2.1 Sampling and analysis

About 32 dust samples were collected from two engineering laboratories, Lab 1 and Lab 2, located within the academic building complex in UiTM Cawangan Pahang (Fig. 1). Lab 1 is a structural lab facing the main campus road. Lab 2, the soil mechanics lab, is on the opposite side. Both laboratories have a comparable occupancy capacity, allowing for the presence of 30 people simultaneously, with an estimated area of 200 m² each. The ventilation systems utilised in the building are of the mechanical supply type, with fans installed in Lab 1, and of the exhaust type, with air conditioning, in Lab 2. These sampling areas were chosen considering students’ use patterns and the presence of ventilation systems that effectively confine pollutants within the building (^{Zhong et al., 2014}; ^{Sulaiman and Suratmin, 2020}).

Fig. 1 Sampling locations in the university laboratories. 

For all sampling locations, samples were taken for four weeks during academic activities between 09.00 and 17.00 LT. Each sampling location is equipped with a combination of exhaust ventilation systems, such as air-conditioning, and mechanical supply, such as fans. A soft brush was used to collect the indoor dust from various sources, such as fume hoods, air conditioners, benches, fans, and windows (^{Sulaiman et al., 2017}).

The dust samples were subjected to a drying process in a laboratory oven set at a temperature of 65 ºC overnight. The dust, weighing approximately 1 g, was subsequently measured and underwent digestion using a mixture of nitric acid (HNO₃ 75% GR, Merck, Germany) and hydrochloric acid (HCl 70% GR, Merck, Germany) at a 4:1 ratio. This process was carried out on a hot plate set at a temperature of 50 ºC for a duration of 1 h under a fume hood. After cooling, the solution was further filtered using filter paper and transferred into a 100 ml volumetric flask. Deionized water was then added to the filtrate until it reached the 100 ml calibration level. Concentrations of Al, Cu, Fe, Pb, and Zn in the samples were assessed using inductively coupled plasma optical emission spectrometry (ICP-OES) (Agilent 5100).

For reference, a sample of topsoil was obtained near the sampling location, about 2 m away from the building designated as the sampling point. The collection was made at a depth of 0.01 m. This study used the neighboring topsoil metal concentrations as background ratio (^{Sulaiman et al., 2017}). The soil samples were dried in a furnace set at a temperature of 450 ºC for 3 h. The reference sample underwent digestion using the same steps as the dust samples.

The accuracy of the analytical process was verified using an analysis of a spiked sample containing a known quantity of metal standard for recovery testing, with the metal recovery rates ranging between 85 and 115%. All analysis devices underwent precleaning treatment in which they were submerged in nitric acid overnight, followed by rinsing with deionized water. An individually new soft paintbrush was used for each sampling event to prevent cross-contamination. Powder-free gloves were worn, especially when dealing with interior dust.

2.2 Enrichment factor (EF)

The enrichment factor (EF) calculation has been applied to ascertain the source of metals in samples, relying on the prevailing metal element found in the Earth’s crust. ^{Nazir et al. (2011)} stated that an EF larger than 10 (EF > 10) signifies that the element is of anthropogenic origin. In contrast, an EF value less than 10 (EF < 10) indicates that the element is of natural origin. The EF was computed using Eq. (1):

in which (Cx/Cref) _dust is the concentration ratio calculated from the concentrations of element x and Fe measured in indoor dust, and (Cx/Cref) _{surface soil} is the concentration ratio in the surface soil (^{Latif et al., 2014}). Due to its dominance in the composition of the Earth’s crust, Fe is widely used as a reference element.

2.3 Statistical analysis

Prior to the statistical analysis, the datasets underwent a verification procedure. Skewness, kurtosis, and Shapiro-Wilk test were employed to assess normality. A t-test was performed to assess potential changes in mean metal levels in the indoor dust throughout the sampling sites. Correlations between mean heavy metal concentrations were analyzed using a Pearson correlation, considering the data had a normal distribution. Cluster analysis (CA) was employed to categorize the metals derived from various sources, while a dendrogram was utilized to evaluate similarities among the heavy metal concentrations in indoor dust. Statistical evaluations were carried out using IBM SPSS v. 20.

