Annals of Occupational Hygiene Advance Access originally published online on May 15, 2009
Annals of Occupational Hygiene 2009 53(5):475-484; doi:10.1093/annhyg/mep033
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ultrafine Particle Characteristics in Seven Industrial Plants
1 Department of Applied Environmental Science, Atmospheric Science Unit, Stockholm University, 106 91 Stockholm, Sweden
2 Department of Occupational and Environmental Medicine, Örebro University Hospital, 701 85 Örebro, Sweden
* Author to whom correspondence should be addressed. Tel: +46-8-674-7763; fax: +46-8-674-7325; e-mail: Karine.Elihn{at}itm.su.se
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
Ultrafine particles are considered as a possible cause of some of the adverse health effects caused by airborne particles. In this study, the particle characteristics were measured in seven Swedish industrial plants, with a special focus on the ultrafine particle fraction. Number concentration, size distribution, surface area concentration, and mass concentration were measured at 10 different job activities, including fettling, laser cutting, welding, smelting, core making, moulding, concreting, grinding, sieving powders, and washing machine goods. A thorough particle characterization is necessary in workplaces since it is not clear yet which choice of ultrafine particle metric is the best to measure in relation to health effects. Job activities were given a different order of rank depending on what particle metric was measured. An especially high number concentration (130 x 103 cm–3) and percentage of ultrafine particles (96%) were found at fettling of aluminium, whereas the highest surface area concentration (up to 3800 µm2 cm–3) as well as high PM10 (up to 1 mg m–3) and PM1 (up to 0.8 mg m–3) were found at welding and laser cutting of steel. The smallest geometric mean diameter (22 nm) was found at core making (geometric standard deviation: 1.9).
Keywords: exposure • particle characterization • size distribution • ultrafine particles • workplace aerosols
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
During recent years, there has been an increasing interest in the smallest particle fraction, called ultrafine particles or nanoparticles (approximately <100 nm) since it has been suggested that exposure to these particles can explain some of the adverse health effects of particulate matter (Seaton et al., 1995; Donaldson et al., 2001).
Occupational exposure to airborne particles may be high as workers often work close to the source. Ultrafine particles in workplaces may originate from several sources (Vincent and Clement, 2000) such as (i) combustion processes, e.g. diesel engines (Wichmann, 2007) and natural-gas burners (Peters et al., 2006), (ii) hot processes, e.g. smelting (Thomassen et al., 2006), welding (Zimmer et al., 2002), and laser cutting (Haferkamp et al., 1998; Klein et al., 1998), and (iii) high-speed mechanical processes, e.g. grinding (Zimmer and Maynard, 2002). Especially, high particle concentrations (1 x 106 to 6 x 106 cm–3) are reported close to welding (Zimmer and Biswas, 2001) and smelting activities (Evans et al., 2008). Many other industrial activities generate 10 000–100 000 of particles (Möhlmann, 2005; Thomassen et al., 2006; Evans et al., 2008).
Ultrafine particles in workplaces can be monitored by particle number and surface area concentrations and size distribution (Brouwer et al., 2004; Handy et al., 2006). Most instruments commercially available today are not suitable for personal sampling since they are too large, too heavy, and additionally are mains operated. Thus, at present, stationary sampling has to be selected in workplaces in order to monitor and assess exposure to ultrafine particles.
It is not yet clear which particle metric (mass, surface area, number, and size distribution) would be the most relevant to measure in relation to health (Wittmaack, 2007) in workplaces. Different health end points may even require different particle metrics (Maynard and Aitken, 2007). It is suggested that large number of ultrafine particles entering the lungs complicate the clearance of the alveolus for the macrophages (Stone and Donaldson, 1998), while recent studies describe the relation between the particulate surface area and their toxic effects (Stoeger et al., 2006; Monteiller et al., 2007). The size of inhaled particles may also be of importance for health effects since the size is crucial for particle deposition in the respiratory tract [International Commission on Radiological Protection (ICRP), 1995].
