The main research methods taken during this experimental
study were NIOSH 0600 and 7500 methods. Following these methods,
personal respirable samples were collected and analysed to
determine the respirable dust and silica dust exposure of workers
during the process of concrete drilling.
4.1. Sampling methods
4.1.1. Personal respirable samples
Personal respirable dust samples were collected using pumps
(GilAir-3) connected to 10 mm nylon cyclones, as shown in
Fig. 2. Each set of sampling equipment was calibrated to a recommended
flow rate of 1.7 l/min using an airflow calibrator. The sampling
medium was a 37 mm, 5.0 lm pore size PVC filter, supported
with backup pad in a two-piece cassette filter holder. This 10 mm
nylon cyclone with its attachments was a light weight, size-selective
particulate collector recommended by NIOSH (1998a,b).
The duration of samples were determined by the limit of detection
(LOD) parameters of laboratory analysis and the actual operational
time of each work process/trade. From LOD perspective, the duration
of samples should be longer enough for the laboratory analysis to detect
the presence of silica, which gives the duration a lower duration
limit (LDL). The LDL could be calculated by the following equation:
LDL ¼ LOD=RELðNIOSHÞ=flow rate ¼ 118 min
ðLOD : 0:01 mg; REL ðRecommended exposure limitÞ by NIOSH
: 0:05 mg=m3; flow rate : 1:7 l=minÞ
Therefore, the duration of samples should be more than
118 min. In order to reduce the variability and narrow the confi-
dent limit, all the respirable samples during the study were taken
for approximately 180 min. During the experimental study, trials
of the block-drilling controls were conducted in two rounds consisting
of 10 trials in each round. Total 20 samples have been collected
with ten samples for each situation: with and without
DustBubble. In addition to the personal respirable samples, other
related variables were also collected and recorded, including filed
blanks, and environmental variables.
4.1.2. Bulk samples
In order to determine the presence of silica and the relative percentage
of different forms of silica (quartz, cristobalite, tridymite),
settled dust was collected as bulk samples since the bulk sample
should be representative of the airborne dust to which the workers
are being exposed NIOSH (1998a,b).
4.1.3. Field blanks
Certain numbers of blank samples are required by NIOSH 0600
and 7500 methods: 2–10 per set (NIOSH 0600, 1998; NIOSH 7500,
2003). One field blank was taken for every ten samples gathered
during the course of the project. The field blank cassettes were
stored in the same conditions as the cassettes used for air sampling.
The protocol used for acquiring field blanks consisted of:
(1) opening the cassette, (2) attaching the cassette to the cyclone
and sequentially to the sampling train, (3) allowing the cassette
Fig. 1. Four steps of using DustBubble.
Fig. 2. A worker wearing the sampling pump during the experimental study.
1286 S. Fan et al. / Safety Science 50 (2012) 1284–1289
to collect dust passively (i.e. no pump operation) during a sampling
period in the onsite sample and worker preparation area, and (4)
returning the cassette to the sealed shipment bag with the other
air samples. Therefore, the field blanks acted as quality control
against potential contamination during preparation for sampling
and shipment, as well as extended cassette exposure to ambient
field conditions. Two filed blank were taken for the twenty samples
during the study.
4.1.4. Environmental and other variables
Environmental variables, e.g. temperature, humidity, wind
direction and velocity, will affect the dust exposures. In order to
control the environmental variables, an indoor area was designed
as the test area to minimise the effects of wind. The two workers
(with or without DustBubble) started the drilling simultaneously
to make sure they were conducted in the same environmental
situations.
As a quality assurance, the environmental conditions were also
measured using a 5-in-1 weather metre (MODEL: AZ 8910).
Humidity, temperature, wind direction and velocity were recorded
at the beginning, middle and end of each sampling processes. The
results indicated that the environmental variables recorded for the
paired samples were the same, which meant that the environmental
variables need not be considered during the comparison.
During the experimental study, the workers’ comments on the
operation and effectiveness of DustBubbles were also collected to
reflect the workers’ perceived satisfaction towards the use of
DustBubble.
