International Journal of

Environmental Research
and Public Health

Article
Characterisation of Particles Emitted during Laser Cutting of
Various Metal Sheets and an Exposure Assessment for the
Laser Operators
Stine Eriksen Hammer , Johanne Østereng Halvorsen, Pål Graff and Torunn Kringlen Ervik *

National Institute of Occupational Health, 0363 Oslo, Norway

* Correspondence: [email protected]

Abstract: Laser cutting is used in many industrial settings to achieve precise cuts of metal sheets.
Laser operators may be exposed to particles formed during cutting when opening the cabinet or
when metal sheets are exchanged. To characterise the potential exposure, particles formed during
laser cutting were studied with scanning electron microscopy equipped with an energy dispersive
X-ray detector and an energy backscatter diffraction detector. The total concentration of particles
(11–615 nm) was determined online with a scanning mobility particle sizer. The chemical composition
of the particles formed during the cutting of the different metal sheets was determined by inductively
coupled plasma mass spectrometry (ICP-MS). X-ray diffraction was applied to determine the phase
composition. The occupational exposure was assessed gravimetrically and by ICP-MS for five laser
operators handling different laser cutters, and materials and were found to be low. Agglomerates
and aggregates of condensation particles were formed during laser cutting, independent of the sheet
type. Iron, present as both magnetite and α-Fe, was the main element found in the particles formed
when cutting steel sheets. The size of the particles generated was mainly below 300 nm. Open laser
cutters may lead to higher metal exposures, which is especially relevant when cutting metal sheets
containing heavy metals.
Citation: Eriksen Hammer, S.;
Halvorsen, J.Ø.; Graff, P.; Ervik, T.K.
Keywords: laser cutting; metal sheets; exposure levels; scanning electron microscopy; ICP-MS; X-ray
Characterisation of Particles Emitted
diffraction; scanning mobility particle sizer
during Laser Cutting of Various
Metal Sheets and an Exposure
Assessment for the Laser Operators.
Int. J. Environ. Res. Public Health 2022,
19, 9888. https://doi.org/10.3390/ 1. Introduction
ijerph19169888 Cutting metal sheets of various steel types, or nickel-(Ni), aluminium-(Al), and
Received: 17 June 2022
titanium-(Ti) sheets is widespread in different metal industries. Automated advanced
Accepted: 8 August 2022
cutting machines are used when manual cutting methods are unsuitable or ineffective.
Published: 11 August 2022
These advanced cutting machines include, but are not limited to, mechanical cutting by
abrasive water jet cutting as well as thermal cutting by plasma or laser beam. Laser cutting
Publisher’s Note: MDPI stays neutral
technologies with pollution strategies are summarised by He et al. [1]. Compared to the
with regard to jurisdictional claims in
other techniques, laser beam cutting is cost-effective and can cut any complex material with
published maps and institutional affil-
a thickness up to 10 cm with high productivity, and leaves relatively thin kerf, yielding
iations.
excellent precision of the cut [2]. An assistant gas jet removes the molten or vapourised
material from the cut zone and can additionally act as an energy source. An inert gas, such
as argon, provides mechanical force to eject material from the cut zone, while a reactive
Copyright: © 2022 by the authors.
gas, such as oxygen, can cause an exothermic reaction to accelerate the cutting process [3].
Licensee MDPI, Basel, Switzerland. The assist gas is chosen dependent on the material and whether cutting speed or cut edge
This article is an open access article quality is more important [2,4].
distributed under the terms and Condensation particles formed during thermal processes have been well characterised
conditions of the Creative Commons by welding [5], laser additive manufacturing [6–9], and in metal fume from other in-
Attribution (CC BY) license (https:// dustrial production processes [10,11]. Generally, these particles are ultrafine agglomer-
creativecommons.org/licenses/by/ ates/aggregates of condensates. The detailed characterisation of particles formed during
4.0/). laser cutting is reported from cutting plastic [12] and composite materials [13]. The particle

Int. J. Environ. Res. Public Health 2022, 19, 9888. https://doi.org/10.3390/ijerph19169888 https://www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2022, 19, 9888 2 of 13

concentration and size distribution of the particles have been studied in the workroom air
in relation to the laser cutting of metal sheets, but the concentrations in the study were
influenced by other sources, such as trucks driving past the sampling equipment [14]. In
a study of laser cutting in carbon steel, galvanised steel, and stainless steel, the particle
number concentration increased with decreasing sheet thickness, and the median particle
diameter was observed to be around 2–2.5 µm [15]. Waste products produced during
the laser cutting of steel sheets have been thoroughly investigated with scanning elec-
tron microscopy (SEM)-energy dispersive X-ray (EDX) analysis, SEM-energy backscatter
diffraction (EBSD), transmission electron microscopy (TEM), and X-ray diffraction (XRD)
to investigate the possibility of reusing these particles in additive manufacturing [16].
Occupational exposure studies have been conducted for metal additive manufacturing
in air [9,17] and by biomonitoring the workers’ blood and urine [18,19]. In all these studies,
the air-dust level was low, and well below respective occupational exposure limits (OELs).
Additionally, low levels of total Cr and hexavalent Cr were determined in the workroom
air. Still, the reports showed increasing trends in urine Cr over one working week and in
blood over one working year for additive manufacturers, indicating relevant exposure to
this metal [18].
In this study, the physicochemical properties of particles emitted during the cutting of
different metal sheets have been studied in detail. This includes the size distribution, num-
ber concentration, morphology, chemical composition, and mineralogy of particles formed
during the laser cutting of metal sheets. Additionally, the size distribution of particles in
the workroom air was studied together with an exposure assessment of particulate matter
(PM) in three different laser cutting facilities in Norway. The purpose of the study was
to gain knowledge about the particles emitted during laser cutting in regular production
settings, which is important for risk assessment for workers in this industry. Knowledge of
the potential exposure in these settings during laser cutting operations is central, as there
are many different materials being cut with this technique, but very little is known of the
potential exposure to laser operators.

