Characterization of welding fume and airborne heavy metals in electronic manufacturing workshops in Hangzhou, China: implication for occupational population exposure

Occupational exposure to contaminants created by electronic manufacturing process is not well characterized. The aim of this study was to carry out risk assessments of exposure to welding fume and airborne heavy metals (HMs) in electronic manufacturing workshops. Seventy-six air samples were collected from five sites in Hangzhou, China. In welding workshops, the most abundant contaminant found was welding fume, followed by Fe, Mn, Zn, Cu, Pb, Cd, and Cr. The concentration of Mn was positively correlated with Fe (r = 0.906). When compared with non-welding workshops, the Fe content in the air of welding workshops increased significantly (P < 0.05), while the Cu content decreased significantly (P < 0.05). Singapore semi-quantitative health risk assessment model and the United States Environmental Protection Agency (US EPA) inhalation risk assessment model were applied to assess the occupational exposure. In welding workshops, the levels of 8-h time weighted average (8 h-TWA) calculated for welding fume (range 0.288 ~ 6.281 mg/m3), Mn (range Nd ~ 0.829 mg/m3), and Fe (range 0.027 ~ 2.234 mg/m3) partly exceeded the permissible limits. While, in non-welding workshops, the average of 8 h-TWA for Cu (0.411 mg/m3) was higher than the limit. The risk rates (RR) assessed for Pb (2.4 vs 1.7), Mn (2.0 vs 1.4), and Fe (1.4 vs 1.0) were higher in welding workshops than that in non-welding workshops, but Cu (1.0 vs 2.2) were lower. The mean excess lifetime cancer risks (ELCR) in welding (5.59E − 06 per 1000 people) and non-welding (1.88E − 06 per 1000 people) workshops were acceptable. The mean non-cancer risk (HQ) estimated for Mn was greater than 10 in both welding (HQ = 164) and non-welding (HQ = 11.1) workshops. These results indicate that there was a risk of occupational exposure implication in the electronic manufacturing workshops. Reducing contaminant exposure through engineering controls and management strategies, such as efficient ventilation and reducing exposure hours, is thus suggested.


Introduction
Electronic manufacturing materials contain potentially hazardous chemicals that may become airborne during the process. Electronic workers are exposed to heavy metals (HMs) and/or welding fume when making metal parts of electronic components in welding and non-welding workshops.
Welding fume's composition is variable, since the main fume components derive from electrode and filler wire and from flux wherever used (Riccelli et al. 2020). HMs, including cadmium (Cd), aluminum (Al), copper (Cu), iron (Fe), chromium (Cr), manganese (Mn), lead (Pb), nickel (Ni), cobalt (Co), and zinc (Zn), are all metal oxide particles that can be present in welding fumes (Zimmer and Biswas 2001;Riccelli et al. 2020;Li et al. 2004). Since the composition of contaminants might vary with specific process tasks and time, contaminants created by electronic manufacturing process in both welding and non-welding workshops are not well characterized so far.
Studies have reported various adverse health effects on workers exposed to HMs in welding fumes, such as chronic obstructive pulmonary disease, pulmonary fibrosis, pneumonia, siderosis, asthma, metal fume fever, and cardiovascular disease (Coggon et al. 1994;Beach et al. 1996;Antonini 2003;Mocevic et al. 2015). Metal constituents of welding fumes pose potential hazards depending on their inherent toxicity. Vaporized Cu, Zn, or Cd present in welding fume may cause metal fume fever (Ohshiro et al. 1988;Sferlazza and Beckett 1991). Cd has been reported to be a cause of acute chemical inhalation lung injury (Anthony et al. 1978). Besides, Parkinson's disease has been hypothesized to be associated with manganese exposure received during the welding operation, but it is controversial and in need of further research (Racette et al. 2001(Racette et al. , 2005Park et al. 2005). In 2017, the International Agency for Research on Cancer declared welding fumes carcinogenic to humans (Guha et al. 2017).
Human health risk assessment is a useful approach to quantify potential health effects of human exposure to environmental contaminants. Yeo and Neo (1998) introduced an analytical hierarchy process in a health-hazard scoring system to quantify the environmental impact of welding processes. Leman et al. (2010) defined an environmental quality index (EQI) for occupational safety and health in the welding workplace. Karkoszka and Soković (2012) applied the integrated risk estimation in the analysis of metal active gas welding and metal inert gas welding processes using a qualitative method of assigning the probability of occurrence, significance of environmental impacts, and risk involved in occupational health and safety. Hariri et al. (2014) developed a welding fume health index (WFHI) that can rank welding workplace associate well with possible health risk of welders. Yang et al. (2018) used US EPA Incremental Lifetime Cancer Risk (ILCR) model to assess the cancer risks due to occupational exposure to Cr and Ni in welding fumes. Dueck et al. (2021) calculated 8-h time-weighted averages to assess health risk for apprentice welders.
Today, quantitative risk assessment is used as a basis for chemical legislation in many organizations, such as the World Health Organization (WHO) and the Environmental Protection Agency (HSE 2005). But very limited studies have carried out semi-quantitative and quantitative risk assessments of occupational exposure to contaminants in electronic manufacturing workshops including both welding and non-welding workshops. Due to the importance of chemical risk assessment for prevention the adverse health effects of chemicals in the electronics industry, this study was conducted with the aim of assessing the health risks of occupational exposure to welding fume and HMs in the electronic manufacturing workshops of Hangzhou City.

