Research Progress of Quantitative Structure-Property Relationship Analysis Method in the Field of Glass Materials

doi:10.16552/j.cnki.issn1001-1625.2025.1143

Abstract

Abstract:

The complex amorphous structure and metastable thermodynamic characteristics of glass materials pose challenges to understanding the mapping relationship between the structure and the performance of this kind of materials. Therefore， the structure-properties relationship becomes one of the core challenges in developing new high-performance glass. The quantitative structure-property relationship （QSPR） analysis method， which can establish a mathematical model to link the microscopic structure with macroscopic properties， provides a new approach to break through the limitation of traditional trial-and-error methods in designing new glass materials. This paper systematically reviews the latest progresses of QSPR analysis method in the field of glass materials. Firstly， it sorts out the complete process of the QSPR analysis method， including the data acquisition， the descriptor extraction， and the model construction. Secondly， it focuses on the classification of descriptor systems and elaborates on the development and evolution of descriptor systems in terms of structure， energy and dynamic mechanisms. Then， successful prediction cases using QSPR analysis method are further summarized， including mechanical properties， thermal properties and chemical durability in silicate， phosphate， and other glass systems， demonstrating its ability in practical material design. Finally， current challenges in understanding the physical meaning of descriptors and improving the data quality are introduced， and outlook of the future improving strategy by integrating multi-scale information and combining multi-methods are proposed， in order to promote the QSPR analysis method from an explanatory tool into a precise design platform.

Key words: quantitative structure-property relationship, glass material, structural descriptor, molecular dynamics simulation, property prediction, material design

CLC Number:

TQ171

YAN Hongying, YAN Jingping, JIANG Fangling, ZHENG Qiuju, DENG Lu. Research Progress of Quantitative Structure-Property Relationship Analysis Method in the Field of Glass Materials[J]. BULLETIN OF THE CHINESE CERAMIC SOCIETY, 2026, 45(3): 755-770.

Figures/Tables 9

Fig.1 Development history of QSPR analysis method in the field of glass materials

Fig.2 Research process of QSPR analysis method for glass materials

Fig.3 Experimentally measured density （a）， hardness （b）， glass transition temperature （c） and use coordination number as structure input to explore correlation between structural descriptor Fnet， which is used to evaluate the overall strength of glass network， and density （d）， hardness （e）， glass transition temperature （f）， and use Q n as structural input to explore correlation between Fnet and density （g）， hardness （h）， glass transition temperature （i）［43］

Fig.4 Workflow of QSPR modeling［50］

Table 1 Fnet descriptor provides values R2 for density， glass transition temperature， CTE， Young’s modulus and hardness［14］

Item	F_net	R²
Density	BSdc/T_m with M_NC	0.865
	BSdc with M_NC	0.838
	SBS/T_m	0.813
T_g	EF/T_mwith M_NC	0.732
T_g	EF with M_NC	0.711
CTE	SBS with M_NC	0.973
	SBS/T_m with M_NC	0.972
	BSdm with M_NC	0.971
	BSdm/T_m with M_NC	0.943
Young’s modulus	EF/T_m	0.988
	EF/T_m with M_NC	0.953
	EF	0.987
	EF with M_NC	0.958
Hardness	EF	0.933
	EF/T_m	0.925
	EF/T_m with M_NC	0.844

Table 2 Ccoe coefficients associated with different properties in SAP series and silicate glass［43］

Structure	Density	Hardness	T_g	CTE	Young’s modulus	η_P	η_Al	η_Si
CN	1	1/R'_M	R_Al	1/R'_M	1	1/R'_p	R $A l'$	R $S i'$
Q ⁿ	R'_M	1/R'_M	1	1/R'_M	1	1/R'_p	R $A l'$	R $S i'$

Table 2 Ccoe coefficients associated with different properties in SAP series and silicate glass［43］

Structure	Density	Hardness	T_g	CTE	Young’s modulus	η_P	η_Al	η_Si
CN	1	1/R'_M	R_Al	1/R'_M	1	1/R'_p	R $A l'$	R $S i'$
Q ⁿ	R'_M	1/R'_M	1	1/R'_M	1	1/R'_p	R $A l'$	R $S i'$

