Vietnam is a developing country with an expanding network of roads and highways to support the growing economy. However, the road system was not well maintained, which makes dust pollution a serious issue in Vietnam, with the level of dust density in some cities several times higher than the average for developed countries (Tuyen et al. 2014). It has been reported that the contribution of roadside dust to the total suspended particles (TSP) in urban air was up to 74% (Hien et al. 1999).

Roadside dust is considered a main reservoir for a mix of pollutants, including metals from various activities occurring on and surrounding the road. Exposure to roadside dust could cause hazardous effects to human health through inhalation, ingestion, and dermal contact absorption as the dust carries such pollutants to the human body. In Vietnam, roadside dust is a more important source of exposure than in many other countries, because people usually live along the road, especially busy roads to open shops, which are highly polluted with dust.

Recently, there have been numerous studies on the level of metals (including heavy metals) in roadside dust/sediment (Liu et al. 2014; Mathur et al. 2016; Nazzal et al. 2013), particularly in China (Wei and Yang 2010). Some studies detected high levels of heavy metals compared with the background concentration (Han et al. 2014; Yu et al. 2016a). Other studies determined that some heavy metals, such as Pb, in roadside dust could represent a potential health risk to people (Dehghani et al. 2017; Zheng et al. 2010). It is important to conduct metal monitoring campaigns in different regions to better understanding the local risk of metals in roadside dust (Apeagyei et al. 2011).

We previously reported the first study in Vietnam on the levels of metal contamination in street dust in Hanoi, Vietnam, where the mean concentration of Pb in urban street dust was relatively low compared with other cities around the world, suggesting a lower risk to human health (Phi et al. 2017). However, to our knowledge, no study was performed in Vietnam on roadside dust along the highways that connect important industrial areas where contamination of heavy metals in soil samples surrounding the industrial sites has been reported (Bui et al. 2016; Martinez et al. 2013). Additionally, there was no study in Vietnam to assess the risk of exposure to metals via roadside dust; many people live next to roads and are highly likely to be exposed to roadside dust. Therefore, it is important to conduct a comprehensive risk assessment for exposure to metals in roadside dust in Vietnam.

Portable X-ray fluorescence (pXRF) spectrometry has progressed considerably in the past decades to provide rapid, cost-effective, nondestructive alternatives to the traditional analytical method for analysis of metals in soils and sediments. Previously, data from pXRF were typically validated by analyzing a subsample with traditional methods, such as AAS or ICP-MS, ICP-OES. Hu et al. (2014) and Rouillon and Taylor (2016) have demonstrated that with careful sample preparation, pXRF could produce data of comparable quality to the traditional methods in a cost-effective way. The method also was validated and used in our previous study (Phi et al. 2017).

In this study, we again utilized the pXRF method to: (1) measure the concentrations of several metals, including Pb, Zn, Cr, Cu, Fe, Mn, Ca, and K, in roadside dust along two major highways, namely Highway No. 5 and Highway No. 18, in northern Vietnam with different land use characteristics and (2) investigate the health risk of heavy metals in roadside dust in those areas through the estimation of inhalation and ingestion of dust for humans.

Materials and Methods

Study Area and Sampling

Highways No. 5 and No. 18 are two major highways in the northeast provinces of Vietnam. Highway No. 5 is the most important highway in northern Vietnam and connects the capital Hanoi with the port city of Hai Phong, passing through two the provinces of Hung Yen and Hai Duong. Highway No. 18 connects Hanoi with three northeast provinces of Bac Ninh, Hai Duong, and Quang Ninh. In this study, we focused on the sections of Highway No. 5 located in Hai Duong province and Highway No. 18 located in Quang Ninh province, because they go through different potential sources of metals (Fig. 1).

Fig. 1
figure 1

Map of the study area with sampling sites

Full size image

A total 55 roadside dust samples were collected from a total of 120 km distance passing through areas of different land use, which included commerce, industrials, power plants, mining, and cement plant (Table S1). In each land use area, four to six roadside dust samples were collected carefully in dry weather condition (5 and 6 days after rainfall). The dust sampling was performed by sweeping an area of 1 m2 (4 × 0.25 m2) using a plastic brush and plastic scoop (Fig. S1). The amount of dust collected in each site was from 10 to 190 g with the average mass presented in Table S1. To avoid resuspension of the finest particles during sampling, the sweeping was slow, and the dust sample was transferred directly into a sealed plastic bag.