In this paper, the characteristics of airborne particles, especially the ultrafine particle fraction, are described for seven Swedish plants, whereas the correlation with the concentration of simultaneously determined inflammatory biomarkers in the blood of some of the workers is described elsewhere (Ohlson et al., 2009).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
Measurement positions
Stationary particle measurements were carried out at a total of 23 positions in seven Swedish industrial plants (Plant A–G). Ten different job activities were examined, including fettling, laser cutting, welding, smelting, core making, moulding, concreting, grinding, sieving, and washing, see Table 1. The particle instruments were positioned as close as possible to the job activities without disturbing the ongoing work. The measurements reflect the particle characteristics at the area close to a job activity, but do not represent personal sampling of the workers. Two to five measurement positions were chosen in each plant. Most of the measurements were performed in the range 1–3 m from the job activities. Several other job activities were sometimes going on simultaneously in the same hall in the plants (>10 m away). Each measurement position is mainly characterized by one job activity; however, other activities in the halls could in some cases also have influenced the aerosol characteristics, which are described in more detail in Table 1 and the Results and Discussion sections. Therefore, the aerosol characteristics presented in this paper do not always exclusively originate from one source. All measurements lasted for 6–8 h, apart from a few complementary measurements of the particle surface area that were shorter (2 h).
|
Industrial plants
A brief description of the job activities and control measures are presented in Table 1. In Plant A, laser cutting and metal active gas (MAG)/metal inert gas (MIG) welding of steel were performed in two separate halls, only connected by a small door that was normally closed during measurements. In Plant B, spot welding and grinding of sheet metal (steel) were performed at two separate locations. In Plant C, particle measurements were performed close to job activities, such as grinding (Grinding 2 without shields and Grinding 3 in a hooded machine) and washing of foundry iron. All measurement positions in Plant C were located in the same hall, but separated by a distance of 20–50 m. In Plant D (iron foundry) and Plant E (aluminium foundry), activities such as smelting, moulding, core making, and grinding were performed and in Plant E also fettling (removal of excess moulding material/casting irregularities from the cast component by band sawing). In Plant D, smelting and moulding were performed in the same hall, while core making and grinding occurred in other rooms connected to the main hall by open doors. In Plant E, smelting, moulding, and core making were performed in the same hall, and in the next hall, connected to the first hall by large openings in the wall, fettling occurred. Grinding activities in Plant E were located in a separate workshop and performed in a booth with exhaust ventilation. In Plant F, concrete constructions were made: at Concreting 1 concrete and iron reinforcing were cut, at Concreting 2 and 3 concrete moulds were prepared and used for the production of concrete constructions, and cutting of concrete and iron reinforcing was performed. Concreting 1 and 2 in Plant F were located in nearby halls, connected by a large opening in the wall. The job activity at Concreting 3 in Plant F was performed along the concrete constructions in a long building, and therefore not always in close vicinity to the particle measuring instruments (2–35 m). In Plant G, steel powders were produced from melted metal. Airborne particles were measured in Plant G at job activities, such as smelting, moulding, and sieving powders. All job activities presented in Plant G were performed at separate locations. Plant A, C, and D sometimes used forklift trucks. However, most transportation within the plants was performed by overhead cranes.
Instruments and analyses of results
Particle number size distributions were monitored with a scanning mobility particle sizer (SMPS), including a differential mobility analyser (DMA, model 3071A, TSI GmbH, Aachen, Germany) and a condensation particle counter (CPC, model 3022, TSI GmbH). Particle sizes from 10 to 760 nm were measured. Occasionally, another CPC (model 3010, TSI GmbH) was used, which measured particle sizes from 10 to 460 nm. A closed loop of sheath air was used for the DMA, including an external filter for the removal of any particles. The sheath air was pumped with an external pump (3 l min–1) and a stainless steel tubing was wound about 20 times (d = 10 cm) and used to cool the gas after passage through the pump. The sample to sheath flow ratio was 1:10. A complete scan of the particle size distribution took 5 min at the set conditions. The number of ultrafine particles is generally high compared to lager particles. Most particles, based on the particle number concentration, were <760 nm in the plants. Therefore, the number concentration and number fraction of particles in the size range 10–100 nm could be calculated from the SMPS data.
