The accurate prediction of the brittleness index (BI) of the rocks is crucial for various geoenergy applications. Mineral-based brittleness index (MBI) is one of the simple methods to calculate the BI of the rocks, but it is associated with errors in various lithologies and mineralogies. The objective of this paper is to develop a new model that improves BI prediction. This new model addresses the limitation of unidimensional analysis of the commonly used MBI in diverse lithologies and mineralogies. To achieve this, MBI was combined with geomechanical information. A dataset of 87 data points, encompassing sedimentary, magmatic, and metamorphic rocks from the GEMex project (Acoculco and Los Humeros region), was utilized. The methodology involved: (1) selecting shear modulus (G), bulk density (<InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="13202_2025_2068_Article_IEq1.gif" Format="GIF" Height="12" Rendition="HTML" Resolution="72" Type="Linedraw" Width="18" /> </InlineMediaObject> <EquationSource Format="TEX">\({\rho }_{b}\)</EquationSource> </InlineEquation>), S-wave (<InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="13202_2025_2068_Article_IEq2.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="18" /> </InlineMediaObject> <EquationSource Format="TEX">\({V}_{s}\)</EquationSource> </InlineEquation>) and P-wave (<InlineEquation ID="IEq3"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="13202_2025_2068_Article_IEq3.gif" Format="GIF" Height="19" Rendition="HTML" Resolution="72" Type="Linedraw" Width="19" /> </InlineMediaObject> <EquationSource Format="TEX">\({V}_{p}\)</EquationSource> </InlineEquation>) as key geomechanical parameters; (2) conducting data processing, including cleaning, transformation, integration, visualization, and interpretation, to ensure data quality; and (3) employing linear regression to develop the improved BI model. The model demonstrated strong performance, achieving an average of <InlineEquation ID="IEq4"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="13202_2025_2068_Article_IEq4.gif" Format="GIF" Height="17" Rendition="HTML" Resolution="72" Type="Linedraw" Width="37" /> </InlineMediaObject> <EquationSource Format="TEX">\({\text{R}}^{2}=\)</EquationSource> </InlineEquation> 95%, MAE of 0.018, and RMSE of 0.05. Notably, the testing data group exhibited the best results: <InlineEquation ID="IEq5"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="13202_2025_2068_Article_IEq5.gif" Format="GIF" Height="17" Rendition="HTML" Resolution="72" Type="Linedraw" Width="19" /> </InlineMediaObject> <EquationSource Format="TEX">\({\text{R}}^{2}\)</EquationSource> </InlineEquation>= 97%, MAE = 0.015, and RMSE = 0.02. Furthermore, G was identified as the most influential parameter. This integration of geomechanical data with MBI provides a more robust and accurate approach to BI prediction. This method considers both the mineralogical composition and the mechanical behavior of the rock, leading to a more precise investigation of the rock and representing its key advantage and novelty.

Quantifying rock brittleness: a refined approach to integrating mineralogy and geomechanics

期刊论文
出版模式：

开放获取
发表日期：

2025年9月25日

查看全文

Amirreza Rouhipour,
Shahin Kord,
Aboozar Soleymanzadeh

摘要

The accurate prediction of the brittleness index (BI) of the rocks is crucial for various geoenergy applications. Mineral-based brittleness index (MBI) is one of the simple methods to calculate the BI of the rocks, but it is associated with errors in various lithologies and mineralogies. The objective of this paper is to develop a new model that improves BI prediction. This new model addresses the limitation of unidimensional analysis of the commonly used MBI in diverse lithologies and mineralogies. To achieve this, MBI was combined with geomechanical information. A dataset of 87 data points, encompassing sedimentary, magmatic, and metamorphic rocks from the GEMex project (Acoculco and Los Humeros region), was utilized. The methodology involved: (1) selecting shear modulus (G), bulk density ( \({\rho }_{b}\) ), S-wave ( \({V}_{s}\) ) and P-wave ( \({V}_{p}\) ) as key geomechanical parameters; (2) conducting data processing, including cleaning, transformation, integration, visualization, and interpretation, to ensure data quality; and (3) employing linear regression to develop the improved BI model. The model demonstrated strong performance, achieving an average of \({\text{R}}^{2}=\) 95%, MAE of 0.018, and RMSE of 0.05. Notably, the testing data group exhibited the best results: \({\text{R}}^{2}\) = 97%, MAE = 0.015, and RMSE = 0.02. Furthermore, G was identified as the most influential parameter. This integration of geomechanical data with MBI provides a more robust and accurate approach to BI prediction. This method considers both the mineralogical composition and the mechanical behavior of the rock, leading to a more precise investigation of the rock and representing its key advantage and novelty.