Sample Analysis

Collected soil samples were air-dried and passed through a 2-mm sieve to remove any husky materials, such as litter, stones, and other debris, as recommended by Rouillon and Taylor (2016). All samples were then sieved into three particle-size fractions (< 75, 75–180, > 180 µm) using standard mesh nylon sieve following the protocol proposed by Duong and Lee (2011).

All fractions of the collected samples were analysed for metals using a Niton XL3t portable X-ray fluorescence spectrometry device (XRF device; Thermo Scientific, Tewksbury, MA) using the benchtop setup in single measurement of 90 s. The XRF method recently has been applied in several studies to measure elemental concentrations in soil and dust samples (Apeagyei et al. 2011; Peinado et al. 2010; Qiao et al. 2014), because it was assessed as acceptable per U.S. Environmental Protection Agency (EPA) Method 6200 (Hu et al. 2014) with appropriate preparation and quality control.

During the analysis, calibration standards from the manufacturer were measured every 20 analyses. Certified standards of heavy metal-containing sediment obtained from the National Research Council, Canada, also were used as quality control. A set of 21 samples was analyzed for heavy metals (Pb and Zn) by the ICP-MS method in an accredited laboratory to validate the XRF method. The limits of detection (LODs) for K, Ca, Mn, Fe, Zn, Pb, Cu, and Cr were 100, 500, 100, 350, 25, 20, 20, and 40 mg/kg, respectively. The accuracy and precision of the XRF device was evaluated by measuring standards and a certified reference material (PACS-2) as well as by comparison with data obtained from ICP-MS measurements. The XRF results have the relative standard deviation (RSD) within ± 20%, which is acceptable according to U.S. EPA Method 6200 (Hu et al. 2014).

Contamination Assessment Indices

To assess the level of contamination in samples that are taken close to different land uses and traffic density, we used two common indices, the enrichment factor (EF) and geo-accumulation index (I geo), which have been used in many previous studies.

The EF is determined by the following equation as suggested by Duong and Lee (2011):

$${\text{EF}} = \frac{{\left( {C_{x} /C_{\text{ref}} } \right)_{\text{sample}} }}{{\left( {C_{x} /C_{\text{ref}} } \right)_{\text{background}} }}$$
(1)

where (C x )sample is the concentration of the element in the street dust sample, (C x )background is the concentration of the same element in the background, and C ref is the concentration of the reference element, which is naturally present in the same sample and background. In this study, background samples were collected from an area away from industries with low traffic and population density.

The geo-accumulation index (I geo) is calculated using the following equation as suggested by Nazzal et al. (2013):

$$I_{\text{geo}} = \log_{2} \left( {\frac{{C_{\text{sample}} }}{{1.5 \; \times \; C_{\text{background}} }}} \right)$$
(2)

where C sample is the metal concentration in the samples and C background is the background value of this metal, and the factor of 1.5 is for correction of potential background variation. The geo-accumulation index consists of seven grades (0–6) indicating various degrees of enrichment compared to the background value, which range from unpolluted to extremely polluted grades (Nazzal et al. 2013).

Exposure Risk Assessment

There are three main exposure routes to metals in roadside dust for human, which include inhalation, ingestion, and dermal contact (Wei and Yang 2010), and exposure is known to be associated with the concentration of metals, the frequency and duration of exposure, as well as the different ages of the exposed. In this study, the exposure estimation for adults and children was based on the method provided by the U.S. EPA and had been applied by Yu et al. (2016b). The calculation of the average daily doses (ADD) of metals via ingestion (ADDing), inhalation (ADDinh), and dermal contact (ADDderm) are:

$${\text{ADD}}_{{{\text{ing}},m}} = \frac{{C_{m} \times {\text{IngR }} \times {\text{EF }} \times {\text{ED}}}}{{{\text{BW }} \times {\text{AT}}_{\text{noncan}} }} \times 10^{ - 6}$$
(3)
$${\text{ADD}}_{{{\text{inh}},m}} = \frac{{C_{m} \times {\text{InhR }} \times {\text{EF }} \times {\text{ED}}}}{{{\text{PEF }} \times {\text{BW }} \times {\text{AT}}_{\text{noncan}} }}$$
(4)
$${\text{ADD}}_{{{\text{derm}},m}} = \frac{{C_{m} \times {\text{SA }} \times {\text{SL }} \times {\text{ABS }} \times {\text{EF }} \times {\text{ED}}}}{{{\text{BW }} \times {\text{AT}}_{\text{noncan}} }} \times 10^{ - 6}$$
(5)

where m is the metal of interest (Cr, Cu, Pb, Zn, Mn); ADDing,m , ADDinh,m , and ADDderm,m are the average daily doses of metal m via ingestion, inhalation, and dermal contact, respectively. C m is the concentration of m in dust samples (mg/kg). IngR and InhR are the ingestion and inhalation rates, respectively. EF is the exposure frequency. ED is the exposure duration. BW is body weight. ATnoncan is the average time for noncarcinogens. PEF is the particle emission factor. SA is the exposed skin area. SL is the skin adherence factor. ABS is the dermal absorption factor. The units and parameter values are shown in Table S2.