At most locations, the aerosol was relatively stable concerning the mean size during the day, but varied in concentration. Therefore, the average of all size distributions as generated by the SMPS during one workday was used to represent the particle size distribution at one measurement position. The average size distributions were fitted to a log-normal distribution in order to obtain the geometric mean diameter (GMD) and geometric standard deviation (GSD). In a couple of positions (Sieving 1 and Welding 2), the size distribution shifted during the day so that the workday average distribution became bimodal. In these cases, two GMD are presented. In Plant D, a bimodal distribution was present during the whole day, but the peak of the smallest mode was below the size range of the SMPS.
The particle number concentration was measured with a P-Trak (model 8525, TSI GmbH) in the size range 20 nm to 1 µm. It was logged each second during the whole working day. The results are presented as median concentration together with the inter-quartile range and peak concentration. Background number concentration was measured prior to the start of job activities.
The total particulate surface area concentration was monitored with a diffusion charging particle sensor LQ1-DC (Matter Engineering AG, Wohlen, Germany) in the size range from a few nanometres up to 10 µm. The sensor measures surface area effectively for particles between 
20 and 100 nm (Ku and Maynard, 2005), and therefore the surface area concentration of aerosols dominated by particles <20 or >100 nm may be underestimated. The surface area was logged every fourth second and is presented here as median surface area concentration together with the inter-quartile range and peak concentration. In Plant A, a diluter had to be used prior to the particle sensor in order not to reach to the upper measuring limit of the surface area concentration. A rotating disc diluter for emissions, MD19-2E (Matter Engineering AG), was used for this purpose.
The mass of particles was measured with a micro-orifice uniform deposit impactor (MOUDI, model 110, MSP Corporation, Shoreview, MN, USA). At 11 impactor stages, it collects particles onto aluminium foils, sprayed with bounce-reducing grease (Spray MountTM, Scotch, 3M). The first stage collects particles >18 µm and the following steps >10, 5.6, 2.5, 1.8, 1, 0.56, 0.32, 0.18, 0.1, and 0.056 µm. No backup filter was used. The MOUDI was run at 30 l min–1. Collected particles were analysed by weighing. Prior to weighing, freshly prepared foils and foils with sampled particles, respectively, were conditioned for 24 h in an air-conditioned room with temperature and relative humidity controlled to 20 ± 0.5°C and 50 ± 2%, respectively. Three to five blanks were prepared and used to correct the sample weights for foil weight change between weighing. PM10 concentrations were calculated from the MOUDI data by the use of Simpson's formula as described by Hinds (1986). PM10 was used in this study to be able to compare to other studies that have measured PM10 in relation to inflammatory biomarkers. PM1 concentrations were estimated by adding the mass of the five last stages of the MOUDI, i.e. from Stage 6 for particles between 1 and 0.56 µm down to Stage 11 for particles between 0.1 and 0.056 µm. The mass for particles <0.056 µm is very low and does not affect the PM1 and PM10 mass concentrations. Particles were collected with the MOUDI for 8 h.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
The results from all particle measurements are presented in Table 2, whereas Table 3a–e gives a generalized picture of the particle exposure due to job activity.
|
|
Particle number concentration
Table 3a shows the median number concentration for the job activities. At fettling, a number concentration of 130 x 103 cm–3 was measured, which is the highest median concentration measured in this study. At typical hot processes, such as laser cutting, welding, and smelting, concentrations between 30 x 103 and 100 x 103 cm–3 were found. At core making and moulding, where most of the preparations occurred at room temperature, concentrations between 20 x 103 and 80 x 103 cm–3 were measured. At concreting, grinding, sieving, and washing lower particle concentrations (10 x 103 to 20 x 103 cm–3) compared to the other activities were generally measured.
Fraction of ultrafine particles
The number percentage of particles <100 nm is presented in Table 3b. Generally, the GMD of the number size distributions at the job activities were <100 nm, which explains why large fractions of ultrafine particles were found. At fettling, >95% of the particles were ultrafine. High fractions of ultrafine particles (90–95%) were also found at core making, moulding, and concreting. Among the hot processes, smelting had a high fraction of ultrafine particles (90–95%), in contrast to laser cutting (<10%) and welding (20–60%). For grinding, sieving, and washing, 60–90% of the particles were ultrafine. At the job activities with high particle concentration, the percentages of ultrafine particles were often high. However, this relation was not valid at laser cutting and welding. The low percentage of ultrafine particles at these job activities is in agreement with the large GMD's found there.
