Size‐Controllable Eu‐MOFs through Machine Learning Technology: Application for High Sensitive Ions and Small‐Molecular Identification

Abstract

high-resolution, especially in the solid state or high concentration liquid states. [5][6][7] Dong et al. constructed two Ln-MOFs {[Ln(H 3 L)(H 2 O)] • 7H 2 O} n (Ln = Eu and Tb, H 3 L = 5-(3',5'-dicarboxylphenyl)picolinic acid), and utilized these two kinds of Ln-MOFs to detect the Fe 3+ and chromate ions. [8] And the detection limits of Fe 3+ / CrO 4 2-/Cr 2 O 7 2are 0.67/0.53/0.32 × 10 −6 m for europium-based metal-organic frameworks (Eu-MOFs) and 1.26/0.75/0.57 × 10 −6 m for Tb-MOFs, respectively. Cui et al. synthesized the functional Eu-MOFs by selecting Eu 3+ and 5-boronobenzene-1,3-dicarboxylic acid (BBDC) as metal node and organic ligand, respectively. [9] Due to the special nucleophilic reaction between H 2 O 2 and boric group, these Eu-MOFs can be used to detect the concentration of H 2 O 2 and glucose, and their low detection limit concentrations are 0.0335 × 10 −6 and 0.0643 × 10 −6 m, respectively. Zhang et al. also synthesized five kinds of 3D Ln-MOFs as [Ln 4 (µ 6 -L) 2 (µ-HCOO)(µ 3 -OH)3(µ 3 -O) (DMF)2(H 2 O) 4 ] n (Ln = Tb, Eu, Gd, Dy, and Er), and reflected that Tb-MOFs showed an high efficiency and selectivity in detecting the acetone and metal (III) cations (Fe 3+ and Ce 3+ ). [10] Expecting for the effect of precursors components (such as rare-salt, organic ligand, solvent and their composition ratio, etc.) discussed above, the morphology and size of MOFs also reported to have an significant influence on the optical properties of AIE characteristic MOFs. [11][12][13] For instance, the size of MOFs not only affects their specific surface area and reaction Metal-organic frameworks (MOFs) with the aggregation-induced emission (AIE) activities exhibit potential applications in the fields of energy and biomedical technology. However, the controllable synthesis of MOFs in the varied particle sizes not only affects their AIE activities, but also restricts their application scenarios. In this work, the varied particle sizes of Eu-MOFs are synthesized by adjusting the synthesis process parameters, and their variation rules combining the single factor analysis method with machine learning technology are studied. Based on the R 2 score, the gradient boosting decision tree (GBDT) regression model (0.9535) is employed to calculate the weight and correlation between different synthesis process parameters and it is shown that all these parameters have synergic effects on the particle sizes of Eu-MOFs, and the Eu-precursors concentration dominates in their synthesis process. Furthermore, it is indicated that the large size of Eu-MOFs and strong structural stability contribute to their high AIE activities. Finally, a screen-printed pattern is fabricated using the sample of "120-0.3-6," and this pattern exhibits a bright red fluorescence under the UV light. More importantly, this kind of Eu-MOFs can also be used to identify varied ions (Fe 3+ , F -, I -, SO 4 2-, CO 3 2-, and PO 4 3-) and citric acid.The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002...

Enhanced Aggregation-Induced Emission Activity of Metal–Organic Frameworks by Using Machine Learning Technology

“…Previous works have shown that the Eu can be used as the metal node and 5-bop and 1,3-BDC can be used as the mixed ligands to form a repeated and ordered 3D spatial structure . Generally, the discrete carboxylate group of 1,3-BDC can be used to link the Eu 1 and Eu 2 atoms alternately and form 1D chains.…”

“…Particularly, Eu-MOFs dissolved in ethanol exhibit the highest intensities in the wavelengths of 591, 615, and 699 nm, but the intensity of Eu-MOFs dissolved in ethanol in a wavelength of 734 nm is quite lower than those in N -hexanol, DMF, and glycerol. Previous works have shown that although the steady PL intensities of Ln-MOFs affected by various solvents, ions, and small-molecule solutions, their 3D structures can still be preserved. − Relevant works have also pointed out that the fluorescence intensity decrease phenomenon of Eu-MOFs in the static quenching state can be summarized as the 5 D 0 → 7 F J ( J = 0, 1, 2, 3, and 4) transitions of Ln 3+ ions, the energy transfer between the Ln-MOFs and solutions, the facilitation of the porous structure of Ln-MOFs on the combination of the Lewis acid ions and the pyridine Lewis base functionality of Ln-MOFs, and the weak effect of ions-O interactions. ,, Herein, it can be found in Figure J that the time-solved PL intensities of Eu-MOFs in the organic solution, such as ethanol, DMF, glycerol, and N -hexanol, are lower than that of Eu-MOFs in the water. However, their steady PL intensities of Eu-MOFs in Figure I are higher than of Eu-MOFs in the water.…”

“…MOFs reveal abundant structural diversity, versatile surface chemistry, low toxicity, high stability without photobleaching, multi-emission characteristic peaks, and good bio-compatibility. , Recently, MOFs with aggregation-induced emission (AIE) activity further show the potential application advantages in the fields of energy, information, electronic, and biomedical technologies. , Generally, these MOFs with AIE activity possess several characteristics: (1) the emission intensity increases with the concentration, especially in the solid state (powder or high-concentration solution); (2) high sensitivity in the solid or high-concentration solution state; (3) high-resolution in cell imaging and related biological imaging technology; and (4) controllable emission characteristics by chemical modification based on metal nodes or organic ligands . Furthermore, recent works also reported that the AIE activity of MOFs is significantly affected by the solvent, anion, and cation; thus, MOFs could be used to identify these solvents and ions. ,, Besides, it is necessary to investigate the various mechanisms of MOFs in various solvents to obtain a high AIE activity of Eu-MOFs, which could be suitable for screen-printing.…”

Reshaping the material research paradigm of electrochemical energy storage and conversion by machine learning

“…Gradient Boosted Decision Tree (GBDT) is also called as Multiple Additive Regression Tree (MART), which is also a decision tree‐based algorithm like RF (Figure 2C). 47,48 But the “boosting thought” decides that these two algorithms work in different ways. The boosting method trains the base classifier in series, and there is dependence between each base classifier.…”

Machine learning prediction of open metal sites in metal-organic framework catalysts

“…Random Forest (RF) and Extreme Gradient Boosting (XGBoost) are two of the numerous algorithms widely implemented in MOF-related fields. They were proved to perform well in predicting adsorption and diffusion properties of various chemical species [35][36][37][38][39][40][41][42] , electronic and force field parameters 43,44 , and catalytic abilities including carbon fixing and photodegradation 45,46 .…”