Research area
Multiple samples of welding fume and airborne heavy metals were procured in September 2021 from five sampling sites involved in electronic manufacturing in Hangzhou City, Zhejiang Province, China. Samples were collected from the worker's breathing zone during the work shift to explore the implication for occupational population exposure. The location and the information of the sampling sites are shown in Fig. 1 and Table 1.

Sample collection and chemical analysis
Welding fume was collected by drawing air at a constant flow rate of 20 L/min for 15 min through a pre-weighed dust measuring filter (Shanghai Xingya Corp., China). After sampling, the filters were transferred to the laboratory and weighed to determine the total mass of welding fume after drying for 2 h. An analytical balance (XSE 205DU, Mettler Toledo, Switzerland) with a minimum measurable weight of 0.1 mg was used for filter weighing. Airborne HM samples were collected using air samplers (CCZ-20(A), Zhejiang Hengda Instrument Corp., China) containing a quartz filter (Shanghai Xingya Corp., China) at a flow rate of 5 L/min for 15 min. Temperature, relative humidity, and pressure at the sampled sites were recorded. Each filter was digested with 5 mL of concentrated HNO 3 at 200 °C until nearly dry. After cooling, the solution was diluted to 10.0 mL with 1% HNO 3 . Concentrations of HMs were determined by a flame atomic absorption spectrophotometer (PinAAcle 900 T, Perkin Elmer, USA) at different wavelengths for each element. The limits of detection (LOD) for Cu, Zn, Cd, Pb, Mn, Fe, and Cr were 0.01, 0.05, 0.02, 0.06, 0.03 0.04, and 0.1 μg/mL, respectively. Recoveries of HMs were in the range of 90 ~ 110%. Each sample was duplicated for reproducibility and accuracy. Two blank filters were also analyzed following the same procedures.

Exposure concentration calculation
To assess health risk for occupational population exposure, exposure concentration of 8-h time weighted average (8 h-TWA ) was calculated using Eq. 1 and compared with permissible exposure limit (PEL) of permissible concentration-time weighted average (PC-TWA).
where 8 h-TWA (mg/m 3 ) is the exposure concentration of 8-h time weighted average, C i (mg/m 3 ) is the concentration in different exposure periods i, and T i (h) is the exposure time in exposure period i.

Singapore semi-quantitative health risk assessment
Risk level of welding fume and HM exposure was assessed using the Singapore semi-quantitative health risk assessment (Heibati et al. 2017). First, the weekly exposure (E) was calculated using Eq. 2.
where E (mg/m 3 ) is the weekly exposure, M (mg/m 3 ) is the exposure value, D (h) is the average time of each exposure, F is the number of exposure times per week, and W (h) is the average working hours per week. Then, the occupational exposure limit (OEL) is used to calculate exposure rate (ER). ER is determined based on the ratio of E/OEL and defined between 1 and 5 according to ER table (Appendix S1). Hazard rate (HR) is determined due to the effects of contaminants on health and according to the classification of American Conference of Governmental Industrial Hygienists (ACGIH) and International Agency for Research on Cancer (IARC) and using the HR table (Appendix S2). Finally, risk rate (RR) is calculated from the root of the product of exposure rate and hazard risk (RR = √(ER × HR) and ranked using the RR matrix (Appendix S3).