2. Materials and Methods

2.1. Stationary Sampling
Particle number concentration and size distribution were analysed in real-time with
a TSI scanning mobility particle sizer (SMPS) instrument (model 3034, TSI Inc., Shoreview,
MN, USA) in the electrical mobility diameter size range of 11–461 nm for the fibre laser,
and 31–615 nm for the CO2 lasers, respectively. The particles were sampled from inside the
protective housing of the laser cutter with the use of an approximately 2 m-long antistatic
tube connected to the SMPS inlet. The sampling was performed during the cutting of
12 different metal sheets with four different laser cutters, one fibre laser (Bystronic, ByStar
Fiber 6 kW), and three different CO2 lasers (Trumpf TruLaser 5030 classic, Bystronic Bystar
4020 and Bystronic Byflex), of which Bysonic Byflex is without protective housing. The fibre
laser and Trumpf CO2 laser included a ventilation system, but the two other lasers had only
one point of air extraction. The lasers were operating under normal drift ensuring exposure
to relevant sampling. Particulate matter formed during the laser cutting was collected with
the use of pumps built in-house (NIOH, Oslo, Norway) and SKC filter cassettes (SKC Inc.,
Eighty-Four, PA, USA) loaded with 25 mm polyvinyl chloride (PVC) membrane filters with
a pore size of 5.0 µm (Merck Millipore, Burlington, MA, USA). This sampling equipment
was placed inside the protective housing of the laser beam cutter (Figure 1). For Bysonic
Byflex, the samples were collected as close as possible to the cut plates, approximately
0.5 m from the laser beam. In one of the workplaces, two lasers were operated at the
same time. In this situation, the SMPS and APS were placed between the two CO2 lasers
(Bystronic Bystar 4020 and Bystronic Byflex), approximately 1.5 m from each laser system.
The sampling time was determined based on the cutting length of the same metal sheet type
and included shifting sheets of the same type. The filter samples for gravimetric, elemental
composition, and mineral phase determination of the bulk material were collected when the
 approximately 1.5 m from each laser system. The sampling time was determined base
on the cutting length of the same metal sheet type and included shifting sheets of the sam
type. The filter samples for gravimetric, elemental composition, and mineral pha
determination of the bulk material were collected when the cutting time was longer tha
Int. J. Environ. Res. Public Health 2022, 19, 9888 3 of 13
30 min. Six parallel samples were collected for XRD analysis to provide the possibility
stacking filters on top of each other, and one sample was collected for ICP-MS analysi
This resulted
cutting in seven
time was longerparallel samples
than 30 min. for each
Six parallel metal
samples sheet
were type.for
collected Additionally,
XRD analysis to colle
particles for microscopy, Cu transmission electron microscopy (TEM)
to provide the possibility of stacking filters on top of each other, and one sample gridswas with hole
carbon
collectedfilms
for (EMresolution, Sheffield,
ICP-MS analysis. This resultedUK) wereparallel
in seven affixed to theforsurface
samples of 25
each metal mm Merc
sheet
type. Additionally, to collect particles for microscopy, Cu transmission electron
Millipore PVC membrane filters in open-faced SKC filter cassettes. Three samples we microscopy
(TEM) grids
collected with material
for each holey carbon films (EMresolution,
to minimise the risk ofSheffield,
overloadingUK) were affixedwith
the filters, to thea variab
surface of 25 mm Merck Millipore PVC membrane filters in open-faced SKC filter cassettes.
sampling time between one minute to 11 min and 30 s.
Three samples were collected for each material to minimise the risk of overloading the
filters, with a variable sampling time between one minute to 11 min and 30 s.

Figure
Figure Schemeof
1. 1.Scheme of the
the sampling
samplingequipment
equipment(total cassettes
(total with filters
cassettes with and associated
filters pumps) used
and associated pumps) use
during laser cutting. The antistatic tube connected to a scanning mobility particle sizer (SMPS) was
during laser cutting. The antistatic tube connected to a scanning mobility particle sizer (SMPS) w
attached to the inside of the protective housing.
attached to the inside of the protective housing.
Information on the sheet type and the laser applied during cutting on the different
Information
sheets is presentedon
in the sheet
Table type
1. TEM and
grids thecollected
were laser applied during
as described and cutting on the
particle size and differe
sheets is presented
number in Table
concentrations were 1. TEM grids
measured for allwere collected
material types ifasnot
described and particle size an
stated otherwise.
number concentrations were measured for all material types if not stated otherwise.
Table 1. Metal sheet information and laser cutting characteristics.

Table 1.Material
Metal sheet information and laser cutting characteristics.
Cutting Characteristics Collection
Thickness
Type Sheet Type Additional Information Laser Type Time Cutting
(min) Site
(mm) Material Collection
Characteristics
Aluminium (Al) 0.5 5754 H34 CO 2 4 B
Thickness Laser Time
Type a (Al)
Aluminium 10 Sheet Type
EN AW 1050A Additional Information CO2 187 C Site
(mm) Type (min)
Black steel a (BKS) 2.3 GGB DU PTFE and lead coating CO2 100 C
Aluminium (Al) 0.5 5754 H34 CO2 4 B
Black steel (BKS) 3.2 Hardox 450 Hot rolled steel Fibre 54 A
Aluminium a (Al) 10 EN AW 1050A CO2 187 C
Galvanized steel (GS) 1.5 DX51D+Z275 CO2 229 B
Black steel a (BKS) 2.3 GGB DU PTFE and lead coating CO2 100 C
Cold-rolled steel covered
BlackMild steel
steel (MS)
(BKS) 2 3.2 DC01AM Hardox 450 with carbon-based Hotoilrolled steelCO2 12
Fibre 54 B A
Galvanized steel
Mild steel (GS)
(MS) 3 1.5 DX51D+Z275
S355MC CO2 CO782 229 B B
Stainless steel (SS) 3 AiSi 316 Cold-rolled steel covered
CO2
Mild steel (MS) 2 DC01AM CO9 2 12 B
B
Stainless steel (SS) 5 AiSi 304 with carbon-based Fibreoil 91 A
Mild steelsteel
Stainless (MS) (SS) 6 3 AiSi 304S355MC Fibre CO2002 78 A B
Stainless steel (SS)
Stainless steel (SS) 15 3 AiSi 316AiSi 316 Fibre CO 422 9 A B
Stainless steel(Ti)
Titanium (SS) 4 5 AiSi 304 CO2 Fibre
4 91 B A
Stainless steel (SS) 6 AiSi 304
a SMPS was not applicable due to restrictions from the factory.
Fibre 200 A
Stainless steel (SS) 15 AiSi 316 Fibre 42 A
Int. J. Environ. Res. Public Health 2022, 19, 9888 4 of 13

An aerodynamic particle sizer (APS) instrument model 3321 (TSI Inc., Shoreview, MN,
USA) was placed within two metres of the laser cutter. This allowed us to measure the con-
centration and size distribution of particles with an aerodynamic diameter of 0.542–19.8 µm
in the workroom air in the different facilities during the laser cutting operations.