Table 3 QSPR hybrid descriptors in the field of glass materials

Development stage	Descriptor	Key improvement	Advantage	Limitation
Early model	F_net-BE^［31］	Combination of CN and BE	Pioneering the fusion of energy and structural parameters.	Dependent on MD simulation， limited application scope
Energy-related developments	F_net-SBS^［13］	Optimized energy parameters： Replace BE with SBS and introduce M_NC	Parameter optimization， significantly enhanced correlation	Dependent on experimental bond energy data
	F_net-FE^［55］	Substitution of EF with formation energy	Good correlation in silicate systems， high prediction accuracy	Strong dependency on FE data， high computational cost
	F_net-CSE^［43］	Using electronegativity to replace bond energy， C_coe is introduced	No experimental input required， strong cross-system applicability	Al CN prediction deviation， neglect of covalent effects
	Average metal oxide dissociation energy^［33］	Integration of metal-oxygen bond strengths and CN	Direct-correlation with dissolution energy barrier， excellent prediction performance	Strong dependency on CN accuracy
Structure-related developments	Three-scale F_net^［16］	Based on multi-scale structural descriptor： CN， Q ⁿ， and ring size	Validation of multi-scale QSPR strategy	Dependent on MD simulation， high computational cost
Structure-related developments	Topological indices descriptor^［60］	Quantification of spatial connectivity complexity in silicate networks	Highly correlated with various properties	Sensitive to methods of model construction
Kinetics-related developments	F_net-variants^［14］	Replace the fixed value m_ij by the NC-based dynamic factor	Identified optimal descriptor forms for different properties	Relatively complex parameters
Kinetics-related developments	Average self-diffusion coefficient of the molten state^［33］	Independent of bond energy parameters	Insensitive to thermal history， good stability	Neglect of external factors such as interface reactions

Table 3 QSPR hybrid descriptors in the field of glass materials

Development stage	Descriptor	Key improvement	Advantage	Limitation
Early model	F_net-BE^［31］	Combination of CN and BE	Pioneering the fusion of energy and structural parameters.	Dependent on MD simulation， limited application scope
Energy-related developments	F_net-SBS^［13］	Optimized energy parameters： Replace BE with SBS and introduce M_NC	Parameter optimization， significantly enhanced correlation	Dependent on experimental bond energy data
	F_net-FE^［55］	Substitution of EF with formation energy	Good correlation in silicate systems， high prediction accuracy	Strong dependency on FE data， high computational cost
	F_net-CSE^［43］	Using electronegativity to replace bond energy， C_coe is introduced	No experimental input required， strong cross-system applicability	Al CN prediction deviation， neglect of covalent effects
	Average metal oxide dissociation energy^［33］	Integration of metal-oxygen bond strengths and CN	Direct-correlation with dissolution energy barrier， excellent prediction performance	Strong dependency on CN accuracy
Structure-related developments	Three-scale F_net^［16］	Based on multi-scale structural descriptor： CN， Q ⁿ， and ring size	Validation of multi-scale QSPR strategy	Dependent on MD simulation， high computational cost
Structure-related developments	Topological indices descriptor^［60］	Quantification of spatial connectivity complexity in silicate networks	Highly correlated with various properties	Sensitive to methods of model construction
Kinetics-related developments	F_net-variants^［14］	Replace the fixed value m_ij by the NC-based dynamic factor	Identified optimal descriptor forms for different properties	Relatively complex parameters
Kinetics-related developments	Average self-diffusion coefficient of the molten state^［33］	Independent of bond energy parameters	Insensitive to thermal history， good stability	Neglect of external factors such as interface reactions

Fig.5 Relationship between Fnet descriptor and experimental density （a）， glass transition temperature （b）， CTE （c）， Young’s modulus （d）， hardness （e）［14］

Fig.6 Correlation between initial dissolution rate r0 and Fnet（a）， Fnet and modified network connectivity （b）， bridge oxygen percentage （c）， overall network connectivity （d）［50］