EPA inhalation risk assessment
Non-cancer risk assessment. Non-cancer risk was estimated for all HMs according to the US EPA procedures. The main pathway of health risk exposure to HMs is chemical daily intake through mouth and nose inhalation. Thus, the exposure concentration (EC) for exposures to the HMs in air via the inhalation pathway was estimated using Eq. 3 (US EPA 2009).
where EC (mg/m 3 ) is the exposure concentration, CA (mg/ m 3 ) is the concentration in air, ET (h/day) is the exposure 1-roll and electric welding process; 2-electric welding process B 18 2 1-electric welding process; 2-carbon dioxide shielded welding process C 3 1 1-electric welding process D 19 4 1-electric welding process; 2-argon arc welding process; 3-carbon dioxide shielded welding process; 4-spot welding process E 6 1 1-electric welding process Non-welding A 9 1 3-chip removing process time, EF (days/year) is the exposure frequency, ED (years) is the exposure duration, and AT (ED in years × 365 days/ year × 24 h/day) is the averaging time.
Calculated EC was subsequently divided by an inhalation reference concentration (RfC) to yield a non-cancer hazard quotient (HQ). The formula for determining HQ is as follow: Cancer risk assessment. The quantitative risk of carcinogens was calculated by using the Office of Environmental Health Hazard Assessment (OEHHA) method, based on the exposure rate, the number of contact hours per day, the lifetime number of years of working, the number of weeks and working days, and the unit risk level (URL) which is a specific amount for each chemical. The URL represents the number of cases of cancer resulting from that particular exposure under specified exposure conditions in every 1000 people. The formula for determining excess lifetime cancer risk (ELCR) is as follows (Mohammadyan et al. 2019a, b): where ELCR is the lifetime risk of increase in cancers per 1000 people, TWA (µg/m 3 ) is the time-weight average of the contaminant, H (h) is the working hours per day, and URL (per µg/m 3 ) is the unit risk level for the contaminant.

Statistical analysis
SPSS version 25.0 (IBM Corp.) were used for the analyses. For data less than the LOD, a value of 1/2 of the LOD was assigned. Since data were lognormally distributed, all data were log transformed prior to perform parametric statistical analysis. One-way analysis of variance was used for comparing the means in welding workshops and non-welding workshops. Pearson analysis and linear regressions were used to test the correlations between contaminants. The statistical significance was considered at P < 0.05.

Concentration of contaminants
Welding fume refers to the solid metal suspended in air that forms when vaporized metal condenses into very small particulates. So the major components of the fume are oxides of metals used in the manufacture of the consumable electrode wire fed into the weld (Antonini 2003). The mean of the concentrations for welding fume and HMs are shown in Table 2. In welding workshops, the average levels of contaminants detected ranked as welding fume (   HMs in welding fume ranked as Fe (22.6%), Mn (2.96%), Zn (1.45%), Cu (0.46%), Pb (0.28%), Cd ( −), and Cr ( −). Moreover, Fe (10/10), Zn (10/10), Cu (9/10), and Mn (8/10) were found to be present in most welding processes. Pb (5/10) was present in half of the welding processes. While Cd (0/10) and Cr (0/10) were non-detected in all welding processes. It was observed that the roll and electric welding process in welding site A-1 produced the highest concentrations of most contaminants, including welding fume (3894 μg/m 3 ), Pb (12.4 μg/m 3 ), Mn (258 μg/m 3 ), and Fe (1307 μg/m 3 ). The electric welding process in welding site B-1 produced the highest concentrations of Zn (58.4 μg/m 3 ). And the highest concentration of Cu (1646 μg/m 3 ) was produced by the chip removing process in non-welding site A-3. In general, air samples collected from welding workshops contained the higher concentrations of Fe, Mn, and Zn, with Fe concentrations being significantly higher (P < 0.05) than that collected from non-welding workshops. In addition, air samples collected from non-welding workshops were found to contain significantly higher (P < 0.05) concentrations of Cu than that in welding workshops.