2.2. Personal Sampling

Personal samples were collected in the breathing zone of five laser operators during
their work shifts, which lasted between 265 and 483 min, respectively. The respirable
cyclones (JS Holdings, Stevenage, UK) were loaded with 37 mm FPVC filters with a pore
size of 5.0 µm (Merck Millipore, Burlington, MA, USA). Battery driven SG5200 pumps
(GSA Messgerätebau, Ratingen, Germany) assured an air flow on 2.2 L/min. The workers
operated the laser cutter during the whole working day. This included, but was not limited
to, computer numerical control (CNC) work, overseeing the process and taking care of
error messages that might appear, sometimes assisting in changing the metal plates, and
maintenance such as changing laser lenses, mirrors, and nozzles on CO2 -lasers. The workers
handled several metal plates during the working day, related to what was planned for
that specific day. Two workers wore sampling equipment at sites A and B, and one did
so at site C (Table 1). All subjects gave their informed consent for inclusion before they
participated in the study. The study protocol was approved by the Norwegian Agency for
Shared Services in Education and Research.

2.3. Scanning Electron Microscopy

The particles collected onto the TEM grids were investigated with a Hitachi SU6600
field emission SEM (Hitachi High-Tech, Tokyo, Japan) equipped with a Bruker EDX detector
(Bruker Nano GmbH, Berlin, Germany) and a NORDIF electron backscatter diffraction
(EBSD) detector (NORDIF, Trondheim, Norway). The particle morphologies were observed
and imaged in secondary electron (SE) imaging mode at a working distance of 10 mm and
an accelerating voltage of 15 keV. The element composition of the agglomerates/aggregates
was determined by SEM-EDX point and area analysis for a selected number of particles
per sample. The crystalline phase was determined with the use of the EBSD detector for
50–100 particles per sample according to the method described in Ervik et al. [20].

2.4. Gravimetric Analysis

The mass of the PM was determined gravimetrically with a Sartorius micro model MC5
balance (Sartorius AG, Göttingen, Germany). The filters were conditioned for a minimum of
two days in a room dedicated to low filter mass measurements (relative humidity 40 ± 2%,
temperature 20 ± 1 ◦ C), and discharged by a 210 Po source prior to weighing. The limit of
detection (LOD), calculated as three times the standard deviation of six field filter blanks,
was 0.005 mg.

2.5. X-ray Diffraction

Three parallel PVC filters were removed from the monitor cassettes and stacked in
the XRD sample holder on a Si crystal zero-diffraction plate. The samples were analysed
in a PANalytical X’Pert3 powder diffractometer, equipped with a PANalytical Empyrean
X-ray tube (Malvern Panalytical, Malvern, Great Britain). In-house reference materials of
magnetite and hematite were also analysed. The resulting diffractograms were examined
in HighScore Plus software (Malvern Panalytical, Malvern, UK) which determined the
background. The search-and-match function of the software was used to identify mineral
candidates in the ICSD database. Element information from SEM and ICP-MS was used in
the restriction settings to limit the number of candidates.
Int. J. Environ. Res. Public Health 2022, 19, 9888 5 of 13

2.6. Inductively Coupled Plasma Mass Spectrometry

All PVC filters, including those with PM and field blanks, were transferred into
separate vessels and added to 2 mL aqua regia (1:3, 65% HNO3 , 37% HCl), 0.05 mL
40% HF (A.R. grade Sigma-Aldrich, Merck KGaA, Darmstadt, Germany), and 200 µL
internal standard (2 µg/mL Rhodium). The samples were digested using a Milestone
mls 1200 digestion module (Milestone, Fatebenefratelli, Italy) and diluted to 15 mL with
MilliQ water.
The quantification of Al, Cr, Cu, Fe, Mn, Ni, Pb, and Zn was performed with an Agilent
8800 QQQ ICP-MS (Agilent Technologies, Santa Clara, CA, USA). Multi-element calibration
solutions (Spectrapure Standards AS, Oslo, Norway) were prepared with acid-matched
matrix solutions. Mild steel welding fume (MSWF-1) reference material (HSE’s Science and
Research Centre, Derbyshire, UK) was used for quality control (Table 2).

Table 2. Recovery (%) of mild steel welding fume reference material.

Element Recovery (%)

MSWF-1 Mild steel welding fume (n = 3)
Fe 99 ± 5.3
Mn 101 ± 5.7
Zn 101 ± 4.6

2.7. Statistics
Geometric mean, median, minimum, and maximum values are presented for PM dust
level and the elemental concentration in the workers’ breathing zone because the variables
are lognormal distributed. These are calculated using R studio version 4.1.2 (R core Team,
Vienna, Austria) [21], and Figures 2 and 5 are plotted with the additional CRAN packages
‘ggplot20 [22],‘ggpubr’ [23], and ‘ggsci’ [24]. The limit of detections was calculated as
three times the standard deviation of six blank samples.

3. Results
3.1. Particle Number Concentration and Size Distribution
Mean particle concentration decreased with the increasing thickness of the metal sheet
considering similar material types cut with a similar laser. This is illustrated for the SS
sheets cut with a fibre laser (Figure 2C). When different material types and/or different
lasers are considered, there are no obvious trends. The laser cutting of the 4 mm Ti sheet
with a CO2 laser resulted in the highest mean particle concentration. Generally, the size
distribution of the particles originating from cutting with the fibre laser of SS sheets was
smaller than particles emitted during laser cutting with a CO2 laser.
blic HealthInt.
2022, 19, x Res. Public Health 2022, 19, 9888
J. Environ. 6 of 13 6 of 13

Figure 2. Particle sizeFigure

distribution forsize
2. Particle condensation
distribution forparticles fromparticles
condensation laser cutting
from laserof cutting
(A) aluminium-
of (A) aluminium-
(Al), titanium- (Ti), galvanized
(Al), titanium-steel
(Ti), (GS), andsteel
galvanized stainless steel
(GS), and (SS) steel
stainless sheets
(SS)cut with
sheets a CO
cut with 2 laser;
a CO 2 laser;(B)
(B) black
black steel (BKS) andsteelmild steel
(BKS) and(MS)
mild sheets
steel (MS)cutsheets
withcutfibre
with and CO2CO
fibre and laser,
2 respectively,
laser, respectively, and
and (C)(C)
stainless
stainless steel (SS) sheets
steelcut
(SS)with
sheetsacut
fibre
withlaser.
a fibre laser.