4.2. Laboratory analysis
Two different analysis methods were used for personal samples
and bulk samples. All analyses were conducted at the laboratory of
Hong Kong Polytechnic University. The LOD and LOQ (limit of
quantification) were 0.01 mg and 0.03 mg respectively.
4.2.1. Personal samples
All personal samples were firstly analysed for total weights
according to NIOSH 0600 method using Mettler-Toledo XP 26 balance,
which gave the concentration of total respirable dust. The
samples were then analysed by X-ray following NIOSH 7500
method.
4.2.2. Bulk samples
A quick qualitative analysis on the bulk samples using X-ray in
accordance with NIOSH 7500 method was conducted to determine
the presence of silica and the relative levels of the three silica
forms. If the presence of silica was confirmed and the major percentage
of silica is quartz, the personal respirable samples collected
during this experimental study were only analysed for
quartz.
4.3. Testing materials and tools
Significant basis of this experimental study was the fact that all
samples should be carried out under comparable conditions. In order
to obtain reproducible results to compare drilling process with
or without Dust Bubble, only a study in a test room was taken into
consideration. With measurements taken at a real-life workplace,
environmental impacts like air change rates, teat area sizes or
changes in material mixtures can seldom be eliminated suffi-
ciently. However, conditions of the experimental study should be
set up as close as possible to those in practice. Therefore, all the
materials used were kept the same as the practice.
It was estimated that it would take around 1 min for 1 hole
when Dust Bubble was used. In order to make comparison, same
number of holes should be drilled during the two drilling processes
(with or without Dust Bubble). Therefore, a total of 180 holes were
drilled for each 3-h sample. A 13 mm (in diameter) masonry drill
was used for drilling. Each hole was around 3 cm deep into the
block. The model of the hand drill was Bosch GBH 2-20 D Professional
(as shown in Fig. 3). A support as shown in Fig. 4 has been
prepared to hold the testing material. The construction material
(concrete block, C30) for treatment was placed vertically. The size
of concrete block was 1 m 1 m 12 cm. The edge distance and
interval distance between holes was 20 cm and 5 cm respectively
so that 196 holes could be drilled on one side. Therefore, a total
392 holes could be achieved on both sides, which was enough for
two samples. A number of 10 blocks were produced to achieve
20 samples.
The assumption of this experimental study was that all the
paired samples should be carried out under comparable conditions.
In order to obtain reproducible results to compare drilling process
with or without DustBubble, two samples were collected at the
same time in the test area to eliminate the environmental impacts
(e.g. wind, temperature, and material mixtures), as shown in Fig. 5.
For each set of paired samples, the work load and duration were
the same. The workers could take a 5-min break for every
20 min. Photos and video record have be taken during the experiment
study.
4.4. Statistical analysis
The main variables tested in this experimental study were total
respirable dust concentrations and respirable quartz dust concentrations.
The reduction in the two variables produced by the units
with controls (DustBubble) compared to that produced by those
without controls during concrete drilling processes were estimated
through the following equation:
Fig. 3. Electric hammer used in the study.
Fig. 4. A support for holding concrete blocks.
S. Fan et al. / Safety Science 50 (2012) 1284–1289 1287
Estimated % reduction ¼ 100x½1
ðcontrol mean=no-control meanÞ
The data are generally assumed to follow the lognormal distribution,
when samples are collected over time (NIOSH, 1977).
Therefore, all the data has been tested firstly for a lognormal distribution.
Descriptive statistics were used to describe total respirable
dust and silica dust concentrations in terms of arithmetic and geometric
means (AM and GM) as well as the corresponding geometric
standard deviations (GSD) and ranges (min–max), while inferential
statistics were conducted to compare the differences of dust generation
between the two situations.
According to Hornung and Reed (1990), for the non-detectable
values, there are two simple ways to reach enough accuracy:
1/ ffiffiffi
2
p of nondectable values when the data are not highly skewed;
1/2 of nondectable values when the data are highly skewed
(Geometric Standard Deviation approximately 3.0 or greater).
However, if more than half of one set of data are non-detectable,
only the percentage of samples below LOD and the range of
remaining samples were reported. Statistical analyses were conducted
with SPSS statistical software (version 17; SPSS Inc.)