For Cr, which is carcinogenic, only the carcinogenic risk from inhalation exposure was considered in this study. The lifetime average daily dose (LADD) for Cr inhalation exposure was used to assess the risk of cancer according to the following equation (USEPA 1996, 2001):

$${\text{LADD}}_{{{\text{inh}},m}} = \frac{{C_{m} }}{{{\text{PEF }} \times {\text{AT}}_{\text{can}} }} \times \left( {\frac{{{\text{InhR}}_{\text{child}} \times {\text{EF}}_{\text{child}} \times {\text{ED}}_{\text{child}} }}{{{\text{BW}}_{\text{child}} }} +\, \frac{{{\text{InhR}}_{\text{adult}} \times {\text{EF}}_{\text{adult}} \times {\text{ED}}_{\text{adult}} }}{{{\text{BW}}_{\text{adult}} }}} \right)$$
(6)

where ATcan is the average time for carcinogens; InhRchild and InhRadult are the inhalation rates for children and adults, respectively. EFchild and EFadult are the exposure frequencies for children and adults, respectively. EDchild and EDadult are the exposure durations for children and adults, respectively. BWchild and BWadult are the average body weights for children and adults, respectively. The units and parameter values used for Eq. (4) also are defined in Table S2.

The average concentration of a metal is used as C m (mg/kg). Values of other parameters are set according to Yu et al. (2016b), because the scenarios are considered similar between the two studies (Table S2).

Risk Characterization

Based on the results of the above calculation, the noncarcinogenic risks of exposure to the metal m in roadside dust were evaluated as follows:

$${\text{HQ}}_{p,m} = \frac{{{\text{ADD}}_{p,m} }}{{{\text{RfD}}_{p,m} }} ;$$
(7)
$${\text{HI}}_{m} = \mathop \sum \limits_{p} {\text{HQ}}_{p,m}$$
(8)

the hazard quotient (HQ p,m ) represents the noncarcinogenic risk for metal m through exposure pathway p (direct ingestion, inhalation, and dermal contact). The reference dose (RfD) are taken from the United States Department of Energy’s RAIS compilation (U.S. Department of Energy 2004), BMRIEP (2009), and MEPC (2013) (Table S2). The hazard index (HI m ) represents the noncarcinogenic risk for metal m through the three exposure pathways. If the HI m (noncancer risk) is < 1, it is believed that there was no significant noncarcinogenic risk; however, if the HI m is > 1, there is a chance that noncarcinogenic effects might occur (USEPA 2001).

For the carcinogenic risk, only Cr was included in the calculation, because it is the only metal among the five (Pb, Zn, Cu, Cr, Mn) that exhibits sufficient evidence as a carcinogen in humans, mainly through inhalation exposure. Therefore, the risk is calculated as:

$${\text{CR}}_{{{\text{inh}},m}} = {\text{LADD}}_{{{\text{inh}},m}} \times {\text{SF}}_{{{\text{inh}},m}} ;$$
(9)

CRinh,m represents the cancer risk from Cr in roadside dust through inhalation. SFinh,m is the cancer slope factor through inhalation.

Statistical Analysis

All statistical analyses were performed through R version 3.3.2 for Windows. Samples with concentrations of metals lower than LOD were included as LOD/sqrt(2) in all calculations (Hornung and Reed 1990). The Mann–Whitney U test was used to examine the differences between groups. Results were considered statistically significant at p < 0.05.

Results and Discussion

Heavy Metals Concentrations in Roadside Dust

The average concentrations of metals measured in roadside dust along the two highways are presented in Table 1. The results show that metal concentrations in roadside dust were usually higher than those in the background soil except for K, which is the background reference element as indicated previously (Phi et al. 2017). The average concentrations increase in the order of Pb < Cu < Cr < Zn < Mn < K < Fe < Ca. The order of metal concentrations in the samples is similar to the order in the background as shown in Table 1.

