Resilient 3D printed porous biodegradable polylactic acid coated with bismuth ferrite for piezo enhanced photocatalysis degradation assisted by machine learning

IF 17.1 1区材料科学 Q1 CHEMISTRY, PHYSICAL

Nano Energy Pub Date : 2025-04-14 DOI:10.1016/j.nanoen.2025.111010

Manshu Dhillon , Tushar Moitra , Shivali Dhingra , Kamalakannan Kailasam , Basab Chakraborty , Aviru Kumar Basu

{"title":"Resilient 3D printed porous biodegradable polylactic acid coated with bismuth ferrite for piezo enhanced photocatalysis degradation assisted by machine learning","authors":"Manshu Dhillon , Tushar Moitra , Shivali Dhingra , Kamalakannan Kailasam , Basab Chakraborty , Aviru Kumar Basu","doi":"10.1016/j.nanoen.2025.111010","DOIUrl":null,"url":null,"abstract":"<div><div>This study focuses on the structural modulation and piezo-photocatalytic performance of BiFeO<sub>3</sub> (BFO) using 3D-printed polylactic acid (PLA) substrates for catalysis. It is crucial to develop cost-effective and reusable three-dimensional (3D)-printed substrates with catalyst coatings. Our research aims to investigate the potential of 3D-printed polymer structures as organized surfaces for anchoring the catalyst coatings. This study introduces a simple dip-coating method for uniformly applying BFO coating onto PLA substrates for piezo-photocatalytic purposes. Systematic material characterization studies confirmed the uniform distribution of BFO nanoparticles on the PLA substrate. The catalyst demonstrates exceptional piezo-photocatalytic activity, efficiently degrading cationic and anionic dyes, such as Methylene Blue (MB) and Congo Red (CR). 98.9 % and 74.3 % degradation were achieved for CR and MB, respectively, within 90 min. Regression modeling techniques are used to estimate degradation behavior. The main machine learning models used are Catboost, XGBoost, Random Forest, Light GBM, and Artificial Neural Networks (ANNs). These techniques are dependable algorithms for determining relationships between dependent and independent variables. These models are advantageous when there is some linear or non-linear complicated relation between the target and input variables, especially if the outcome variables are continuous. The regression models used in the prediction for photocatalysis, piezo-catalysis, and piezo-photocatalysis showed R<sup>2</sup> score of 0.93 (Mean Square Error (MSE) is 0.0044), 0.99 (MSE is 0.00006), 0.99 (MSE is 0.000479) respectively and are well-suited for anticipating and aligning the experimental data regarding the percentage removal of CR and MB dyes. The piezo-photocatalytic performance of BFO-coated PLA for CR (98.9 %) and MB (74.3 %) degradation makes them strong contenders for purifying wastewater.</div></div>","PeriodicalId":394,"journal":{"name":"Nano Energy","volume":"140 ","pages":"Article 111010"},"PeriodicalIF":17.1000,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Nano Energy","FirstCategoryId":"88","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2211285525003696","RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, PHYSICAL","Score":null,"Total":0}