References 86

[1]	CHEN H Y， QIAN M， CHEN Y K， et al. Monolithic edge-cladding process for the elliptical disk of N31-type Nd-doped high-power laser glass［J］. High Power Laser Science and Engineering， 2022， 10： e14.
[2]	MALKOVSKY V I， YUDINTSEV S V， OJOVAN M I， et al. The influence of radiation on confinement properties of nuclear waste glasses［J］. Science and Technology of Nuclear Installations， 2020， 2020（1）： 8875723.
[3]	FU Q， SAIZ E， RAHAMAN M N， et al. Bioactive glass scaffolds for bone tissue engineering： state of the art and future perspectives［J］. Materials Science and Engineering： C， 2011， 31（7）： 1245-1256.
[4]	PANTULAP U， ARANGO-OSPINA M， BOCCACCINI A R. Bioactive glasses incorporating less-common ions to improve biological and physical properties［J］. Journal of Materials Science： Materials in Medicine， 2021， 33（1）： 3.
[5]	JIANG Z H， ZHANG Q Y. The structure of glass： a phase equilibrium diagram approach［J］. Progress in Materials Science， 2014， 61： 144-215.
[6]	ZANOTTO E D， COUTINHO F A B. How many non-crystalline solids can be made from all the elements of the periodic table？［J］. Journal of Non-Crystalline Solids， 2004， 347（1/2/3）： 285-288.
[7]	MAURO J C. Decoding the glass genome［J］. Current Opinion in Solid State and Materials Science， 2018， 22（2）： 58-64.
[8]	KATRITZKY A R， DOBCHEV D A， KARELSON M. Physical， chemical， and technological property correlation with chemical structure： the potential of QSPR［J］. Zeitschrift für Naturforschung B， 2006， 61（4）： 373-384.
[9]	CYPCAR C C， CAMELIO P， LAZZERI V， et al. Prediction of the glass transition temperature of multicyclic and bulky substituted acrylate and methacrylate polymers using the energy， volume， mass （EVM） QSPR model［J］. Macromolecules， 1996， 29（27）： 8954-8959.
[10]	PETROSYAN L S， SIZOCHENKO N， LESZCZYNSKI J， et al. Modeling of glass transition temperatures for polymeric coating materials： application of QSPR mixture-based approach［J］. Molecular Informatics， 2019， 38（8/9）： e1800150.
[11]	GUBSKAYA A V， KHOLODOVYCH V， KNIGHT D， et al. Prediction of fibrinogen adsorption for biodegradable polymers： integration of molecular dynamics and surrogate modeling［J］. Polymer， 2007， 48（19）： 5788-5801.
[12]	KARELSON M， LOBANOV V S， KATRITZKY A R. Quantum-chemical descriptors in QSAR/QSPR studies［J］. Chemical Reviews， 1996， 96（3）： 1027-1044.
[13]	LU X N， DENG L， GIN S， et al. Quantitative structure-property relationship （QSPR） analysis of ZrO₂-containing soda-lime borosilicate glasses［J］. The Journal of Physical Chemistry B， 2019， 123（6）： 1412-1422.
[14]	LU X N， DU J C. Quantitative structure-property relationship （QSPR） analysis of calcium aluminosilicate glasses based on molecular dynamics simulations［J］. Journal of Non-Crystalline Solids， 2020， 530： 119772.
[15]	严静萍，王飞梅，李波源，等. 基于实验、分子动力学模拟和QSPR分析的玻璃力学性能机器学习模型［J］. 硅酸盐学报， 2025， 53（10）： 2808-2819.
	YAN J P， WANG F M， LI B Y， et al. Machine learning models for predicting mechanical properties of glass based on experiments， molecular dynamics simulations and QSPR analysis［J］. Journal of the Chinese Ceramic Society， 2025， 53（10）： 2808-2819 （in Chinese）.
[16]	YAN J P， ZHANG Y J， WANG F M， et al. Effect of Al₂O₃ on the structure-property relationship of sodium aluminophosphate glasses： a combined study of experiments， MD simulations， and QSPR analysis［J］. Journal of the American Ceramic Society， 2023， 106（11）： 6510-6526.