Table 1 Average concentrations of metals (mg kg−1) measured in roadside dust samples and soil samples in different studies
Full size table

The results in Table 1 also indicate that concentrations of studied metals (except K) in fine particle size fractions were higher than in coarse particle size fraction, which is consistent with data previously reported (Acosta et al. 2011; Duong and Lee 2011; Kabadayi and Cesur 2010). The similarity of K concentrations in all particle size fractions again indicated that this element is in the background and could be used as a reference element to calculate the EF (Phi et al. 2017).

The average concentrations of Pb, Zn, Cu, and Cr in roadside dust samples of this study are below the standard limits for heavy metals in agricultural soils in Vietnam (Table 1). However, concentrations of some individual samples exceed these national limits (Fig. S2a, b). For example, only one individual dust sample (< 75 µm) from site S4 has a concentration of Pb of 86.33 mg/kg, which is higher than the Vietnam standard of 70 mg/kg. Meanwhile 47 individual samples (of the total 155 samples analyzed) from S1, S2, S3, S4, S5, S6, and S9 have Zn concentrations higher than the standard of 200 mg/kg with the highest concentration of 1167.7 mg/kg from site S9. There are also 25 samples with concentration of Cu greater than the standard limit of 100 mg/kg with the highest Cu concentration of 626.7 mg/kg from site S6. Concentrations of Cr in some samples, mainly from sites S4 and S5, are higher than the standard of 150 mg/kg with the highest Cr concentration measured at 841.9 mg/kg from site S5.

Pearson correlation coefficients among metal concentrations in this study show significant correlations among Pb, Zn, Cu, Fe, Mn, and Cr in the fine particle fraction (< 75 µm), which become weaker in the medium particle fraction (75–180 µm) and almost lost in the coarse particle fraction (> 180 µm; Table 2). The correlation among this group of metals (Pb, Zn, Cu, Fe, Mn, Cr) indicate that the main source for them is likely traffic-related one (particles worn off brake-pads and tires) as suggested by other authors (Zafra et al. 2011; Wawer et al. 2015). Two background metals (Ca, K) correlate with each other, again confirming their background reference status.

Table 2 Pearson’s correlation matrix for metal concentrations in roadside dust samples
Full size table

Despite the exceedance of standards in some individual samples, the average concentrations of metals reported in this study are comparable or lower than data reported for roadside dust from highways in other studies (Faiz et al. 2009; Nazzal et al. 2013; Pagotto et al. 2001; Suryawanshi et al. 2016). The concentrations of roadside dust from highways reported in this study also are similar to those from urban street dust of Hanoi reported by Phi et al. (2017). This suggests that the level of metal contamination in roadside dust in Vietnam is not as severe as in some other countries in the world, probably because the fleet of motor vehicles in Vietnam is still small and developed more recently compared to other countries. A similar observation is reported by Phi et al. (2017).

Levels of Metal Contamination in Roadside Dust

Enrichment Factor

Because K was determined in previous sections as the reference element, it was used to calculate the EF values for other metals of interest (Pb, Cu, Cr, Zn, Mn). The EF values for dust samples from each site in all particle size fractions are presented in Table S4. Although some samples have a significant enrichment factor (EF > 5), most samples are moderately enriched (EF = 2–5) or minimally enriched (EF < 2) according to the categorization suggested by Mathur et al. (2016). The highest enrichment occurred in samples of the smallest particle sizes (< 75 µm), whereas the average enrichment factors of samples from the two other fractions are comparable (Table S4).

This observation is similar to several previous studies (Duong and Lee 2011; Phi et al. 2017), which might be due to the fact that roadside dust with smaller particle sizes has more available surface area per unit of mass for deposition or coating of the metals as well as the particles of smaller sizes may be directly emitted by traffic and industrial activities.

Geo-Accumulation Index (I geo)

Igeo values estimated for samples from all the sites are presented in Table S5. In general, most of the I geo values are < 1, indicating a low level of contamination in our roadside dust samples compared with the average metal levels in different soils of Vietnam (Table 1).

Impact of Different Land Uses and Traffic Density

In this study, we attempted to classify the sampling sites into five categories of land uses (commerce, industry, power plant, cement plant, and mining) based on the majority of land occupants in the surrounding areas. Figure 2 shows that industry seems to contribute more to the contamination of heavy metals in roadside dust, followed by commerce. Other land uses do not enrich the dust samples significantly (i.e., EF > 5). It should be noted that the nearby presence of power plants, cement plants, or open mining sites do not significantly contaminate the roadside dust, which is similar to the findings of George et al. (2015) (for coal-fired power plant), Masto et al. (2017), and Martinez et al. (2013) (for coal mining area).

