Abstract

Research in the source apportionment of trace and major elements for the successful implementation of strategies in environmental management has expanded following growth trends in industrial activities, urban transportation systems, and urbanization in developing countries. This study adopted a novel approach to identify and quantify the contribution of geogenic sources along with other potential sources. Enrichment factor (EF), principal component factor analysis (PCFA), positive matrix factorization (PMF), UNMIX receptor model, and the SPECIATE database were used for source apportionment to determine the role of twenty-four trace and major elements as pollutants in the road dust of Najafabad (Iran). The statistical analyses of the geochemical data were conducted via GraphPad Prism, version 9.0, and SPSS, version 22. According to the results, the concentration levels of all the trace and major elements, except Cd and K, were higher than the background value. The mean EF value for Ce was the highest (10.93), followed by Ba (6.09), Al (5.30), Mg (5.71), Sr (3.63), and Y (3.17), while the other elements were of minimum enrichment in the dust samples (EF < 2). The PCFA results resolved four sources with their respective contributions, namely geogenic (57%), industrial (24%), traffic (12%), and unidentified sources (6%). The PMF and UNMIX models revealed three sources with their respective contributions based on marker species: geogenic (80 and 94%), traffic (13 and 5%), and industrial (1 and 7%). The spatial variation analysis of source contribution by UNMIX and PMF revealed that the contribution of industrial and traffic sources corresponded to the substantial activity of factories and transport by light and heavy vehicles in the study area. Overall, the EF results were in agreement with 43% of the PCFA results, 33% of the UNMIX results, and 23% of the PMF results. The study concluded that the PMF model gave acceptable results, but those of EF, PCFA, and UNMIX were unsatisfactory.

Introduction

This study focuses on the source apportionment of trace and major elements in road dust in Najafabad, Iran—an industrial area with increasing environmental challenges due to traffic and outdated vehicle technologies. While past research has explored pollution in urban road dust, few have quantitatively assessed the geogenic (natural) versus anthropogenic (human-made) contributions. This work applies multivariate statistical tools—Enrichment Factor (EF), Principal Component Factor Analysis (PCFA), Positive Matrix Factorization (PMF), and the UNMIX model—to identify and quantify pollutant sources, aiming to support future environmental policy and management.

Materials and Methods

Road dust samples were collected from 30 locations across Najafabad’s roads and intersections. After drying, sieving, and acid digestion (HF, HCl, HNO₃, HClO₄), elemental concentrations (Al, Fe, Pb, Zn, etc.) were analyzed using ICP-OES. EF was used to assess enrichment relative to geogenic background. Multivariate techniques like PCFA, PMF (EPA PMF 5.0), and UNMIX (EPA UNMIX 6.0) were applied to determine sources. The SPECIATE database helped interpret source profiles. Data processing included KMO/Bartlett tests, scaled residuals, Q-value optimization, and Spearman correlation for statistical relationships.

Results and Discussion

Concentration Patterns: Al and Fe were most abundant; Cd and Yb were least. Compared to Iranian cities, some elements (e.g., Cu, Pb, Zn) showed elevated levels, suggesting industrial and vehicular inputs.

EF Analysis: High enrichment was found for Ce, Ba, Al, Mg, Sr, and Y, indicating anthropogenic sources. Many elements had EF < 2, implying geogenic origin.

PCFA & Clustering: Four principal components were identified:

PC1: Geogenic (Al, K, Li, etc.).

PC2: Mixed geogenic/anthropogenic (Ba, Fe, Ni, etc.).

PC3: Industrial/traffic (Cd, Cu, Pb, Zn).

PC4: Unidentified (Ce, La).

PMF & UNMIX Modeling: Both models identified three sources:

PMF: Geogenic (80%), traffic (13%), industrial (7%).

UNMIX: Geogenic (94%), traffic (5%), industrial (1%).
The PMF model showed higher consistency and was deemed more reliable.

SPECIATE Matching: Najafabad’s dust profiles had close overlaps with US EPA profiles, supporting source identifications.

Conclusion

Geogenic sources were the dominant contributors to road dust-bound elements in Najafabad, but industrial and traffic sources also played significant roles. Among the models used, PMF provided the most reliable results, while EF, PCFA, and UNMIX had partial agreement. These findings underscore the utility of multivariate receptor models in urban environmental management and can guide pollution mitigation strategies in similar regions.