[17]	BENMORE C J， WEBER J R， SAMPATH S， et al. A neutron and X-ray diffraction study of calcium aluminate glasses［J］. Journal of Physics： Condensed Matter， 2003， 15（31）： S2413-S2423.
[18]	HENNET L， DREWITT J W E， NEUVILLE D R， et al. Neutron diffraction of calcium aluminosilicate glasses and melts［J］. Journal of Non-Crystalline Solids， 2016， 451： 89-93.
[19]	SHI Y， GULBITEN O， NEUEFEIND J， et al. Temperature-induced structural change through the glass transition of silicate glass by neutron diffraction［J］. Physical Review B， 2020， 101（13）： 134106.
[20]	BISBROUCK N， BERTANI M， ANGELI F， et al. Impact of magnesium on the structure of aluminoborosilicate glasses： a solid-state NMR and Raman spectroscopy study［J］. Journal of the American Ceramic Society， 2021， 104（9）： 4518-4536.
[21]	TRICOT G. The structure of Pyrex^® glass investigated by correlation NMR spectroscopy［J］. Physical Chemistry Chemical Physics， 2016， 18（38）： 26764-26770.
[22]	MARPLE M A T， JESUIT M， HUNG I， et al. Structure of TeO₂ glass： results from 2D ¹²⁵Te NMR spectroscopy［J］. Journal of Non-Crystalline Solids， 2019， 513： 183-190.
[23]	PARK J H. Structure-property relationship of CaO-MgO-SiO₂ slag： quantitative analysis of Raman spectra［J］. Metallurgical and Materials Transactions B， 2013， 44（4）： 938-947.
[24]	FERNANDES H R， KAPOOR S， PATEL Y， et al. Composition-structure-property relationships in Li₂O-Al₂O₃-B₂O₃ glasses［J］. Journal of Non-Crystalline Solids， 2018， 502： 142-151.
[25]	单传丽，柯震坤，郝梦尧，等. B₂O₃和Al₂O₃对硼硅酸盐玻璃网络结构、热学性能和热安全性能的影响［J］. 硅酸盐通报， 2025， 44（7）： 2647-2655.
	SHAN C L， KE Z K， HAO M Y， et al. Effects of B₂O₃ and Al₂O₃ on glass network structure， thermal properties， and thermal safe performance of borosilicate glass［J］. Bulletin of the Chinese Ceramic Society， 2025， 44（7）： 2647-2655 （in Chinese）.
[26]	SHULL C G， WOLLAN E O. X-ray， electron， and neutron diffraction［J］. Science， 1948， 108（2795）： 69-75.
[27]	KUO P H， DU J C. Atomistic understanding of ion exchange strengthening of boroaluminosilicate glasses： insights from molecular dynamics simulations and QSPR analysis［J］. The Journal of Physical Chemistry B， 2022， 126（9）： 2060-2072.
[28]	DU J， CORMACK A N. The medium range structure of sodium silicate glasses： a molecular dynamics simulation［J］. Journal of Non-Crystalline Solids， 2004， 349： 66-79.
[29]	LUSVARDI G， MALAVASI G， MENABUE L， et al. Density of multicomponent silica-based potential bioglasses： quantitative structure-property relationships （QSPR） analysis［J］. Journal of the European Ceramic Society， 2007， 27（2/3）： 499-504.
[30]	CHRISTIE J K， TILOCCA A. Molecular dynamics simulations and structural descriptors of radioisotope glass vectors for in situ radiotherapy［J］. The Journal of Physical Chemistry B， 2012， 116（41）： 12614-12620.
[31]	LUSVARDI G， MALAVASI G， TARSITANO F， et al. Quantitative structure-property relationships of potentially bioactive fluoro phospho-silicate glasses［J］. The Journal of Physical Chemistry B， 2009， 113（30）： 10331-10338.
[32]	XIE W Q， FU Q， YOUNGMAN R E， et al. Revealing structure and property relationships in ZrO₂-containing phosphosilicate bioactive glasses［J］. Journal of the American Ceramic Society， 2025， 108（12）： e70154.
[33]	GONG K， WHITE C E. Predicting CaO-（MgO）-Al₂O₃-SiO₂ glass reactivity in alkaline environments from force field molecular dynamics simulations［J］. Cement and Concrete Research， 2021， 150： 106588.