Correlation analysis for contaminants
A Pearson correlation and linear regression on logtransformed welding fume and HM concentrations are shown in Table 3 and Fig. 2. Correlation analysis was not performed in non-welding workshops due to insufficient sample size. In welding workshops, relatively strong correlations were found for Cu and fume, Mn and fume, Fe and fume, Pb and Cu, Mn and Cu, and Fe and Cu with r ranging from 0.483 to 0.702 (R 2 ranging from 0.3006 to 0.5293). A strong correlation (r = 0.906, R 2 = 0.8204) was observed between Fe and Mn.

Occupational exposure assessment results
In this study, the exposure concentration calculation, Singapore semi-quantitative health risk assessment model, and EPA inhalation risk assessment model were performed to assess health risk for occupational population exposure. As Cd and Cr were non-detected in all of the samples, occupational exposure assessment was not performed for them.

Exposure concentration calculation
Contaminant concentrations were converted into 8 h-TWA and compared with PC-TWA. The exposure time of each process in sampling sites is listed in ET table (Appendix S4). And the overall results were shown in Table 4.
The PC-TWA recommended by the Ministry of Health of the People's Republic of China and the ACGIH are 4 mg/ m 3 for welding fume, 0.2 mg/m 3 for Cu, 2 mg/m 3 for Zn, 0.03 mg/m 3 for Pb, 0.15 mg/m 3 for Mn (as MnO 2 ), and 1 mg/m 3 for Fe. For the workers in welding workshops, the levels of occupational exposure to welding fume, Mn, and Fe were 1.025 (range 0.288 ~ 6.281) mg/m 3 , 0.055 (range Nd ~ 0.829) mg/m 3 , and 0.246 (range 0.027 ~ 2.234) mg/ m 3 , respectively, which exceeded the permissible limits in some sites. For the workers in non-welding workshops, the average of occupational exposure to Cu (0.411 mg/m 3 ) was found to be higher than the permissible limit. In addition, the levels of occupational exposure to other contaminants in both welding and non-welding workshops were lower than the permissible limits. These results suggest that the workers may be at risk of overexposure to welding fume, Mn and Fe in welding workshops, and at risk of overexposure to Cu in non-welding workshops.

Singapore semi-quantitative health risk assessment
According to the classification of ACGIH and IARC and using the HR table, the HR for studied contaminants is The ER was calculated based on 8 h of work per day, 5 days a week, and was compared with the PC-TWA of each contaminant which is shown in Table 4. And the results of Singapore semi-quantitative health risk assessment were shown in Table 5. For the workers in welding workshops, the average risk rate for contaminants ranked as welding fume (risk level = 2.4, low), Pb (risk level = 2.4, low), Mn (risk level = 2.0, low), Fe (risk level = 1.4, negligible), Cu (risk level = 1.0, negligible), and Zn (risk level = 1.0, negligible). For the workers in non-welding workshops, that ranked as Pb (risk level = 1.7, low), Cu (risk level = 2.2, low), Mn (risk level = 1.4, negligible), Fe (risk level = 1.0, negligible), and Zn (risk level = 1.0, negligible). It is noteworthy that the high risk rate due to exposure to welding fumes (risk level = 3.5, high), and the medium risk rates due to exposure to Pb (risk level = 3.0, medium) and Mn (risk level = 3.2, medium), all occurred in welding  workshops. These results indicate that workers in welding workshops may be at a higher risk rate due to exposure to welding fume, Pb, and Mn than that in non-welding workshops. While, workers in non-welding workshops may be at a higher risk rate due to exposure to Cu than that in welding workshops.