Particle surface area concentration
The median particle surface area concentration is presented in Table 3c. The surface area concentration followed the same trend as the number concentration measured at the same position. The largest surface area concentration (up to 3800 µm2 cm–3) was measured at the hot processes laser cutting and welding, whereas the other job activities resulted in a considerable lower surface area (most of them <300 µm2 cm–3). The smallest surface area concentrations were measured at grinding, sieving, and washing (50–200 µm2 cm–3).
Particle mass concentration
Table 3d shows the particle mass concentration in terms of PM10 for the job activities. The highest PM10 values (0.4–1.0 mg m–3) were measured at laser cutting and welding. This fact agrees with the high number concentration of particles >100 nm for these activities. Most job activities had PM10 values between 0.2 and 0.4 mg m–3, see Table 3d. The lowest PM10 values were found at job activities such as grinding and washing, most of them between 0.1 and 0.2 mg m–3. The mass concentration does not generally correlate with the particle number concentration. Due to their small size, the total mass of ultrafine particles is very small compared to the mass of larger particles, even though the ultrafine particles are often very numerous (see Table 3b).
Fraction of sub-micron particles
Table 3e shows the percent by mass of the PM10 particles which are <1 µm. The results are presented as mass-% of particles <1 µm for the job activities. At the hot processes, laser cutting, and welding, a majority of the mass was for particles <1 µm (70–90%). The lowest mass fractions were found at core making and moulding (10–30%) and sieving (10%). The use of sand and handling of micron-sized powders during these job activities agree with these findings.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
Comparison with other studies
Other studies of industrial processes that present particle number concentration are compared with the results presented in this paper.
Fettling of aluminium, as presented here, generated a particle number concentration of 130 x 103 cm–3, which is within the concentration range reported for steel fettling (118 x 103 to >500 x 103 cm–3) (Wake et al., 2002) and cutting of aluminium (10 x 103 to 260 x 103 cm–3) (Schneider et al., 2007).
MIG and MAG welding [gas metal arc welding (GMAW) methods] without local exhaust ventilation (LEV) was performed at Welding 1. The particle number concentration measured at this job activity (69 x 103 cm–3) was within the concentration range found at couple of other welding facilities, but lower than several other reported concentrations. At GMAW using varying voltage, the number concentration is reported to be approximately between 40 x 103 and 150 x 103 cm–3 (Hovde and Raynor, 2007) and at MAG welding with LEV 20 x 103 to 100 x 103 cm–3 (Möhlmann, 2005). Higher particle concentrations are reported for MAG welding without LEV (100 x 103 to 1000 x 103 cm–3) (Möhlmann, 2005) and for MIG welding (117 x 103 to >500 x 103 cm–3) (Wake et al., 2002). Especially, high particle concentrations are presented for measurements above the arc (19.2 cm) during GMAW (5720 x 103 cm–3) (Zimmer and Biswas, 2001).
Smelting activities in potrooms of a primary aluminium smelter reports particle concentrations in the range 10 x 103 to 50 x 103 cm–3 (Thomassen et al., 2006), which agrees well with the concentrations during smelting of aluminium at Smelting 3 (50 x 103 cm–3). Smelting in an iron foundry resulted in much higher particle concentrations (1600 x 103 cm–3) (Evans et al., 2008) compared to all smelting activities presented in this study (39 x 103 to 96 x 103 cm–3). In the same iron foundry, particle concentration measured at core making and moulding (120 x 103 cm–3) (Evans et al., 2008) is about five times higher than at corresponding activities in Plant D (25 x 103 cm–3).
Particle concentrations at grinding of iron (35 x 103 cm–3) (Evans et al., 2008), steel (80 x 103 cm–3), and aluminium (120 x 103 cm–3) (Zimmer and Maynard, 2002) generated two to five times more particles than at grinding of iron/steel (15 x 103 to 23 x 103 cm–3) and aluminium (64 x 103 cm–3) as presented here. In another paper, grinding of metal is reported to generate particle concentrations between 10 x 103 and 130 x 103 cm–3 (Schneider et al., 2007).
Particle number concentration at sieving of precious metal black (23 x 103 to 71 x 103 cm–3) (Wake et al., 2002) is higher than the median concentration measured during the sieving of steel powders in Plant G presented here (12 x 103 to 20 x 103 cm–3).
