Skip to main navigation Skip to search Skip to main content

Exploratory machine learning analysis of research trends in photocatalytic cementitious materials

  • Shin Deuk Park
  • , Sung Sik Park
  • , Madhan K
  • , Wha Jung Kim*
  • *Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

We investigated research trends in photocatalytic cementitious materials over the past decade by analyzing 130 peer-reviewed papers (systematically selected from an initial corpus of approximately 1400 publications) using computational approaches, including topic modeling, citation network analysis and temporal trend evaluation. Our methodology achieved satisfactory validation metrics, with a coherence score of 0.68, a perplexity of 142.3 and a cross-validation accuracy of 0.71 (±0.03), as confirmed by an expert review. The research field has evolved through four distinct phases: foundational studies (2014–2016), performance enhancement (2017–2019), early applications (2020–2022) and practical implementation (2023–2024). Topic analysis revealed eight major research themes with environmental cleanup dominating the current work (55.4 % of publications), followed by self-cleaning applications (34.6 %) and antimicrobial functions (20.0 %). The frequency analysis identified TiO2 (85 occurrences), performance (81 occurrences) and photocatalysis (76 occurrences) as the most prominent terms. Notable technological progress includes enhanced NOx removal efficiency, improving from 15 % to 25 % in early studies to a peak performance of 70 % by 2024. Visible-light activation has advanced significantly, with graphene-TiO2 nanocomposites achieving efficiencies of 60–75 %. Long-term field studies have demonstrated encouraging durability, with 75–85 % of the initial performance maintained after seven years. Economic viability has also improved, with manufacturing costs decreasing from a 300 % to a 150 % premium over conventional cement production. The Technology Readiness Level (TRL) assessment indicates that the scientific foundations have reached TRL 4–5, while commercial implementation remains at TRL 2–3, identifying a substantial development gap that requires focused attention on standardization and operational optimization. We found the intensification of international collaboration particularly interesting, which increased from approximately 29 % during 2014–2017 to nearly 47 % in recent years, reflecting the growing interdisciplinary integration among the domains of materials science, environmental engineering, and construction.

Original languageEnglish
Article number103068
JournalApplied Materials Today
Volume48
DOIs
StatePublished - Feb 2026
Externally publishedYes

Bibliographical note

Publisher Copyright:
Copyright © 2025. Published by Elsevier Ltd.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 8 - Decent Work and Economic Growth
    SDG 8 Decent Work and Economic Growth
  2. SDG 9 - Industry, Innovation, and Infrastructure
    SDG 9 Industry, Innovation, and Infrastructure

Keywords

  • Air pollution reduction
  • Environmental remediation
  • Machine learning analysis
  • Photocatalytic cement
  • Research trend analysis
  • Technology readiness assessment
  • TiO nanoparticles

ASJC Scopus subject areas

  • General Materials Science

Fingerprint

Dive into the research topics of 'Exploratory machine learning analysis of research trends in photocatalytic cementitious materials'. Together they form a unique fingerprint.

Cite this