EPA inhalation risk assessment
EPA inhalation risk assessment was finally conducted and shown in Table 6. Non-cancer risk was estimated for all HMs, whereas cancer risk was calculated with Pb. The lifetime cancer risk (ELCR) of Pb was calculated based on 8 h-TWA (see in Table 4) and URL for lead, which is 1.20 × 10 −5 μg/m 3 (Mohammadyan et al. 2019a, b). The ELCR in welding workshops was 5.59E − 06 per 1000 people (range 5.89E − 07 ~ 2.03E − 05) and in nonwelding workshops was 1.88E − 06 per 1000 people (range 3.53E − 06 ~ 1.07E − 05). According to EPA standards, the acceptable risk level for exposure to chemicals is defined as one per 1,000,000 in environmental and 1 in 1000 in occupational exposures (Heibati et al. 2017). Given that the average cancer risk in the welding and non-welding workshops was both far less than one per 1000 people, over a 70-year life span, the risk of carcinogenesis was acceptable.
For non-cancer risk (HQ) estimations, the essential parameters associated with health risk assessment have been listed in Appendix S6. When HQ < 1, then non-cancer effects are impossible. When HQ > 1, there is the potential for unfavorable health consequences. If HQ > 10, then adverse a high chronic risk exists (Leung et al. 2008). In this study, it was observed that non-cancer risks (HQs) for Zn, Cu, and Fe were all less than 1.0 for both welding and non-welding workshops, indicating that non-cancer effects owing to Zn, Cu, and Fe exposure were impossible. However, the range of estimated HQ for Pb (range 0.0793 ~ 2.73) was greater than 1.0 in welding workshops, suggesting the potential for unfavorable health consequences owing to Pb exposure. Notably, the non-cancer risks estimated for Mn were greater than 10 in both welding (HQ = 164) and non-welding workshops (HQ = 11.1), manifesting a chronic high risk owing to Mn exposure. Furthermore, since the concentrations of HMs in different sites differed greatly, the EPA inhalation risk assessment was conducted according to the welding process and shown in Appendix S7. It was observed that the HQs estimated for Zn, Cu, and Fe were acceptable (HQ < 1) in all processes. However, the HQs estimated for Pb and Mn were above the limits in some processes. And the roll and electric welding process in welding site A-1 had the highest non-cancer risk of Pb (HQ = 1.48) and Mn (HQ = 919) when compared with the other processes. The average value of HQ for Mn estimated in the studied processes ranked as roll and electric welding process (HQ = 919), electric welding process (HQ = 56.4), gas-shielded welding process (HQ = 35.7), chip removing process (HQ = 11.1), and spot welding process (HQ = 2.97).

Characterization of contaminants
A summary of the contaminant concentrations in the air of different workplaces compared with other studies is shown in Table 7. In our study, Fe was found to be the chief component (22.6%) in welding fume generated from welding process, which is consistent with the previous findings (Dueck et al. 2021;Ellingsen et al. 2006;Insley et al. 2019;Hariri et al. 2018). Additionally, Mn is present in most welding fumes as shown in Table 7. Hassani et al. (2012) evaluated occupational exposure to Mn-containing welding fumes among natural gas transmission pipeline welders. In their study, the average calculated for percent of Mn in welding fume is 1.35% (range 0.88 ~ 1.57%), which is slightly lower than ours (2.96%, range 0.34 ~ 6.63%). The commonly presence of Mn in welding fume is probably due to a widely use of Mn as a flux agent and as an alloying element in electrodes. And some special steels containing a high content of Mn may produce a high concentration of Mn oxide in welding fume (Antonini 2003).  Comparison of HM concentrations in the air of welding and non-welding workshops has been previously described as shown in Table 7. Li et al. (2004) determined airborne Mn levels in the breathing zones of welders from a vehicle manufacturer and controls from a nearby food factory with no history of occupational exposure to Mn and other metals. Their finding showed that the concentration of ambient Mn in welding workshop was 10,000-fold magnitude higher than that in non-welding workshop. Insley et al. (2019) conducted a survey of airborne metal exposures to welders and bystanders in small fabrication shops. They found that for HMs detected in the welder samples, the corresponding bystander sample concentrations were lower. In our study, the concentrations of Fe, Mn, and Zn in the air of welding workshops were much higher than that in the air of nonwelding workshops. However, the concentration of Cu was found to be higher in the air of non-welding workplace than that in the air of welding workshops. In non-welding workplace, chip removing process was carried out there. Different from the welding process, the chip removing process is one of the last stages of manufacturing by machining, which is the finishing treatment and deburring of the product's edges. During chip removing process, the high content of Cu in the deburred materials could result in high emissions into the air depending on the type of process, operating temperature, impact force, and other factors. Our results indicate that the composition of contaminants in the air of workshops varies considerably with specific process tasks. We also found that the concentration of Mn was positively correlated with Fe (r = 0.906) in welding workshops. A similar strong correlation between Mn and Fe was also observed by Flynn and Susi (R 2 = 0.69) and Dueck et al. (r 2 = 0.92). Dueck et al. (2021) conducted the health risk assessment for welding fume exposure in a cohort of apprentice welder, and found the positive correlation between Mn and Fe which may be related to the similar ratio of Mn to Fe in electrodes. Flynn and Susi (2010) analyzed three databases consisting of welding exposure to Mn, Fe, and total particulate mass. The results showed strong correlations among Mn, Fe, and total particulate mass exposures, suggesting simple equations to estimate one fume component from any of the others. Results from the correlation and linear regression may be used to estimate Fe concentration as a first approach and then to determine if there is a risk of overexposure to Mn in the welding workshop. And this approach could allow the facility to perform more frequent monitoring of welders at a low cost.