3.2. Particle Morphology

The particles collected during the laser cutting of the various materials consisted of
agglomerates of spherical primary particles (Figure 3A) and agglomerates/aggregate of
coalescing primary particles (Figure 3B,C). A magnification of the aggregation is shown
in Figure 3C. Most particles were in the size range of 5–50 nm; however, many of the
 Int. J. Environ. Res. Public Health 2022, 19, 9888 7 of 13

3.2. Particle Morphology

The particles collected during the laser cutting of the various materials consisted of
agglomerates of spherical primary particles (Figure 3A) and agglomerates/aggregate of
coalescing primary particles (Figure 3B,C). A magnification of the aggregation is shown in
Figure
Int. J. Environ. Res. Public Health 2022, 19, x 3C. Most particles were in the size range of 5–50 nm; however, many of the samples
7 of 13
also included some larger condensation particles with sizes ranging from 100–200 nm
(Figure 3D).

Figure 3. Secondary electron images of: (A) agglomerate of spherical primary particles, (B)
Figure 3. Secondary electron images of: (A) agglomerate of spherical primary particles, (B) agglomer-
agglomerate/aggregate of primary particles, (C) magnification of an aggregate, and (D) larger
ate/aggregate of primary particles, (C) magnification of an aggregate, and (D) larger condensation
condensation particles.
particles.

3.3.
3.3.Chemical
ChemicalComposition
Composition of CondensationParticles
of Condensation Particles
The
Theamount (mg/m33) )of
amount (mg/m of PM
PM was highest
highest for
forthe
thesample
samplecollected
collected during
during thethe laser
laser
cutting
cutting of GS sheets
of GS sheets(Table
(Table
3). 3).
As As for particle
for the the particle
number number concentration,
concentration, the PM
the PM mass mass
seems
to decrease
seems with increasing
to decrease sheet thickness.
with increasing One exemption
sheet thickness. is the sample
One exemption collected
is the sampleduring
collected
the COthe
during 2 laser
CO2cutting of the 3ofmm
laser cutting theMS
3 mmwhich
MS yielded the lowest
which yielded themass
lowestconcentration.
mass concentration.
The XRD pattern for GS 1.5 mm, SS 6 mm and a magnetite in-house reference material,
is shown in Figure
Table 3. Geometric 4. Aconcentration
mean good match(mg/mwas found between matter
3) of particulate magnetite
fromand theparallel
seven samples in
samples
terms
with of peak and
minimum position and the
maximum in relative
brackets.intensity of the diffraction
Concentrations (µg/m ) ofpeaks.
3 elementsFrom the XRD
in one of these
pattern,ofitparticulate
samples is shown matter
that magnetite was found
freshly emitted duringinthe
both GS (red
cutting line) and
of various SS aluminium
sheets: (green line)(Al),
samples,
black steel whereas α-Fe is only
(BKS), galvanized steelseen instainless
(GS), the SS 6 steel
mm sample.
(SS), andAs indicated
mild in Table
steel (MS). 3, themineral
The main GS
phase waswas
sample determined
the only by SEM-EBSD,
sample analysedand/or
withXRD,
XRDfor threeα-Fe
where to four
was samples.
not found.

GS 1.5 mm MS 3 mm BKS 3.2 mm SS 5 mm SS 6 mm Al 10 mm SS15 mm

Particulate 0.1 [0.09, 0.4 ** [0.3,
1.5 [1.2, 1.9] 1.0 [0.7, 1.9] 0.8 [0.5, 1.2] 0.4 * [0.4, 0.5] 0.3 [0.2, 0.3]
matter a 0.13] 0.4]
Al 2.4 2.1 5.5 5.7 5.8 40 4.2
 Int. J. Environ. Res. Public Health 2022, 19, 9888 8 of 13

The XRD pattern for GS 1.5 mm, SS 6 mm, and magnetite reference material is shown
in Figure 4.
J. Environ. Res. Public Health 2022, 19, x 8 of
Table 3. Geometric mean concentration (mg/m3 ) of particulate matter from seven parallel samples
with minimum and maximum in brackets. Concentrations (µg/m3 ) of elements in one of these
samples of particulate matter freshly emitted during the cutting of various sheets: aluminium (Al),
black steel (BKS), galvanized steel (GS), stainless steel (SS), and mild steel (MS). The main mineral
Pb 0.008 0.002 0.1 0.03 0.02 11 0.03
phase was determined by SEM-EBSD, and/or XRD, for three to four samples.
Zn <LOD <LOD <LOD <LOD <LOD 4.2 <LOD
GS 1.5 mm MS 3 mm
BKS b,c
3.2 mm SS 5 mm b,c SS 6 mm Alb,c10 mm SS15 mm
Magnetite and Magnetite and Magnetite and
ineral phase Magnetite
Particulate
1.5
b,c Magnetite c
[1.2, 1.9] 0.1 [0.09, 0.13] α-Fe
1.0 [0.7,
b 1.9] 0.8 [0.5, 1.2] b 0.4 * [0.4, 0.5] α-Fe
0.3b[0.2, 0.3]
Hercynite c Magnetit
0.4 ** [0.3, 0.4]
matter a α-Fe
Al 2.4 a mg/m3, b XRD,
2.1 c SEM-EBSD,
5.5 * n = 5, **5.7
n = 6. 5.8 40 4.2
Cr 0.3 0.04 1.9 1.2 0.5 0.3 0.5
Cu 1.2 The XRD
0.05 pattern4.3for GS 1.5 0.6 mm, SS 6 mm 0.3 and a magnetite
20 in-house
0.4 referen
a 0.7 material, is0.05shown in 0.3
Fe Figure 4. A 0.4good match0.2was found0.04between magnetite 0.1 and t
Mn 3.3 samples in1.1 9.9 5.6 2.6 1.0
terms of peak position and the relative intensity of the diffraction peaks. Fro 2.1
Ni 0.2 0.2 1.7 0.7
the XRD pattern, it is shown that magnetite was0.4found in both0.2 0.5
GS (red line) and SS (gre
Pb 0.008 0.002 0.1 0.03 0.02 11 0.03
line) samples, whereas α-Fe is only seen in the SS 6 mm sample. As indicated in Table
Zn <LOD <LOD <LOD <LOD <LOD 4.2 <LOD
the GS sample was the only b,c
sample analysed b,c
with XRDb,c
where α-Fe was not found.
c Magnetite Magnetite Magnetite
Magnetite b,c The XRD pattern Hercynite c Magnetite c
andfor
α-FeGS 1.5 mm, SSb 6 mm,
andand
α-Fe bmagnetite reference material is show
Mineral phase Magnetite b and α-Fe
in Figure 4. b XRD, c SEM-EBSD, * n = 5, ** n = 6.
a mg/m3,

Figure 4. X-ray
Figure diffraction
4. X-ray diffractionpatterns formagnetite
patterns for magnetite reference
reference material
material and PMand PM collected
collected from cutt
from cutting
on GSon1.5
GSmm andand
1.5 mm SSSS6 mm.
6 mm.