2.4 Health risk estimation

The average daily dosage (ADD) of pollutants and possible risks from indoor lab dust were assessed. According to the United States Environmental Protection Agency (USEPA), Cu, Fe, Pb, and Zn may represent non-carcinogenic concerns. Eqs. (2)-(4) were used to compute the ADD values of pollutants. The hazard quotient (HQ) was calculated to estimate the non-carcinogenic risk (Eq. [5]). Eq. (6) was used to estimate the total non-cancer risk or hazard index (HI), encompassing the summation of HQ values (^{USEPA, 2011}). High hazards are likely when the HQ and HI values surpass a value of 1 and vice versa.

The possibility of cancer risk (CR) from indoor dust was assessed using Eqs. (7)-(9). Lifetime cancer risk (LCR) was calculated by adding the CR from three exposure routes (Eq. [10]). Only Pb was considered to be carcinogenic. The parameter values adopted to calculate health risk are shown in Table I.

3.1 Indoor dust metal composition

Table II presents the metal concentrations of Al, Fe, Cu, Pb, and Zn measured in samples of indoor dust and surface soil collected from the sampling locations. Fe had the highest concentration relative to other metals in indoor dust samples. For Lab 1, the average Fe concentration was 79.85 ± 33.35 mg kg^-1, followed by Al (65.60 ± 21.90 mg kg^-1), Zn (5.05 ± 2.23 mg kg^-1), Cu (1.53 ± 0.62), and Pb (0.68 ± 0.18 mg kg^-1). Lab 2 showed the same trend, with average values for Fe of 127.63 ± 50.70 mg kg^-1, followed by Al (70.40 ± 21.20 mg kg^-1), Zn (4.35 ± 1.45 mg kg^-1), Cu (0.93 ± 0.24 mg kg^-1), and Pb (0.81 ± 0.24 mg kg^-1). Compared to nearby surface soils, indoor dust contained more metals (Table II). The results show that the interior environment exhibits greater metal pollution. The higher coefficient of variation (CV) levels specify that the metal likely is of an anthropogenic source. Metals primarily influenced by natural sources exhibit a low coefficient of variation (CV), while metals affected by man-made sources display a relatively high CV (^{Guo et al., 2012}).

There were statistically significant variations (p < 0.05) in metal concentrations found in indoor dust across different sampling locations. Fe and Al concentrations in indoor dust from Lab 2 were higher than those in Lab 1. Lab 2 also showed a slightly higher Pb concentration than Lab 1. The variations in Fe and Al concentration in indoor dust collected from these two laboratories may be attributed to the introduction of soil dust into the interior environment through student activities and the nature of the laboratory. Lab 2, as the soil mechanics lab, focuses on soil as the primary material for the practical laboratory exercises.

Table III compares selected metal concentrations in this study’s indoor dust samples to those in previous studies’ indoor dust samples. The metal concentrations in laboratories’ indoor dust were comparatively lower than those reported by ^{Latif et al. (2011)}, ^{Sulaiman et al. (2017)}, and ^{Fan et al. (2021)}. This could be related to the nature of the laboratories since previous studies have been documented from biological or chemical-based labs. However, the laboratories’ indoor dust metal concentrations in this study were higher than those in Nigeria (^{Ajayi et al., 2023}). This study found lower concentrations of indoor metals compared to indoor dust in urban homes (^{Kelepertzis et al., 2021}; ^{Asvad et al., 2023}), except for a study by ^{Harb et al. (2015)}.

3.2 Enrichment and correlation between metals

Both laboratories had a similar trend of EFs, wherein Zn had the highest EF value, followed by Al, Cu, and Pb, respectively (Fig. 2). Indoor dust indicated a higher enrichment of Zn compared to other metal elements. Based on the EF value, this study suggests anthropogenic sources of Zn (EF > 10). Lab 1 had a higher EF value of Zn compared to Lab 2. ^{Han et al. (2011)} reported that the presence of Zn in road dust has been identified as a significant component, primarily due to its release from tires, vehicle brakes, and motor oil. Since Lab 1 is located near a road and has a mechanical supply type (i.e., a fan), wind from the outside may have contributed to the Zn enrichment (^{Latif et al., 2014}).

Fig. 2 Estimated enrichment factor (EF) value of metal concentrations in indoor laboratory dust. 