[34]	刘小青，何　峰，房　玉. 硼硅酸盐玻璃结构与其熔体性质研究［J］. 硅酸盐通报， 2013， 32（5）： 804-807+813.
	LIU X Q， HE F， FANG Y. Structures and melt properties of borosilicate glasses［J］. Bulletin of the Chinese Ceramic Society， 2013， 32（5）： 804-807+813 （in Chinese）.
[35]	DENG L， DU J C. Development of boron oxide potentials for computer simulations of multicomponent oxide glasses［J］. Journal of the American Ceramic Society， 2019， 102（5）： 2482-2505.
[36]	CORMACK A N， DU J， ZEITLER T R. Alkali ion migration mechanisms in silicate glasses probed by molecular dynamics simulations［J］. Physical Chemistry Chemical Physics， 2002， 4（14）： 3193-3197.
[37]	CORMACK A N， DU J， ZEITLER T R. Sodium ion migration mechanisms in silicate glasses probed by molecular dynamics simulations［J］. Journal of Non-Crystalline Solids， 2003， 323（1/2/3）： 147-154.
[38]	DU J C， CORMACK A N. Molecular dynamics simulation of the structure and hydroxylation of silica glass surfaces［J］. Journal of the American Ceramic Society， 2005， 88（9）： 2532-2539.
[39]	LI X， SONG W Y， YANG K， et al. Cooling rate effects in sodium silicate glasses： bridging the gap between molecular dynamics simulations and experiments［J］. The Journal of Chemical Physics， 2017， 147（7）： 074501.
[40]	BHATIA N M， GAIKWAD V L， MANE R V， et al. Quantitative structure-property relationship modeling for the prediction of hydrophilic drug entrapment in liposomes for lung targeted delivery［J］. New Journal of Chemistry， 2018， 42（6）： 4384-4393.
[41]	GAIKWAD V L， BHATIA N M， DESAI S A， et al. Quantitative structure property relationship modeling of excipient properties for prediction of formulation characteristics［J］. Carbohydrate Polymers， 2016， 151： 593-599.
[42]	LU X N， VIENNA J D， DU J C. Glass formulation and composition optimization with property models： a review［J］. Journal of the American Ceramic Society， 2024， 107（3）： 1603-1624.
[43]	YAN J P， ZHANG Y J， WANG F M， et al. Electronegativity-based QSPR analysis for understanding structure-property relationships of glass materials［J］. The Journal of Physical Chemistry B， 2025， 129（20）： 5033-5046.
[44]	LIU P X， LONG W. Current mathematical methods used in QSAR/QSPR studies［J］. International Journal of Molecular Sciences， 2009， 10（5）： 1978-1998.
[45]	LE T， CHANDANA EPA V， BURDEN F R， et al. Quantitative structure-property relationship modeling of diverse materials properties［J］. Chemical Reviews， 2012， 112（5）： 2889-2919.
[46]	马　倩，宋项宁，苑　媛，等. QSPR中变量选择和建模算法的研究进展［J］. 化学研究与应用， 2025， 37（3）： 513-523.
	MA Q， SONG X N， YUAN Y， et al. Research progress of variable selection and modeling algorithm in QSPR［J］. Chemical Research and Application， 2025， 37（3）： 513-523 （in Chinese）.
[47]	HELGUERA A M， COMBES R D， GONZÁLEZ M P， et al. Applications of 2D descriptors in drug design： a dragon tale［J］. Current Topics in Medicinal Chemistry， 2008， 8（18）： 1628-1655.
[48]	KATRITZKY A R， KUANAR M， SLAVOV S， et al. Quantitative correlation of physical and chemical properties with chemical structure： utility for prediction［J］. Chemical Reviews， 2010， 110（10）： 5714-5789.
[49]	SCIOR T， MEDINA-FRANCO J L， DO Q T， et al. How to recognize and workaround pitfalls in QSAR studies： a critical review［J］. Current Medicinal Chemistry， 2009， 16（32）： 4297-4313.
[50]	DU J C， LU X N， GIN S， et al. Predicting the dissolution rate of borosilicate glasses using QSPR analysis based on molecular dynamics simulations［J］. Journal of the American Ceramic Society， 2021， 104（9）： 4445-4458.