Occupational exposure assessment
Occupational exposure to welding fume and HMs has been studied in other studies as well. Balkhyour et al. (2010) investigated workers' welding fume exposure levels in six industries in Jeddah, Saudi Arabia. In their study, the calculated 8-h average welding fume concentrations in four industries were above the permissible exposure limit value (5 mg/m 3 ) established by Saudi Standards. In this study, the calculated 8 h-TWA for welding fume was 1.025 (range 0.288 ~ 6.281) mg/m 3 , which partially exceeded the permissible limit (4 mg/m 3 ) recommended by China Standards. In addition, Flynn and Susi (2010) examined three databases suggesting that arc welding produces Mn exposures that are frequently close to or in excess of ACGIH threshold limit value (TLV). Dueck et al. (2021) conducted health risk assessment for 12 metals in a cohort of apprentice welders. Their findings showed that 8 h-TWA of Mn was in excess of ACGIH TLV for welding participants for day 50, indicating that welder apprentices were at risk for overexposure to Mn. The result of our study also indicated a risk of overexposure to Mn for the workers in some welding workshops.
Semi-quantitative risk assessment of occupational exposure to welding fume and HMs has been carried out in very limit studies. Mohammadyan et al. (2019a, b) conducted the semi-quantitative risk assessment of occupational exposure to Pb in solderers in the electronic industry of Neyshabur City. In their study, Pb exposure in solderers was 93 μg/m 3 and Pb exposure risk was high (risk level = 3.87). The lower exposure risk in this study compared with Mohammadyan et al. (2019a, b)'s study was because of the lower lead exposure (5 μg/m 3 ) among the welders of this study compared with Mohammadyan et al. (2019a, b)'s study (93 μg/m 3 ).
Quantitative assessment for occupational exposure to heavy metals in welding fumes has been studied in other studies as well. Yang et al. (2018) applied US EPA's Incremental Lifetime Cancer Risk model calculating the cancer risks of exposure to Cr and Cr (VI) in welding fumes that exceeded the acceptable level of occupational exposure (10 −3 ). A quantitative assessment has been done by Mohammadyan et al. (2019a, b) according to the OEHHA method, to assess occupational exposure to Pb among electrical solders in Neyshabur, Iran. In their study, the mean excess lifetime cancer risk (ELCR) was 0.11 per 1000 people and the mean non-carcinogenic risk (HQ) was 7.20. The results of their study indicate that there is a risk of non-carcinogenic complications among electronic solderers. Moreover, Dehghani et al. (2021) conducted health risk assessment of occupational exposure to HMs in a steel casting unit of a steelmaking plant. In their study, the calculated HQ due to Mn (range 368 ~ 7410) was far higher than the permissible level in all measured stations. In the present study, the non-cancer risk of occupational exposure to Pb (rang 0.0793 ~ 2.73) and Mn (2.96 ~ 2480) was over the acceptable value in some welding workshops. This may present an increased risk of adverse health consequences, particularly due to Mn, depending on the individual's susceptibility, duration of the exposures, and other factors.