引用次数: 0

Abstract

This study focuses on the structural modulation and piezo-photocatalytic performance of BiFeO₃ (BFO) using 3D-printed polylactic acid (PLA) substrates for catalysis. It is crucial to develop cost-effective and reusable three-dimensional (3D)-printed substrates with catalyst coatings. Our research aims to investigate the potential of 3D-printed polymer structures as organized surfaces for anchoring the catalyst coatings. This study introduces a simple dip-coating method for uniformly applying BFO coating onto PLA substrates for piezo-photocatalytic purposes. Systematic material characterization studies confirmed the uniform distribution of BFO nanoparticles on the PLA substrate. The catalyst demonstrates exceptional piezo-photocatalytic activity, efficiently degrading cationic and anionic dyes, such as Methylene Blue (MB) and Congo Red (CR). 98.9 % and 74.3 % degradation were achieved for CR and MB, respectively, within 90 min. Regression modeling techniques are used to estimate degradation behavior. The main machine learning models used are Catboost, XGBoost, Random Forest, Light GBM, and Artificial Neural Networks (ANNs). These techniques are dependable algorithms for determining relationships between dependent and independent variables. These models are advantageous when there is some linear or non-linear complicated relation between the target and input variables, especially if the outcome variables are continuous. The regression models used in the prediction for photocatalysis, piezo-catalysis, and piezo-photocatalysis showed R² score of 0.93 (Mean Square Error (MSE) is 0.0044), 0.99 (MSE is 0.00006), 0.99 (MSE is 0.000479) respectively and are well-suited for anticipating and aligning the experimental data regarding the percentage removal of CR and MB dyes. The piezo-photocatalytic performance of BFO-coated PLA for CR (98.9 %) and MB (74.3 %) degradation makes them strong contenders for purifying wastewater.

Abstract Image

查看原文本刊更多论文

弹性三维打印多孔可生物降解聚乳酸（Polylactic Acid）涂覆铁氧体铋（Bismuth Ferrite），在机器学习辅助下实现压电增强光催化降解

本研究的重点是利用三维打印聚乳酸（PLA）基底催化 BiFeO3（BFO）的结构调制和压电光催化性能。开发具有催化剂涂层的经济高效且可重复使用的三维（3D）打印基底至关重要。我们的研究旨在探讨三维打印聚合物结构作为锚定催化剂涂层的有组织表面的潜力。本研究介绍了一种简单的浸涂方法，可将 BFO 涂层均匀地涂在聚乳酸基底上，用于压电光催化。系统的材料表征研究证实了 BFO 纳米粒子在聚乳酸基底上的均匀分布。该催化剂具有优异的压电光催化活性，可高效降解阳离子和阴离子染料，如亚甲蓝（MB）和刚果红（CR）。在 90 分钟内，CR 和 MB 的降解率分别达到 98.9% 和 74.3%。回归建模技术用于估计降解行为。使用的主要机器学习模型有 Catboost、XGBoost、随机森林、Light GBM 和人工神经网络（ANN）。这些技术是确定因变量和自变量之间关系的可靠算法。当目标变量和输入变量之间存在某种线性或非线性的复杂关系时，尤其是当结果变量是连续的时候，这些模型就会发挥优势。用于光催化、压电催化和压电光催化预测的回归模型的 R2 值分别为 0.93（均方误差为 0.0044）、0.99（均方误差为 0.00006）和 0.99（均方误差为 0.000479），非常适合预测和调整有关去除 CR 和 MB 染料百分比的实验数据。BFO 涂层聚乳酸对 CR（98.9%）和 MB（74.3%）降解的压光催化性能使其成为净化废水的有力竞争者。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Nano Energy CHEMISTRY, PHYSICAL-NANOSCIENCE & NANOTECHNOLOGY

CiteScore

30.30

自引率

7.40%

发文量

1207

审稿时长

23 days

期刊介绍： Nano Energy is a multidisciplinary, rapid-publication forum of original peer-reviewed contributions on the science and engineering of nanomaterials and nanodevices used in all forms of energy harvesting, conversion, storage, utilization and policy. Through its mixture of articles, reviews, communications, research news, and information on key developments, Nano Energy provides a comprehensive coverage of this exciting and dynamic field which joins nanoscience and nanotechnology with energy science. The journal is relevant to all those who are interested in nanomaterials solutions to the energy problem. Nano Energy publishes original experimental and theoretical research on all aspects of energy-related research which utilizes nanomaterials and nanotechnology. Manuscripts of four types are considered: review articles which inform readers of the latest research and advances in energy science; rapid communications which feature exciting research breakthroughs in the field; full-length articles which report comprehensive research developments; and news and opinions which comment on topical issues or express views on the developments in related fields.