3.4. Exposure Measurements

3.4. Exposure Measurements
The exposure measurements indicate that the general exposure levels during laser
The exposure
cutting measurements
are low and indicate
below the current thatOELs
Norwegian the general exposurerespectively
for all components, levels during la
(Table 4). The variation between workers can be explained by
cutting are low and below the current Norwegian OELs for all components,the fact that the samples
respectiv
were collected in different factories, each one cutting metal sheets for their own specific
(Table 4). The variation between workers can be explained by the fact that the samp
application. The open laser cutting of Pb containing BS pipes/sheets was performed by
were collected in different factories, each one cutting metal sheets for their own spec
application. The open laser cutting of Pb containing BS pipes/sheets was performed
one of the workers, which resulted in a much higher air-Pb concentration measured in t
breathing zone compared to the other workers.
 Int. J. Environ. Res. Public Health 2022, 19, 9888 9 of 13

one of the workers, which resulted in a much higher air-Pb concentration measured in the
breathing zone compared to the other workers.

Table 4. Particulate matter and elemental concentrations [µg/m3 ] measured in the breathing zone
of workers operating laser cutters, n = 5, and the current Norwegian occupational exposure limit
Int. J. Environ. Res. Public Health 2022, 19, x 9 of 13
(eight-hour).

Median Minimum Maximum Norwegian OEL (2022)

Particulate
Ti 52 0.1 29 0.01 115 2 5000 a 5000 e
matter
Fe Mn 5 0.08 2 0.01 14 0.3 3000 b 50 a
Al Cr 2 0.07 1 <LOD 2 0.08 5000 c 500 b
Zn 1 0.3 4 5000 d
Cu 0.03 0.02 24 100 c
Ti 0.1 0.01 2 5000 e
Mn Pb 0.08 0.01 0.01 0.002 0.3 12 50 a 50 c
Cr Sn 0.07 0.002 <LOD <LOD 0.08 4 500 b 2000 b
Cu 0.03 0.02 100 c
a respirable dust, b not particle size dependent, c welding 24
fume/smoke, d respirable as oxide, e as
Pb 0.01 0.002 12 50 c
dioxide. Sn 0.002 <LOD 4 2000 b
a respirable dust, b not particle size dependent, c welding fume/smoke, d respirable as oxide, e as dioxide.
The particle number concentration in the work environment during laser cutting
with two Thelaser cutters
particle numberoperating at the same
concentration in thetime,
workone with and one
environment without
during a cabinet, is
laser cutting
presented in Figure 5. The laser cutter with a cabinet was cutting
with two laser cutters operating at the same time, one with and one without a cabinet, 10 mm Al sheets
continuously
is presented between
in Figure 09:20
5. Theand 12:27.
laser cutterThe laser
with cutterwas
a cabinet without a cabinet
cutting 10 mm Al was operated
sheets
manually at intervals,
continuously betweenfor two
09:20 andminutes around
12:27. The laser 09:30, from 10:20
cutter without to 11:00
a cabinet wasand from 11:30
operated
to manually at intervals,
12:30 (start-time for two
marked in minutes around
red, Figure 5). 09:30, from tubes
The steel 10:20 to
and11:00 and from
sheets were11:30
manually
to 12:30 (start-time marked in red, Figure 5). The steel
handled between the laser cutting. The peaks of the higher concentrations tubes and sheets were manually
of particles
handled
larger thanbetween
1 µm daethe laser cutting.
occurred The peaks
frequently relatedof to
thethe
higher
laserconcentrations
cutting. However,of particles
the highest
larger than 1 µm dae occurred frequently related to the laser cutting. However, the highest
concentration of particles was measured before the manually handled laser cutting
concentration of particles was measured before the manually handled laser cutting started,
started,
shortlyshortly after
after 10:00. 10:00.this
During During
time, thethis time,
laser the laser
operator operator
collected collected
and prepared and
metal prepared
sheets
metal sheets for laser cutting. The higher concentration of particles, including
for laser cutting. The higher concentration of particles, including particles larger than 1 µm particles
larger than 1 µm
dae , indicates thedresuspension
ae, indicates the resuspension
of settled dust. of settled dust.

Figure 5. Particle size and number concentrations measured with APS during laser cutting. The red
Figure 5. Particle size and number concentrations measured with APS during laser cutting. The red
line indicates the start of cutting with the open laser and the black line indicates the stop. The closed
line indicates the start of cutting with the open laser and the black line indicates the stop. The closed
laser cutter was operating in the background throughout the entire time.
laser cutter was operating in the background throughout the entire time.
The size distribution of particles with a dae between 0.549–19.8 µm was similar in
theThe size distribution
workroom of particles
air of the three differentwith a dae between
factories, 0.549–19.8
with the highest µm was similar
concentration aroundin the
workroom air of
dae 0.9 µm [25]. the three different factories, with the highest concentration around dae
0.9 µm [25].