Many of the median particle number concentrations reported in this study are lower or in the same range as many other particle measurements in industrial plants. The quality of the general ventilation and use of LEV play an important role for the number concentration in workplaces (Lee, 2007) but also the use of forklift trucks. LEV was used at six job activities, see Table 1, where the particle number concentrations probably would have been higher without LEV, especially at laser cutting and Grinding 5 where new LEV had been installed. Three of the plants (Plant A, C, and D), see Table 1, used forklift trucks, which affected the particle concentrations at the time they were passing the instruments. However, since these trucks were only used briefly, the contribution to the median particle concentration was low.
Background particle number concentration was quite low in comparison to the median number concentration in many of the industrial plants in this study (3–10%). Where the particle concentrations were in the lower range of measured concentrations (10 x 103 to 25 x 103 cm–3), the major part of the particles originated from the studied job activities, although the particle contribution from the background was higher (20–25%).
Fettling
The highest median number concentration in this study was found at fettling of aluminium (Plant E). Even though aluminium was band-sawn at a relatively low speed (which means that it cannot be considered a high-speed process), it still resulted in a large number fraction of ultrafine particles (96%) as well as a high particle number concentration (130 x 103 cm–3). Processing of the metal aluminium at fettling may have given a higher particle number concentration at this job activity compared to iron processing (which was processed in the other plants) due to the higher vapour pressure of aluminium. The particle concentration at fettling was not considered to be influenced by particles emanating from the adjoining hall, where smelting and the preparation of moulds and cores were performed (Smelting 3, Moulding 2, and Core making 1 in Plant E), since concentrations found in there were all lower than at fettling.
Laser cutting, welding, and smelting
Hot processes are generally considered to generate high concentrations of ultrafine particles in the air. In this paper, particle characteristics were measured at three typical hot processes; laser cutting, welding, and smelting. At these job activities, relatively high number concentrations of particles were found. However, only at smelting did the fraction of ultrafine particles exceed 60% (85–94%, GMD = 30–50 nm). At laser cutting and welding, a relatively low number of the particles were ultrafine (laser cutting 8%, Welding 1: 26%, and Welding 2: 59%). Several different welding methods were used in the plants, which may have affected the particle size (Schneider et al., 2007). Considerably, larger particles were measured at Welding 1 (GMD = 235 nm) using MAG and MIG welding compared to Welding 2 (GMD = 33 and 100 nm) using spot welding (and also MIG welding in the same hall). The measured size distributions reveal that most particles in terms of number at laser cutting (92%) and Welding 1 (74%) were in the sub-micron range (0.1–1 µm). At these job activities, the sub-micron particles constitute the major fraction of the particle mass, as seen from the fraction of particle mass <1 µm (82–88%), see Table 2.
At laser cutting and Welding 1 (Plant A), a rotating disc diluter was used prior to the LQ1-DC in order not to reach the maximum measuring limit of the surface area concentration. According to Bernemyr (2007), there is no difference in dilution performance of the rotating disc diluter for particles >50 nm. Observations for particles <50 nm (10 and 27 nm) gave somewhat higher dilution, which was associated with diffusion losses of small particles. The aerosol in Plant A was dominated by particle >50 nm (GMD of 235 and 450 nm), and therefore no correction of the surface area concentration due to the use of the diluter was needed.
The largest surface area concentrations were found in Plant A. However, since the LQ1-DC only efficiently measures particles <100 nm, the surface area concentrations in Plant A may have been even larger in reality.
Core making and moulding
Core making and moulding are among the job activities where >90 number-% of the particles were ultrafine (GMD = 22–48 nm). Core making 1 and 2 and Moulding 2 and 3 are room temperature processes. The number of airborne ultrafine particles generated at room temperature processes is generally considered to be low. However, relatively high concentrations of particles (median concentrations of 25 x 103 to 75 x 103 cm–3) were measured at these positions, in combination with high percentages of ultrafine particles. In these cases, however, the measured number concentration presumably was influenced by particles generated at metal smelting and filling melted metal into moulds (Smelting 1 and 3) that occurred close (5–25 m) to the preparation of cores and moulds. The mass of airborne particles in terms of PM10 was 0.3 mg m–3 at most of the core making and moulding activities. At Moulding 2, however, PM10 was much higher (1.13 mg m–3) due to a cloud of dust that was leaking out at the time of measurement and is not considered as a normal PM10 value at Moulding 2. Instead, it can be reasonable to expect a PM10 of 0.3 mg m–3 at Moulding 2 under normal conditions. The particle number as measured by P-Trak and the size distribution as measured by SMPS were not affected by this incident and it can therefore be concluded that the dust cloud only contained particles >1 µm. At Moulding 1, the moulds were heated during preparation, which may explain the relatively high concentration of particles (81 x 103 cm–3).
