决策树
化学
结晶
树(集合论)
相(物质)
二叉树
二进制数
二元决策图
形态学(生物学)
组合化学
计算机科学
算法
有机化学
人工智能
数学
组合数学
算术
生物
遗传学
作者
Kaiyu Guan,Jingxian Wu,Jing Zhou,Yang Li,Lingnan Pei,Xin Shi
标识
DOI:10.1021/acs.inorgchem.3c03263
摘要
The morphology control of metal phosphonates is always a difficulty because there are many challenges derived from the complexity of crystallization and the multivariable synthesis system. Responding to challenges, we propose a synthesis strategy guided by a decision tree for morphology control of metal phosphonates, through which directional design of the morphology-controlled synthesis can be realized. Specifically, any one synthetic condition involving the synthesis of metal phosphonates can be regarded as a decision problem to construct a binary decision tree. By means of the classification principle of the binary decision tree, the samples synthesized under the boundary value of each synthesis condition are classified based on crystal phase and morphology. The key synthetic conditions determining crystal phase and morphology can be precisely screened out to serve as decision nodes for the binary decision tree and are also rapidly optimized by the recursion level by level, whereas others cannot. Here, the β-polymorph of copper phenylphosphonate (β-CuPP) is selected as an example to elaborate the decision-tree-guided synthesis strategy for morphology control of metal phosphonates. From the constructed binary decision tree, it is clear that the right amount of methanol in the solvent is vital to obtain β-phase of CuPP, whereas the reactant concentration, pH value, and reaction time are important for morphology and phase transformation. Under the optimal synthetic conditions screened out by the binary decision tree, β-CuPP can thus be controlled to be hierarchically flower-like microsphere morphology through either the direct synthesis route or the solid-to-solid phase transformation route. This research work confirms that the decision-tree-guided synthesis is highly efficacious for the morphology control of metal phosphonates. Furthermore, the morphology-controlled synthesis guided by a decision tree may provide some valuable inspiration for morphology control of metal-organic frameworks (MOFs) and even coordinate compounds.
科研通智能强力驱动
Strongly Powered by AbleSci AI