Fig. 2
figure 2

Enrichment factor for selected metals in different land use areas

Full size image

As discussed earlier, traffic could be a major source of metals, such as Zn, Cu, and Cr, through particulates emitted from wear and tear of tires and brake linings. Our comparison indicated that the concentrations of metals in roadside samples from Highway No. 5 are significantly higher than those from Highway No. 18 (t test, p < 0.05; Fig. 3). The high traffic density of Highway No. 5, which is > 10 times higher than Highway No. 18, is probably partly responsible for the higher levels of metals in roadside dust from Highway No. 5.

Fig. 3
figure 3

Enrichment factor for selected metals in two different highways

Full size image

Exposure Risk Assessment

Noncarcinogenic Risks

The average daily doses of five heavy metals (Pb, Zn, Cu, Mn, Cr) were calculated for all exposure pathways and were compared with the corresponding RfDs to yield their hazard quotients (HQ) as presented in Table 3. In our risk assessment, we used the average concentrations of metals in the smallest particle size (< 75 µm), because it has the highest potential for human uptake and will provide a conservative assessment.

Table 3 Hazard quotient and exposure risk for selected metals in roadside dust of this study
Full size table

According to our calculations, exposure through ingestion of dust is the route of exposure that results in the highest intake followed by the route of dermal contact for both adults and children. It is reasonable, because dust ingestion through hand-to-mouth activity is a well-known exposure pathway for humans, especially children (Xue et al. 2007). The intake due to inhalation of dust is low compared with intake from the other two pathways mentioned; for some metals, such as Pb, Zn, and Cu, it could be several orders of magnitude lower. This observation also was reported by Zheng et al. (2010), stating that heavy metal exposure risk through inhalation is almost negligible compared with the other routes of exposure.

The HIs from all exposure routes also were calculated for those elements (Table 3). All HI values are < 1, suggesting no risk of adverse health effect due to exposure to those five metals in roadside dust along the studied highways. However, it is noted that the HI values of children are several times greater than those of adults, similar to what was found by other previous studies (Masto et al. 2017; Yu et al. 2016b). Children have a higher risk from exposure to metals in roadside dust than adults, probably due to more frequent hand-to-mouth activities (Xue et al. 2007).

Among the five metals assessed, Cr could pose the highest noncarcinogenic risk for both adults and children, with the HI values of 0.195 and 0.537, respectively, not too far from the safe threshold of 1. It is followed by Mn, having the HI values of 0.031 and 0.097 for adults and children, respectively. The overall ranking of noncarcinogenic risk of metals in this study is Cr > Mn > Pb > Cu > Zn for both adults and children. The ranking of noncarcinogenic is different than the ranking of average metal concentrations presented in Table 1 where Mn has the highest average concentration, followed by Zn, Cr, Cu, and Pb. This difference reflects the combined effect of metal concentration and its toxicity (expressed through the value of RfD). For example, Mn has an average concentration approximately three times that of Cr; however, the risk it poses is lower than that of Cr, because the RfDs of Cr for ingestion and dermal contact are much lower than the corresponding values of Mn.

Carcinogenic Risk of Cr

The CR value calculated for Cr is 3.92E−7, which means that the cancer risk caused by Cr in roadside dust is negligible in the studied area. The acceptable range of CR value is from 1E−06 to 1E−04, whereas CR values > 1E−04 will require some remediation responses (García-Rico et al. 2016). It is noted that although the carcinogenic risk of Cr is low/negligible, its noncarcinogenic risk is still the highest among studied metals with the concentration range comparable to what were found in soils across the Red River Delta (northern Vietnam) as reported by Tra and Egashira (2001), Phuong et al. (2010), and Van Nguyen et al. (2016) (Table 1).

Conclusions

This is the first study to monitor metal concentrations and their potential risks in roadside dust along two important highways in northern Vietnam. The concentrations of metals increased with the decreasing particle size in all samples, except for K. The evaluation of enrichment factor and geo-accumulation index indicated low-to-moderate pollution of metals in the collected dust samples. The concentrations of metals varied between the two highways, with the higher concentrations of metals observed in roadside sample from Highway No. 5, probably due to considerable higher traffic density. Among the five types of land uses, industry seems to be the highest potential source of heavy metals in roadside dust, followed by commercial area and power plant. The health risks of five heavy metals were assessed and considered to be lower than the safety threshold. Overall, the health risk of exposure to metals in roadside dust in two major highways in northern Vietnam was relatively low, but the risk for children is higher than for adults.