Concreting
At Concreting 1, a cutting machine was continuously cutting concrete and reinforcing in a shielded system during the whole day (number concentration 22 x 103 cm–3). Only 65 number-% of the particles were ultrafine at Concreting 1, which agrees with the GMD (98 nm). At Concreting 2 and 3, concrete moulds were prepared and used for production of concrete constructions. Most of the particles at concreting activities, however, were formed during the cutting and sawing of concrete, including reinforcing. More than 90 number-% of the particles were ultrafine particles at Concreting 2 and 3, which agree with the small particle size found at these positions (GMD: 63 and 68 nm, respectively). The cutting activities did not occur continuously at Concreting 2 and 3, so the median particle concentration during the whole day turned out to be lower (12 x 103 cm–3) compared to at Concreting 1. During cutting, however, peaks from 50 x 103 up to over 100 x 103 cm–3 (duration 5–20 min) were measured at Concreting 2 and 3, see Fig. 1.
|
Grinding
Grinding is not considered as a true hot process, but could be a high-speed process where the metal may become heated and evaporated, and is therefore also likely to generate a relatively high proportion of ultrafine particles. At Grinding 1–4, ultrafine particle fractions between 63 and 84 number-% were found and relatively low median number concentrations compared to the other job activities in this study (15 x 103 to 23 x 103 cm–3). The low median concentration of particles may be due to a relatively low frequency of grinding activities at these sites at the time. During grinding activities, however, peak concentrations of 32 x 103 to 500 x 103 cm–3 were measured. At Grinding 5 in Plant E, a higher percentage of the particles were ultrafine (95%) and a higher concentration (64 x 103 cm–3) was found compared to the other grinding activities. This may be a result of continuous grinding during the day. The fact that aluminium was processed at Grinding 5 may also have influenced the particle number concentration since aluminium has a higher vapour pressure than iron (which was processed at Grinding 1–4). However, the particle concentration at Grinding 5 was considerably lower than at fettling of aluminium, which may be explained by the fact that Grinding 5 was performed in a booth with LEV.
Sieving
Ninety percent of the particle mass at Sieving 1 and 2 originated from particles >1 µm, which was expected due to the production of micron-sized powders. However, many ultrafine particles were found (80 number-%) which also seemed to originate from the sieving process since the particle concentration decreased to background levels during non-operational conditions. The powders, however, are generated by the spraying of melted metal, where ultrafine particles are also most likely to be formed. Release of these particles during sieving may explain the relatively high concentration of ultrafine particles in this process.
Washing
Washing of machine goods was performed in the same building as Grinding 2 and 3 in Plant C. The measured particle concentrations at washing could have been affected by the surrounding grinding activities as judged from the similar values, especially those at Grinding 3, which was closest (20 m) to the washing activity.
Number size distribution
The number size distribution of particles was commonly approximately unimodal as measured during one working day, with GMDs between 20 and 50 nm for many of the job activities, see e.g. the size distribution at Smelting 2 in Fig. 2. GSDs varied between 1.3 and 2.2, which are typical values for size distributions with one dominating source. GMDs between 22 and 37 nm were mainly found in Plant E (a foundry). Generally, the GMDs of particles were more similar within plants than between the same job activities (in different plants). At Concreting 2 and 3, the GMDs were about the same (63 and 68 nm). However, at Concreting 1 in Plant F, a larger GMD (98 nm) was found, although the cutting activity was similar at all three measurement positions. The largest particles in this study were found at laser cutting (GMD = 450 nm) and Welding 1 (GMD = 235 nm). At Welding 2, two size distribution modes (GMD at 33 and 100 nm) appeared during the day, either one at a time or occasionally at the same time. Seventy-five percent of the total particle concentration was present in the second size distribution mode. At Sieving 1, two size distribution modes were also present, GMDs at 26 and 70 nm, either one at a time or occasionally at the same time. Sixty percent of the total particle concentration was present in the second size distribution mode. Bimodal distributions were found in Plant D during the whole day of measurement (Core making 2, Moulding 3, and Grinding 4). The peak of the first size distribution at these activities was not within the size range of the SMPS (<10 nm), while the second size distribution mode had a GMD of 50 nm.