The Pearson’s correlation between indoor dust metals is shown in Table IV. Fe showed strong positive correlations with Pb (r = 0.870). The Cu concentration had a moderate relationship (r = 0.432) with Zn. To further confirm the correlation between metals, the data went through a cluster analysis, wherein Euclidean distances were computed to determine similarities. Figure 3 depicts the cluster analysis outcomes performed on the metals found in indoor dust. The study identified three distinct clusters: C1, consisting of Cu and Zn; C2, consisting of Pb and Fel, and C3 consisting solely of Al. Metals in C1 and C2 may be influenced by anthropogenic sources, while those in C3 could be considered to have natural origins. Vehicle emissions are a significant source of Cu, Zn, and Pb (^{Han et al., 2011}; ^{Hassan, 2012}). Another possible anthropogenic source of Zn and Pb could be linked to the use of paint containing Zn and Pb compounds (^{McAlister et al., 2008}). Furthermore, the presence of Zn in the composition of indoor dust can be attributed to the process of corrosion experienced by objects coated with Zn inside laboratory settings (^{Zhong et al., 2014}).

Fig. 3 Dendrogram of the indoor lab dust cluster analysis. 

3.3 Potential health risk

HQ and HI values were calculated to evaluate non-carcinogenic health hazards associated with the exposure to individual metals via oral, dermal contact, and inhalation, as well as the combined exposure to metals (Cu, Fe, Pb, and Zn). In contrast, the possible health risk of Al was not considered due to the absence of input from the World Health Organization (WHO) or the USEPA. For Fe and Cu, only the oral pathway was considered, as no information on dermal or inhalation routes has been published by USEPA or WHO. Due to the greater vulnerability of children compared to adults towards the impacts of metals, differences in the duration of exposure and body weight compared to adults (^{Guney et al., 2010}), an independent assessment of risk was conducted for this population subgroup.

Table V lists the potential non-carcinogenic risks associated with metal composition in indoor dust. Generally, the HQ values of indoor dust exposure were below the unity value (HQ < 1), indicating low non-cancer risks to students and staff. The HQ for oral pathways suggests higher values than HQs for dermal contact and inhalation. Children have higher non-carcinogenic risks than adults, except for Fe. In terms of the hazard index (HI), children are more susceptible to non-cancer risks than adults (Fig. 4). The HI shows increment values of Pb < Zn < Fe < Cu for children; however, for adults, it shows increment values of Pb < Fe < Cu < Zn.

Fig. 4 Hazard index (HI) of metal exposure from indoor dust. 

The findings of this study confirmed that the oral route is the most hazardous exposure pathway for metals and all population subgroups, consistent with prior research (^{Chabukdhara and Nema, 2013}; ^{Mohamad et al., 2016}; ^{Wang et al., 2021}; ^{Asvad et al., 2023}). Indoor dust samples recorded considerably lower HQ values for dermal contact and inhalation, indicating that these pollutants are less harmful when inhaled. The relative contributions of HQingest, HQdermal, and HQinhalation to children’s overall health index (HI) were 99.2, 0.73, and 0.02%, respectively. For adults, the related contributions were 97.9, 2.04, and 0.05, respectively.

The order of cancer risk (CR) values was CRingest > CRdermal > CRinhale for both children and adults (Table VI). The risk associated with all exposure routes was shown to be greater in children than in adults (Fig. 5), similar to the results reported in Iran by ^{Asvad et al. (2023)}. Despite the lifetime cancer risk (LCR) values in children being found to be two times greater than in adults, the values fell within the tolerated range (10^-6 < LCR < 10^-4).

Fig. 5 Lifetime cancer risk (LCR) of metal exposure from indoor dust. 

This study found that the health risk parameters (HI and LCR) remained within acceptable ranges. However, it is crucial to maintain regular monitoring of the concentrations of hazardous elements, as prior research has indicated considerable health hazards from indoor dust exposure in universities (^{Latif et al., 2011}; ^{Zhong et al., 2014}; ^{Sulaiman et al., 2017}; ^{Fan et al., 2021}). Considering the combination of human-induced and naturally occurring emissions, it is imperative for universities and risk managers to focus not only on mitigating levels of particulate matter (PM) but also on reducing the concentration of hazardous elements and human exposure to indoor dust.