[51]	MOUNTJOY G， AL-HASNI B M， STOREY C. Structural organisation in oxide glasses from molecular dynamics modelling［J］. Journal of Non-Crystalline Solids， 2011， 357（14）： 2522-2529.
[52]	RIMSZA J M， DU J C. Structural and mechanical properties of nanoporous silica［J］. Journal of the American Ceramic Society， 2014， 97（3）： 772-781.
[53]	MALAVASI G， PEDONE A， MENZIANI M C. Towards a quantitative rationalization of multicomponent glass properties by means of molecular dynamics simulations［J］. Molecular Simulation， 2006， 32（12/13）： 1045-1055.
[54]	SUN K H. Fundamental condition of glass formation［J］. Journal of the American Ceramic Society， 1947， 30（9）： 277-281.
[55]	ZHAO Z Y， ZHAO M， LANG Y D， et al. Unlocking a positive correlated descriptor of Young’s modulus in silicate glasses using quantitative structure-property relationship analysis［J］. Journal of the American Ceramic Society， 2025， 108（9）： e20691.
[56]	TANTARDINI C， OGANOV A R. Thermochemical electronegativities of the elements［J］. Nature Communications， 2021， 12（1）： 2087.
[57]	OELKERS E H. General kinetic description of multioxide silicate mineral and glass dissolution［J］. Geochimica et Cosmochimica Acta， 2001， 65（21）： 3703-3719.
[58]	AUGHENBAUGH K L， WILLIAMSON T， JUENGER M C G. Critical evaluation of strength prediction methods for alkali-activated fly ash［J］. Materials and Structures， 2015， 48（3）： 607-620.
[59]	KUCHARCZYK S， ZAJAC M， STABLER C， et al. Structure and reactivity of synthetic CaO-Al₂O₃-SiO₂ glasses［J］. Cement and Concrete Research， 2019， 120： 77-91.
[60]	DEEPIKA S， ARATHI P. Computing eccentricity based topological indices of silicate network with applications to QSPR/QSAR analysis［J］. Chemometrics and Intelligent Laboratory Systems， 2025， 263： 105424.
[61]	XIE W Q， LI H， NEUVILLE D R， et al. Effect of Li₂O on the structure and properties of low-boron aluminosilicate fiber glasses from molecular dynamics simulations and quantitative structure-property relationship analysis［J］. Journal of the American Ceramic Society， 2024， 107（12）： 7810-7824.
[62]	胡丽丽，王　欣，张丽艳，等. 从加和法到AI辅助数据驱动的玻璃性质模拟计算［J］. 硅酸盐学报， 2025， 53（10）： 2929-2938.
	HU L L， WANG X， ZHANG L Y， et al. From summation to AI-assisted data-driven methods of glass property calculations［J］. Journal of the Chinese Ceramic Society， 2025， 53（10）： 2929-2938 （in Chinese）.
[63]	FRUGIER P， GIN S， MINET Y， et al. SON68 nuclear glass dissolution kinetics： current state of knowledge and basis of the new GRAAL model［J］. Journal of Nuclear Materials， 2008， 380（1/2/3）： 8-21.
[64]	DENG L， DU J C. Development of effective empirical potentials for molecular dynamics simulations of the structures and properties of boroaluminosilicate glasses［J］. Journal of Non-Crystalline Solids， 2016， 453： 177-194.
[65]	ZHANG Y C， BERTANI M， PEDONE A， et al. Decoding crystallization behavior of aluminoborosilicate glasses： from structural descriptors to quantitative structure-property relationship （QSPR） based predictive models［J］. Acta Materialia， 2024， 268： 119784.
[66]	NORDBERG M E， MOCHEL E L， GARFINKEL H M， et al. Strengthening by ion exchange［J］. Journal of the American Ceramic Society， 1964， 47（5）： 215-219.
[67]	GY R. Ion exchange for glass strengthening［J］. Materials Science and Engineering： B， 2008， 149（2）： 159-165.
[68]	VARSHNEYA A K. Chemical strengthening of glass： lessons learned and yet to be learned［J］. International Journal of Applied Glass Science， 2010， 1（2）： 131-142.