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
At laser cutting and welding, especially large particles and surface area concentrations were found, compared to those at the other work activities. At these hot processes, the particle concentrations were generally higher compared to those at colder processes. The material being processed may also be of importance for the particle concentration. In the plant that was fettling and grinding aluminium, higher particle concentrations were generally measured compared to plants grinding iron. At welding, the use of different methods may influence the particle characteristics, as seen in the comparison between MAG/MIG and spot/MIG welding in this study and as reported by Schneider et al. (2007).
Different order of rank depending on what particle metric was measured appeared at the job activities. The best choice of ultrafine particle metric in relation to health effects in workplaces is not yet fully understood. Therefore, it is important to measure several particle metrics in an attempt to find suitable particle characteristics to use in relation to health effects. The particle number is generally a good metric in the measurement of nano-sized particles since a high particle concentration is often connected to the presence of ultrafine particles. Although at laser cutting the particle number concentration was found to be high, the particles still turned out to be in the sub-micron range. It is important to determine the particle size from the health aspect since this metric is a determinant for the deposition probability in the different compartments of the respiratory system. Further, the surface area is suggested to be a good metric in relation to particle toxicity since the surface area that is available for contact with e.g. lung cells may be an important factor from that perspective. It can also be of importance to determine the chemical composition of the ultrafine particles through studies of their effect on human health. In this study, one aluminium foundry was included, while most of the other plants processed iron-based materials (and concrete), but the exact elemental composition of the particles is not presented here. Measurement of the particle mass concentration gives, in principle, only a measure of the amount of micron and sub-micron particles in the air and is generally not a good measure of ultrafine particles. However, it may still be important to measure the mass concentration of large particles in relation to health.
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
Swedish Council for Working Life and Social Research (FAS) (2004-1122).
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
We gratefully acknowledge all industrial plants and workers participating in this study.
|
FOOTNOTES |
|---|
The free full text of this article can be found in the online version of this issue.
Received August 21, 2008; in final form March 13, 2009
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
Bernemyr H. Volatility and number measurement of diesel engine exhaust particles (2007) Stockholm, Sweden. Doctoral Thesis in Machine Design. Available at http://kth.diva-portal.org/smash/record.jsf?pid=diva2:12482. Accessed 6 May 2009.
Brouwer DH, Gijsbers JHJ, Lurvink MWM. Personal exposure to ultrafine particles in the workplace: exploring sampling techniques and strategies. Ann Occup Hyg (2004) 48:439–53.
Donaldson K, Stone V, Seaton A, et al. Ambient particle inhalation and the cardiovascular system: potential mechanisms. Environ Health Perspect (2001) 109:523–7.[Web of Science][Medline]
Evans DE, Heitbrink WA, Slavin TJ, et al. Ultrafine and respirable particles in an automotive grey iron foundry. Ann Occup Hyg (2008) 52:9–21.
Haferkamp H, Bach Fr-W, Goede M, et al. Emissions generated during laser-cutting and safety precautions. Welding World (1998) 41:169–76.
Handy RG, Jackson MJ, Robinson GM, et al. The measurement of ultrafine particles: a pilot study using a portable particle counting technique to measure generated particles during a micromachining process. J Mater Eng Perform (2006) 15:172–7.[CrossRef]
Hinds WC. Cascade impactor, Sampling and data analysis. In: Chapter 3: Data analysis—Lodge JP, Chan TL, eds. (1986) Akron, OH: American Industrial Hygiene Association. 45–78.
Hovde CA, Raynor PC. Effects of voltage and wire feed speed on welding fume characteristics. J Occup Environ Hyg (2007) 4:903–12.[CrossRef][Web of Science][Medline]
International Commission on Radiological Protection (ICRP). Human respiratory tract model for radiological protection (1995) 21. Oxford, UK: Pergamon Press. ICRP Publication 66.