Health effects of contaminants
Occupational exposure to welding fumes is a serious global health problem. When inhaled, welding fumes can enter the lungs, blood, brain, and other organs and result in short-term and long-term health effects. Acute exposure to welding fumes can lead to eye, nose, and throat irritation; metallic taste in the mouth; nausea; chills; fever; and muscle pain. The most frequently observed acute respiratory effect of welders is metal fume fever, which often simulates a flulike illness. Chronic exposure to welding fumes can cause respiratory effects, including decreased pulmonary function, coughing, and occupational asthma. The degree of the risk to welder's health from fume exposure depends on the concentration, composition, and length of exposure time (Antonini 2003;Antonini et al. 2004;Balkhyour and Goknil 2010).
Cu is an essential element. Interestingly, there is little published or reported evidence that occupational exposure to Cu causes adverse health effects. Haase et al. (2022) conducted a cross-sectional study of workers employed at a Cu smelter and found that there was no indication of an alteration of the lung function or chronic inflammation of Cu-exposed workers.
Fe is considered with little likelihood of causing chronic lung disease after inhalation. But welders with long-term exposure to Fe oxide particles may cause siderosis, a pneumoconiosis without complicated lesions or progressive fibrosis (Antonini 2003).
Pb is a toxic metal with widespread use. Pb poisoning has been a well-known disease which can affect various organs, including hematopoietic system, nervous system, reproductive system, and digestive system. Pb poisoning in workers is primarily due to respiratory exposure. It is known as a silent environmental disease with life-long adverse health effects (Mohammadyan et al. 2019a, b;Karrari et al. 2012).
Mn is an essential element but excess exposure induces neurotoxic effects. It has been hypothesized that Mn-containing welding fumes are a possible neurological hazard. After inhalation, the absorbed Mn is transported in the blood and crosses the blood-brain barrier, and preferentially damages different areas of the brain (Alici et al. 2022). Symptoms such as headache, sleep disturbance, and mood disorder can be observed after exposure to Mn. High-dose and chronic exposure to Mn can lead to the chronic neurological condition known as "manganism," which is a neurological syndrome that resembles Parkinson's disease, but can be differentiated based on their clinical, imaging, and pathological features (Racette et al. 2001). However, the neurotoxicity of low-dose and long-term exposure to Mn in welding fumes remains to be determined. Notably, most welders are continuously exposed to airborne concentrations of Fe that are considerably higher than that of Mn. It has been hypothesized that the neurotoxic effects of Mn-containing welding fumes may be influenced by the accumulation of Fe in the lungs/brain of welders because of the chemical similarities between the two metals (Riederer et al. 2001). The neurotoxic mechanism of Mn exposure in welders has not yet been fully elucidated due to paucity of adequate scientific reports.

Strengths and limitations
Although we have identified welding fumes and seven HMs in electric manufacturing environment, it is important to assess all welding fume components to ensure healthy welding environments. The small sample size of our study may limit the ability to identify the true distribution of contaminant concentrations. Future lager environmental monitoring studies coupled with biological monitoring of workers are required to verify the health effects of exposure to the measured contaminants.

Conclusion
Through collection of air samples, it was found that electronic manufacturing workers were exposed to contaminant concentrations which may be detrimental to their health. Compared with PC-TWA, welding fume, Mn, and Fe were identified overexposure in some welding workshops, while Cu was identified overexposure in most non-welding workshops. Moreover, the workers in some welding workshops were at a high risk rate due to exposure to welding fumes (risk level = 3.5), and at a medium risk rate due to exposure to Pb (risk level = 3.0) and Mn (risk level = 3.2). Although the calculated cancer risks were all below the standards, the non-cancer risks estimated for Mn were greater than the acceptable reference value in both welding (HQ = 164) and non-welding workshops (HQ = 11.1), thus manifesting a high chronic risk owing to Mn exposure. As a result, taking preventive actions, implementing engineering measures, developing health and safety training, and performing periodic medical examinations are highly recommended to protect the worker's health against contaminant exposure.

Declarations
Ethical approval Not applicable.
Consent to participate Not applicable. The study did not involve human participants.

Consent for publication
All authors declare they have given consent to publish this article.

Competing interests
The authors declare no competing interests.