[69]	DACHEUX N， CLAVIER N， ROBISSON A C， et al. Immobilisation of actinides in phosphate matrices［J］. Comptes Rendus Chimie， 2004， 7（12）： 1141-1152.
[70]	DAY D E， WU Z， RAY C S， et al. Chemically durable iron phosphate glass wasteforms［J］. Journal of Non-Crystalline Solids， 1998， 241（1）： 1-12.
[71]	BINGHAM P A， HAND R J， HANNANT O M， et al. Effects of modifier additions on the thermal properties， chemical durability， oxidation state and structure of iron phosphate glasses［J］. Journal of Non-Crystalline Solids， 2009， 355（28/29/30）： 1526-1538.
[72]	KALAHE J， LU X N， RILEY B J， et al. Structure-property relations of sodium iron phosphate nuclear waste glasses： effects of iron redox ratio and glass composition［J］. Journal of the American Ceramic Society， 2025， 108（2）： e20174.
[73]	YUE Y， WANG Y J， CAO Y B， et al. Effect of Al₂O₃ on structure and properties of Al₂O₃-K₂O-P₂O₅ glasses［J］. Optical Materials Express， 2018， 8（2）： 245.
[74]	LINATI L， LUSVARDI G， MALAVASI G， et al. Qualitative and quantitative structure-property relationships analysis of multicomponent potential bioglasses［J］. The Journal of Physical Chemistry B， 2005， 109（11）： 4989-4998.
[75]	PRIVEN A I. Evaluation of the fraction of fourfold-coordinated boron in oxide glasses from their composition［J］. Glass Physics and Chemistry， 2000， 26（5）： 441-454.
[76]	LU X N， DENG L， KUO P H， et al. Effects of boron oxide substitution on the structure and bioactivity of SrO-containing bioactive glasses［J］. Journal of Materials Science， 2017， 52（15）： 8793-8811.
[77]	MIRKHANI S A， GHARAGHEIZI F， ILANI-KASHKOULI P， et al. An accurate model for the prediction of the glass transition temperature of ammonium based ionic liquids： a QSPR approach［J］. Fluid Phase Equilibria， 2012， 324： 50-63.
[78]	CHENG Y Q， MA E. Atomic-level structure and structure-property relationship in metallic glasses［J］. Progress in Materials Science， 2011， 56（4）： 379-473.
[79]	DING J， CHENG Y Q， SHENG H， et al. Universal structural parameter to quantitatively predict metallic glass properties［J］. Nature Communications， 2016， 7： 13733.
[80]	ZHU F， SONG S X， REDDY K M， et al. Spatial heterogeneity as the structure feature for structure-property relationship of metallic glasses［J］. Nature Communications， 2018， 9（1）： 3965.
[81]	WEI D， YANG J， JIANG M Q， et al. Revisiting the structure-property relationship of metallic glasses： common spatial correlation revealed as a hidden rule［J］. Physical Review B， 2019， 99： 014115.
[82]	KANG H， LIU H R， LI Q G， et al. Invariant topological feature of atomic packing in a model metallic glass［J］. Acta Materialia， 2025， 289： 120925.
[83]	ZAKERY A， ELLIOTT S R. Optical properties and applications of chalcogenide glasses： a review［J］. Journal of Non-Crystalline Solids， 2003， 330（1/2/3）： 1-12.
[84]	SHARMA P， SHARMA N， SHARDA S， et al. Recent developments on the optical properties of thin films of chalcogenide glasses［J］. Progress in Solid State Chemistry， 2016， 44（4）： 131-141.
[85]	LORETZ R A， LORETZ T J， RICHARDSON K A. Predictive method to assess chalcogenide glass properties： bonding， density and the impact on glass properties［J］. Optical Materials Express， 2022， 12（5）： 2012.
[86]	MASTELINI S M， CASSAR D R， ALCOBAÇA E， et al. Machine learning unveils composition-property relationships in chalcogenide glasses［J］. Acta Materialia， 2022， 240： 118302.