitting with experimental input
2
data. The correlation between descriptors and the ZT helps identify the new composition's ZT through the
ML model. Figure 6B shows the ZT of the experimentally obtained value of doped elements and the
predicted value. The optimized content of Bi Po CuSeO and Bi Cs CuSeO has improved ZT by 104%
0.14
0.86
0.12
0.88
and 98% at 923 K, respectively. The employed ML approach identified how to analyze the suitable doping
element among the more elements for TE performance enhancement with the input help of descriptors.
Minhas et al. used the database of synthesized materials trained with ML models to predict the suitable
doping elements to dope into GeTe, SnTe, PbTe, Sn Se, Bi Sb Te ), skutterudite (CoSb , As Te ), clathrates
3
3
2
3
x
1-x
2-x
(Ba Ga Ge ), and transition metal-based chalcogenides (Cu Se,Ag Te ) . Among various ML models,
[57]
2
8
2-x
30
2-x
16
eXtreme gradient boosting regression (XGBR) has the best fitting with experimental database. The
maximum ZT of 2.20 at 1,000 K was identified for Bi Sb Te . The correlation between the materials
0.1
3
1.9
descriptors and the output parameter helped to find the new high-performance material. The descriptors
selected were based on the material chemistry influencing the structural and transport properties. Also, the
classification of the material was based on the performance of the RF classifier ML model in the different
structures.
The ML model is accelerating to find a new TE material and highly efficient dopant. Parse et al. constructed
an ML model to discover a new dopant to place in the Bi site in BiCuSeO . The model design focused on
[58]
the accuracy improvement by normalizing ZT of doped BiCuSeO with pristine BiCuSeO. The developed
2
model produced the best fitting data with experimental. The R value of 0.93 from the developed extra tree
regression model was much closer to 1, indicating more accuracy than the initial model. New dopants were
discovered through this model without wasting the material, which satisfied the thermoelectric principle.
Based on various descriptors, the selected Si as a dopant for the Bi site improves the ZT by increasing
mobility. Table 1 lists the survey of this review and refers to the descriptors, targets, and findings of the
study related to ML in material chemistry.
Machine learning in TE materials synthesis
The efficient synthesis of TE materials requires understanding the connection between various parameters
involving synthesis conditions. In this direction, ML has great potential to control the synthesis conditions
of novel TE materials. By learning existing synthesis information, ML can recommend efficient synthesis
conditions with few trials. Thus, the ML can be used to understand complex relationships and predict
optimal synthesis conditions with a high probability of success by employing existing initial synthesis data
of TE material. Tang et al. have explored the feasibility of ML for guiding the MoS synthesis by Chemical
2
Vapor Deposition (CVD) and hydrothermal synthesis of carbon nanostructures for the first time .
[59]
[59]
Figure 7 shows the schematic of ML workflow for materials synthesis developed by Tang et al. . This study
has constructed an ML model for the general control of synthesis parameters for future experiments. The
model could predict the probability of success and recommend the optimal synthesis conditions.
Furthermore, a progressive adaptive model (PAM) to maximize the outcome of synthesis experiments
through minimum trials has been introduced. This successful ML methodology for CVD of MoS has the
2