4. Discussion
The particle number concentration of the condensation particles emitted during the
Int. J. Environ. Res. Public Health 2022, 19, 9888 10 of 13

4. Discussion
The particle number concentration of the condensation particles emitted during the
cutting of metal sheets seems to be dependent on the thickness of the sheets when consid-
ering similar material types and similar laser types and/or laser beam energy (Figure 2).
This result is consistent with what was reported by Pena et al. [15]. They explained this
finding by suggesting that the excess energy from the laser cutting results in vaporisation
rather than fusion. More energy is concentrated in less material in a thinner sheet, which
leads to the high generation of aerosol. A similar trend is seen in the PM mass concentra-
tion (Table 3). However, it is not only the sheet thickness that influences the generation
of particles. The heat-affected zone will also be influenced by other factors such as the
composition of the metal sheets, cutting speed, and laser power [2]. Slower cutting may
lead to more absorbed energy, when considering the same material, and may increase the
width of the cut. On the other hand, cutting at high speed might result in an insufficient
melting of bottom regions during the cutting and increase the risk of spatter from melted
material. The speed and power must thus be adjusted according to the physical proper-
ties of the material, such as thermal conductivity and reflectivity. These properties affect
the cutting process and thereby the generation of particles. A material with low thermal
conductivity can be cut at a higher speed than materials with high thermal conductivity
since the energy is concentrated in the cutting zone instead of being dissipated into the
material [3]. In the present study, laser cutting into a 4 mm Ti sheet resulted in the highest
number of condensation particles. This may be a result of the low thermal conductivity of
Ti, which, as described in the thickness study by Pena et al. [15], will allow more energy to
be concentrated in less material. In contrast, Al has a very high thermal conductivity and is
reflective. However, the Al sheet was only 0.5 mm, which may be the reason why cutting
this sheet resulted in the second highest particle number concentration. In general, the
particle number concentration depends on the material properties, sheet thickness, laser
power, cutting speed, and assistant gas.
Processes affecting particle size distribution are particle growth, agglomeration, source,
and removal processes [8,26]. The smallest particle size distribution was observed for
condensation particles from fibre laser cutting into 5 mm, 6 mm, and 15 mm thick SS sheets,
yielding a maximum dmob peak around of 40 nm. The size distribution of the particles
collected during laser cutting with a CO2 laser in 3 mm SS sheets is higher, with the
maximum peak around a dmob of 200 nm. This may indicate that the laser type influences
the size distribution of the condensation particles. The fibre laser is known to achieve
a higher cutting speed compared to the CO2 laser, but a study by [27] illustrated that
this only applied to sheet thicknesses below 6 mm when they tested SS material up to
a thickness of 10 mm. Additionally, condensation particles from the fibre laser cutting of
3.2 mm BKS resulted in a larger particle size distribution than for the 5 mm, 6 mm, and
15 mm SS. This indicates that other parameters, like chemical composition, also influence
the particle size distribution.
The collected particles show typical morphologies associated with thermal-generated
condensation particles. The agglomerates/aggerates were often a mixture of agglomerates
of loosely bound spherical primary particles (Figure 3A) and/or sintered or coalesce
primary particles forming aggregates (Figure 3B,C), which is similar to the findings in
Noskov et al. [7]. The size and shape of particles will depend on the relative rates of
particle collision and coalescence. When coalescence is fast, spherical particles are formed,
whereas if the collision rate dominates and coalescence is slow, fractal-like structures are
formed [28]. These mechanisms will be determined by temperature and residence time.
For laser cutting on steel sheets, the size of the primary particles may also be influenced
by Fe vapour concentration. It has been suggested, by modelling fume formation in arc
welding, that higher Fe concentrations result in larger primary particles while lower initial
Fe concentrations result in smaller primary particles of Fe [29]. This also has consequences
for the crystalline phases of particles as explained further below.
Int. J. Environ. Res. Public Health 2022, 19, 9888 11 of 13

With SEM-EBSD, it was found that magnetite dominated the mineral phase for parti-
cles collected during the laser cutting of steel sheets. Magnetite was confirmed with XRD
on bulk filter samples from SS 5 mm, SS 6 mm, BKS 3.2 mm, and GS 1.5 mm. Additionally,
α-Fe was determined in the SS and BKS samples. Metallic Fe has earlier been observed
in the core of condensation particles collected during additive manufacturing [7]. When
particles are exposed to air at room temperature, an oxide shell will instantly be formed [30].
According to the model by Sanibondi et al. [29], nanoparticles in arc welding are formed
mainly by FeO nucleation in the vapour phase, which oxidizes further to Fe3 O4 . Larger
primary particles are formed when the Fe concentrations are higher. The larger particles
may not be fully oxidized resulting in a core of α-Fe. Because SEM-EBSD was performed
on relatively few selected particles, and because the EBSD signal only comes from the
shell, α-Fe was not detected by EBSD. A crystalline pattern indexed as Hercynite, which
is a spinel mineral with formula FeAl2 O4 , was detected with SEM-EBSD in the sample
from cutting of 10 mm Al [31]. Both results from SEM-EDS and ICP-MS show that the
PM also contained Cu. Substitution with Cu ions has earlier been found in this spinel
phase [32]. Unfortunately, the Hercynite phase could not be confirmed with XRD as the
diffractogram was indistinct. The amount of PM on the filter influenced the quality of the
XRD diffractogram. TiO2 and Ti2 O3 were indexed with SEM-EBSD in the sample collected
from the Ti sheet cutting, but this was not confirmed with XRD as the cutting time was too
short to collect enough material.
In general, the chemical composition of the condensation particles reflected the origi-
nating metal sheet, but there are some indications of contamination from previously cut
sheets, from condensation particles formed during cutting with another simultaneous cut-
ting laser or PM mixed in from the ventilation. An example of a sintered particle originating
from a previously cut metal sheet is shown [25]. Another example of contamination is
the relatively high Pb concentration measured in PM originating from 10 mm Al sheets
(Table 3). These samples were collected at the same time as PTFE- and Pb-coated BKS
pipe/sheets were manually handled on the laser without protective housing and cut only
a few metres away. Another possibility may be that the samples were contaminated by
fume and/or sputter/sintered particles from the grid supporting the metal sheet, or the
workroom air entering via the ventilation system of the laser cutter.
A higher Mn/Cr ratio than in the originating metal sheet was found in the metal fume
from laser cutting in SS (Table 2). Originally, the relative amount of Cr was about 18–20%
in SS sheets and only 2% Mn. In the fume, on the other hand, there was about five times
more Mn than Cr. This may be due to Mn being more volatile than Cr and readily forming
oxides [8,33]. The enrichment of Mn in agglomerated ultrafine condensation particles was
also reported by Ljunggren et al. [18], but the same was not reported for samples collected
in the breathing zone of the laser additive manufacture workers in their study. In our study,
however, a higher concentration of Mn than Cr was also found in the workers’ breathing
zone. This may imply that the laser additive manufacturing workers were mainly exposed
to the original powder, but the laser cutting workers in this study were exposed to the
condensate particles formed during laser cutting.
The exposure levels to metals during laser cutting were below the respective OELs
for the workers in the three different manufacturing plants. This is an unsurprising result,
as the laser cutting was mostly enclosed in a cabinet and included point air extraction
above the laser cutter or a built-in ventilation system. Further, concentrations of PM, Cr, Fe,
Mn, Ni, and Pb were all well below what is reported in laser additive manufacturing [18].
Still, the particles generated during laser cutting are small, respirable particles which may
penetrate deep into the lungs [34,35]. The exposure during open laser cutting was measured
for one of the workers. This worker handled BKS tubes/sheets coated with PTFE and
Pb. Laser cutting into this material resulted in a potential exposure to Pb of 0.01 mg/m3
which is below the current OEL of inorganic lead in workroom air. However, prolonged
Pb exposure at these levels has been shown to be a risk for lead poisoning [36]. Measures
should thus be taken to reduce the Pb exposure when laser cutting these materials. One
Int. J. Environ. Res. Public Health 2022, 19, 9888 12 of 13