Klein R, Dahmen M, Putz H, et al. Workplace exposure during laser machining. J Laser Appl (1998) 10:99–105.[Web of Science]
Ku BK, Maynard AD. Comparing aerosol surface-area measurements of monodisperse ultrafine silver agglomerates by mobility analysis, transmission electron microscopy and diffusion charging. J Aerosol Sci (2005) 36:1108–24.[CrossRef][Web of Science]
Lee MH, McClellan WJ, Candela J, et al. Reduction of nanoparticle exposure to welding aerosols by modification of the ventilation system in a workplace. J Nanopart Res (2007) 9:127–36.[CrossRef]
Maynard AD, Aitken RJ. Assessing exposure to airborne nanomaterials: current abilities and future requirements. Nanotoxicology (2007) 1:26–41.[CrossRef][Web of Science]
Monteiller C, Tran L, MacNee W, et al. The pro-inflammatory effects of low-toxicity low-solubility particles, nanoparticles and fine particles, on epithelial cells in vitro: the role of surface area. Occup Environ Med (2007) 64:609–15.
Möhlmann C. Vorkommen ultrafeiner Aerosole an arbeitsplätzen. Gefahrst Reinhalt Luft (2005) 65:469–71. (in German).
Ohlson C-G, Berg P, Bryngelsson I-L, et al. Biomarkörer för inflammatoriska svar vid exponering för luftföroreningar i olika arbetsmiljöer. (2009) Report Dnr 2004-1122 at the Swedish Council for Working Life and Social Research. (in Swedish). Available at http://www.fas.se/. Accessed 6 May 2009.
Peters TM, Heitbrink WA, Evans DE, et al. The mapping of fine and ultrafine particle concentrations in an engine machining and assembly facility. Ann Occup Hyg (2006) 50:249–57.
Schneider T, Jansson A, Jensen KA, et al. Evaluation and control of occupational health risks from nanoparticles (2007) Copenhagen, Denmark: Nordic council of ministers. TemaNord 2007:581. Available at http://www.norden.org/publications. Accessed 6 May 2009.
Seaton A, MacNee W, Donaldson K, et al. Particulate air pollution and acute health effects. Lancet (1995) 345:176–8.[CrossRef][Web of Science][Medline]
Stoeger T, Reinhard C, Takenaka S, et al. Instillation of six different ultrafine carbon particles indicates a surface area threshold dose for acute lung inflammation in mice. Environ Health Perspect (2006) 114:328–33.[Web of Science][Medline]
Stone V, Donaldson K. Small particles—big problem. Aerosol Soc Newsl (1998) 33:12–4.
Thomassen Y, Koch W, Dunkhorst W, et al. Ultrafine particles at workplaces of a primary aluminium smelter. J Environ Monit (2006) 8:127–33.[CrossRef][Web of Science][Medline]
Vincent JH, Clement CF. Ultrafine particles in workplace atmospheres. Phil Trans R Soc Lond A (2000) 358:2673–82.[CrossRef]
Wake D, Mark D, Northage C. Ultrafine aerosols in the workplace. Ann Occup Hyg (2002) 46(Suppl 1):235–8.
Wichmann H-E. Diesel exhaust particles. Inhal Toxicol (2007) 19:241–4.[CrossRef][Web of Science][Medline]
Wittmaack K. In search of the most relevant parameter for quantifying lung inflammatory response to nanoparticle exposure: particle number, surface area, or what? Environ Health Perspect (2007) 115:187–94.[Web of Science][Medline]
Zimmer AT, Baron PA, Biswas P. The influence of operating parameters on the number-weighted aerosol size distribution generated from a gas metal arc welding process. J Aerosol Sci (2002) 33:519–31.[CrossRef][Web of Science]
Zimmer AT, Biswas P. Characterization of the aerosols resulting from arc welding processes. J Aerosol Sci (2001) 32:993–1008.[Web of Science]
Zimmer AT, Maynard AD. Investigation of the aerosols produced by a high-speed, hand-held grinder using various substrates. Ann Occup Hyg (2002) 46:663–72.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||