limitation is the relatively scarce number of workers included in this study. Further research
should be performed to get a better understanding of the exposure to laser operators, with
a special focus on exposure during open laser cutting. Further studies should also include
biomonitoring when handling material containing heavy metals such as Pb.

5. Conclusions
In this study, we found that the number of particles formed during laser cutting
depended on the material type, thickness of the metal sheets, and the laser cutter char-
acteristics. The size distribution may be dependent on the type of laser or the chemical
composition of originating metal sheets. The laser cutting of the different sheets resulted
in ultrafine agglomerates/aggregates of condensation particles, with the enrichment of
more volatile compounds like Mn. The assessment of the workroom air collected in the
breathing zone of laser cutting workers showed that the potential exposure to metals was
low. However, the exposure to Pb-containing material during open laser cutting may
potentially result in significant Pb exposure. Traditional occupational PM air measurements
are mass-based, which may not be sufficient in this work environment, as most particles
are in the ultrafine range. Future studies in this environment should include size to account
for the exposure of particles contributing with very little mass. Additionally, biomonitoring
of open laser operators may give a better indication of the extent of exposure.

Author Contributions: S.E.H.: conception and design, sampling, analysis and interpretation of ICP-
MS data, writing and revising manuscript. J.Ø.H.: sampling, analysis and interpretation of ICP-MS
and SEM-EDS/EBSD data, revising manuscript. P.G.: conception and design, sampling, revising
manuscript. T.K.E.: conception and design, sampling, analysis and interpretation of SEM-EDS/EBSD
and XRD data, writing and revising manuscript. All authors have read and agreed to the published
version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Acknowledgments: The authors would like to thank Kari Dahl for assisting with the XRD and
gravimetrical analysis. We also acknowledge the participating industries, with a special thank you to
the workers carrying the sampling equipment.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. He, Y.; Xie, H.; Ge, Y.; Lin, Y.; Yao, Z.; Wang, B.; Jin, M.; Liu, J.; Chen, X.; Sun, Y. Laser Cutting Technologies and Corresponding
Pollution Control Strategy. Processes 2022, 10, 732. [CrossRef]
2. Sharma, A.; Yadava, V. Experimental analysis of Nd-YAG laser cutting of sheet materials—A review. Opt. Laser Technol. 2018, 98,
264–280. [CrossRef]
3. Ivarson, A. On the Physics and Chemical Thermodynamics of Laser Cutting. Ph.D. Thesis, Luleå Tekniska Universitet, Luleå,
Sweden, 1993.
4. Riveiro, A.; Quintero, F.; Lusquiños, F.; Comesaña, R.; del Val, J.; Pou, J. The Role of the Assist Gas Nature in Laser Cutting of
Aluminum Alloys. Phys. Procedia 2011, 12, 548–554. [CrossRef]
5. Berlinger, B.; Benker, N.; Weinbruch, S.; L’Vov, B.; Ebert, M.; Koch, W.; Ellingsen, D.G.; Thomassen, Y. Physicochemical
characterisation of different welding aerosols. Anal. Bioanal. Chem. 2011, 399, 1773–1780. [CrossRef] [PubMed]
6. Graff, P.; Ståhlbom, B.; Nordenberg, E.; Graichen, A.; Johansson, P.; Karlsson, H. Evaluating Measuring Techniques for Occupa-
tional Exposure during Additive Manufacturing of Metals: A Pilot Study. J. Ind. Ecol. 2017, 21, S120–S129. [CrossRef]
7. Noskov, A.; Ervik, T.K.; Tsivilskiy, I.; Gilmutdinov, A.; Thomassen, Y. Characterization of ultrafine particles emitted during
laser-based additive manufacturing of metal parts. Sci. Rep. 2020, 10, 20989. [CrossRef]
8. Sutton, A.T.; Kriewall, C.S.; Karnati, S.; Leu, M.C.; Newkirk, J.W. Characterization of AISI 304L stainless steel powder recycled in
the laser powder-bed fusion process. Addit. Manuf. 2020, 32, 100981. [CrossRef]
Int. J. Environ. Res. Public Health 2022, 19, 9888 13 of 13

9. Azzougagh, M.N.; Keller, F.-X.; Cabrol, E.; Cici, M.; Pourchez, J. Occupational exposure during metal additive manufacturing:
A case study of laser powder bed fusion of aluminum alloy. J. Occup. Environ. Hyg. 2021, 18, 223–236. [CrossRef]
10. Ervik, T.K.; Benker, N.; Weinbruch, S.; Thomassen, Y.; Ellingsen, D.G.; Berlinger, B. Size distribution and single particle
characterization of airborne particulate matter collected in a silicon carbide plant. Environ. Sci. Processes Impacts 2019, 21,
564–574. [CrossRef]
11. Hammer, S.E.; Ervik, T.K.; Ellingsen, D.G.; Thomassen, Y.; Weinbruch, S.; Benker, N.; Berlinger, B. Particle characterisation and
bioaccessibility of manganese in particulate matter in silico-and ferromanganese smelters. Environ. Sci. Processes Impacts 2021, 23,
1488–1499. [CrossRef]
12. Chan, Y.-J.; Yuan, T.-H.; Sun, H.-C.; Lin, T.-C. Characterization and Exposure Assessment of Odor Emissions from Laser Cutting
of Plastics in the Optical Film Industry. Aerosol. Air Qual. Res. 2016, 16, 2216–2226. [CrossRef]
13. Noskov, A.; Thomassen, Y.; Berlinger, B.; Olsen, R.; Ervik, T.K.; Weinbruch, S.; Gilmutdinov, A. Characterization of air contam-
inants emitted during laser cutting of carbon fiber-reinforced composite materials. Anal. Bioanal. Chem. 2019, 411, 305–313.
[CrossRef] [PubMed]
14. Elihn, K.; Berg, P. Ultrafine Particle Characteristics in Seven Industrial Plants. Ann. Occup. Hyg. 2009, 53, 475–484. [CrossRef] [PubMed]
15. Pena, A.C.; Soler, J.G.; Caicedo, G.R. The Characterization of Aerosols Generated During the Cutting of Metallic Materials with
Lasers. Environ. Technol. 1998, 19, 83–90. [CrossRef]
16. Zubko, M.; Loskot, J.; Świec, P.; Prusik, K.; Janikowski, Z. Analysis of Stainless Steel Waste Products Generated during Laser
Cutting in Nitrogen Atmosphere. Metals 2020, 10, 1572. [CrossRef]
17. Bau, S.; Rousset, D.; Payet, R.; Keller, F.-X. Characterizing particle emissions from a direct energy deposition additive manufacturing
process and associated occupational exposure to airborne particles. J. Occup. Environ. Hyg. 2020, 17, 59–72. [CrossRef] [PubMed]
18. Ljunggren, S.A.; Karlsson, H.; Ståhlbom, B.; Krapi, B.; Fornander, L.; Karlsson, L.E.; Bergström, B.; Nordenberg, E.; Ervik, T.K.;
Graff, P. Biomonitoring of Metal Exposure During Additive Manufacturing (3D Printing). Saf. Health Work 2019, 10, 518–526.
[CrossRef] [PubMed]
19. Ljunggren, S.A.; Ward, L.J.; Graff, P.; Persson, A.; Lind, M.L.; Karlsson, H. Metal additive manufacturing and possible clinical
markers for the monitoring of exposure-related health effects. PLoS ONE 2021, 16, e0248601. [CrossRef]
20. Ervik, T.K.; Benker, N.; Weinbruch, S.; Skogstad, A.; Thomassen, Y.; Ellingsen, D.G.; Berlinger, B. Phase identification of individual
crystalline particles by combining EDX and EBSD: Application to workplace aerosols. Anal. Bioanal. Chem. 2018, 410, 2711–2721.
[CrossRef] [PubMed]
21. R Studio Team. R Studio: Integrated Development for R. Version 4.1.2; R Studio Team: Vienna, Austria, 2015.
22. Wickham, H. ggplot2: Elegant Graphics for Data Analysis, Version 3.3.2; CRAN; Springer: New York, NY, USA, 2016.
23. Kassambara, A. ‘ggplot2’ Based Publication Ready Plots. Version 0.4.0.999. CRAN. 2020. Available online: https://rpkgs.
datanovia.com/ggpubr/ (accessed on 7 August 2022).
24. Xiao, N. ggsci: Scientific Journal and Sci-Fi Themed Color Palettes for ‘ggplot2’. Version 2.9, CRAN. 2018. Available online:
https://nanx.me/ggsci/ (accessed on 7 August 2022).
25. Halvorsen, J.Ø. Exposure Assessment and Particle Characterization of Workplace Aerosols in Norwegian Metal Laser-Cutting
Industry. Master’s Thesis, Norwegian University of Life Sciences, Ås, Norway, 2021.
26. Eggersdorfer, M.L.; Pratsinis, S.E. Agglomerates and aggregates of nanoparticles made in the gas phase. Adv. Powder Technol.
2014, 25, 71–90. [CrossRef]
27. Stelzer, S.; Mahrle, A.; Wetzig, A.; Beyer, E. Experimental Investigations on Fusion Cutting Stainless Steel with Fiber and CO2
Laser Beams. Phys. Procedia 2013, 41, 399–404. [CrossRef]
28. Lehtinen, K.E.J.; Windeler, R.S.; Friedlander, S.K. Prediction of nanoparticle size and the onset of dendrite formation using the
method of characteristic times. J. Aerosol Sci. 1996, 27, 883–896. [CrossRef]
29. Sanibondi, P. Numerical investigation of the effects of iron oxidation reactions on the fume formation mechanism in arc welding.
J. Phys. D 2015, 48, 345202. [CrossRef]
30. Wang, C.; Baer, D.R.; Amonette, J.E.; Engelhard, M.H.; Antony, J.; Qiang, Y. Morphology and Electronic Structure of the Oxide
Shell on the Surface of Iron Nanoparticles. J. Am. Chem. Soc. 2009, 131, 8824–8832. [CrossRef]
31. Hill, R.J. X-ray powder diffraction profile refinement of synthetic hercynite. Am. Mineral. 1984, 69, 937–942.
32. Mistri, R.; Maiti, S.; Llorca, J.; Dominguez, M.; Mandal, T.K.; Mohanty, P.; Ray, B.C.; Gayen, A. Copper ion substituted hercynite
(Cu0.03Fe0.97Al2O4): A highly active catalyst for liquid phase oxidation of cyclohexane. Appl. Catal. A Gen. 2014, 485, 40–50. [CrossRef]
33. Simonelli, M.; Tuck, C.; Aboulkhair, N.T.; Maskery, I.; Ashcroft, I.; Wildman, R.D.; Hague, R. A Study on the Laser Spatter and the
Oxidation Reactions During Selective Laser Melting of 316L Stainless Steel, Al-Si10-Mg, and Ti-6Al-4V. Metall. Mater. Trans. A
2015, 46, 3842–3851. [CrossRef]
34. Braakhuis, H.M.; Park, M.V.D.Z.; Gosens, I.; De Jong, W.H.; Cassee, F.R. Physicochemical characteristics of nanomaterials that
affect pulmonary inflammation. Part. Fibre Toxicol. 2014, 11, 18. [CrossRef] [PubMed]
35. Bartley, D.L.; Vincent, J.H. Sampling Conventions for Estimating Ultrafine and Fine Aerosol Particle Deposition in the Human
Respiratory Tract. Ann. Occup. Hyg. 2011, 55, 696–709. [CrossRef]
36. Wu, Y.; Gu, J.-M.; Huang, Y.; Duan, Y.-Y.; Huang, R.-X.; Hu, J.-A. Dose-Response Relationship between Cumulative Occupational
Lead Exposure and the Associated Health Damages: A 20-Year Cohort Study of a Smelter in China. Int. J. Environ. Res. 2016, 13,
328. [CrossRef] [PubMed]