Você está na página 1de 12

Diesel Particulate Matter Control Technologies and

Measurements in US Mines
S E Mischler1, A D Bugarski1 and G H Schnakenberg Jr1

ABSTRACT operate at exhaust temperatures typical of a large number


of heavy-duty vehicles operated in underground mining
An investigation was carried out at a US underground mine to
characterise diesel aerosols and to asses the effects of various control applications. In addition, disposable filters using high
technologies on the concentrations and properties of particulate matter temperature disposable filter elements from Donaldson
and gases emitted by heavy-duty diesel-powered equipment. This study Company, Inc and Filter Systems, both widely used by the coal
was dedicated to evaluating the impact on mine air quality of mining industry, are also considered by some metal mining
contemporary diesel particulate filter (DPF) systems installed on mining companies to be viable DPM controls for underground mining
equipment. Additional efforts were made to assess the effect of blended vehicles.
biodiesel fuels, yellow grease and soy, ultra-low sulfur fuel, a fuel Other DPM control strategies of interest are water fuel
emulsion and of selected diesel oxidation catalysts on air quality and
emulsions, biodiesel blends, and ultra-low sulfur diesel fuel.
emissions. The carbon analysis performed on the collected samples
showed that the tested DPF systems reduced the mass concentrations of These alternative fuels are currently used by some metal and
elemental carbon in mine air by between 70 per cent and 92 per cent. In non-metal underground mines. Preliminary surveys showed
addition, TEOM measurements, collected during these tests, showed that promising results but effectiveness of fuel emulsions in
the total diesel particulate matter concentration in the mine air was controlling DPM emissions from underground mining vehicles
reduced by greater than 65 per cent when DPF systems were used. The was not accurately quantified.
tested biodiesel blends were found to reduce elemental carbon The study reported here was organised to evaluate the in situ
concentrations in the mine air by between 49 per cent and 33 per cent, for effectiveness of the aforementioned technologies for controlling
the B20 (20 per cent biodiesel/80 per cent #2 diesel) of soy and yellow
grease respectively and between 66 per cent and 56 per cent for the B50
DPM and gaseous emissions from underground diesel-powered
(50 per cent biodiesel/50 per cent #2 diesel) of soy and yellow grease, mining equipment.
respectively. The water-in-diesel fuel emulsion was shown to reduce
elemental carbon concentrations by as much as 85 per cent. NO2 MATERIALS AND METHODS
measurements showed an increase in the maximum concentrations of
NO2 when catalysed DPF systems were used instead of mufflers. The control technologies evaluated in this study are presented in
Table 1. The technologies were tested in an underground mine
INTRODUCTION using mining vehicles that were operated under conditions that
closely resembled actual production scenarios.
In January 2001 the US Mine Safety and Health Administration The effects of the selected control technologies on
(MSHA) promulgated rule 30 CFR 57.5060 limiting exposures concentrations of DPM and gases in the mine air were assessed
of underground metal and non-metal miners to diesel particulate by conducting two groups of tests in the isolated zone of active
matter (DPM). As a result of this rule, the United States underground metal mine (see Table 1). The first group of tests
underground mining community is currently working on was used to evaluate the effects of various fuel formulations,
identifying technically and economically feasible controls for the namely the water-fuel emulsions, blended biodiesel fuels,
curtailment of DPM and gaseous emissions from existing and ultra-low sulfur diesel fuel, and #1 diesel. The second group of
new diesel powered vehicles in underground mines. To date, the tests evaluated diesel exhausts filter systems, namely the two
replacement of older diesel engines with cleaner modern engines, DPF systems supplied by ArvinMeritor and Clean Air
improvements in mine ventilation, diesel engine maintenance, Power/ETG and high temperature disposable filter elements
and the implementation of various diesel emission control (DFE) from Donaldson and Filter Services.
technologies, including diesel particulate filters (DPFs) and
reformulated fuels, are viewed by the industry as the most viable
Vehicles and engines
methods to reduce DPM concentrations in mine air.
Previous laboratory evaluations (Mayer et al, 1999; Larsen et Two load-haul-dump (LHD) vehicles, designated as LHD1 and
al, 1999) and underground mine tests (Watts et al, 1995; McGinn LHD2, served as test platforms for this study. These LHDs were
et al, 2002; Bugarski and Schnakenberg, 2001; Bugarski and selected because they are typical of a mine production fleet.
Schnakenberg, 2002; Bugarski et al, 2005) have shown that These LHDs are classified as heavy-duty production machines,
current DPF technologies are efficient in reducing DPM and their engines are routinely heavily loaded in the course of
emissions. One potential problem with using catalysed DPF their normal duty cycle.
systems in underground operations is an undesired increase in The basic specifications for the vehicles and engines are given
the ambient concentration of nitrogen dioxide (NO2) (Bugarski et in Table 2.
al, 2005). This increase was found to be related to the processes
promoted by a platinum-based catalyst introduced into these Preparation of the vehicle for the study
systems to support the regeneration of the DPF at relatively low
exhaust temperatures. The Caterpillar 3126B DITA AA engine in LHD1 was equipped
with its original muffler for eight of the ten tests on fuel
To avoid the potential NO2 problem, a DPF system with
formulations. For two of the fuel tests the engine was fitted with
a diesel fuel burner, developed by ArvinMeritor, and a catalytic
a DCL Mine-X Catalytic Converter, Model #3206-MD. The
particle oxidiser (CPO) from Clean Air Power attracted
Caterpillar 3126B DITA AA engine has an open crankcase
the attention of mine operators. Those systems are designed to breather that was not filtered in this study.
The modifications to LHD2 for the filter tests were related
1. National Institute for Occupational Safety and Health, Pittsburgh to the temporary installation of various exhaust system
Research Laboratory, PO Box 18070, Pittsburgh PA 15236, USA. configurations. The ArvinMeritor (AM) DPF system had been

Eighth International Mine Ventilation Congress Brisbane, QLD, 6 - 8 July 2005 263
S E MISCHLER, A D BUGARSKI and G H SCHNAKENBERG Jr

TABLE 1
Control technologies evaluated in this study.

Test group (test vehicle) Exhaust system Fuel formulation


Fuel formulations Muffler #1 diesel
(LHD1 powered by Muffler PuriNOx cold-weather water-fuel emulsion
Caterpillar 3126B
DITA AA) Muffler PuriNOx warm-weather water-fuel emulsion
Muffler 20% soy biodiesel and 80% #1 diesel blend
Muffler 50% soy biodiesel and 50% #1 diesel blend
DOC (DCL International) 50% soy biodiesel and 50% #1 diesel blend
Muffler 20% yellow grease biodiesel and 80% #1 diesel blend
Muffler 50% yellow grease biodiesel and 50% #1 diesel blend
Muffler ULS (10 ppm sulfur) diesel
DOC (DCL International) ULS (10 ppm sulfur) diesel
Filtration systems Muffler #1 diesel
(LHD2 powered by Deutz ArvinMeritor fuel burner DPF system with platinum (Pt) DOC #1 diesel
BF4M1013 FC/MVS)
ArvinMeritor fuel burner DPF system with palladium (Pd) DOC #1 diesel
Donaldson disposable filtration element P604516 #1 diesel
Filter services disposable filtration element #1 diesel
Clean air power/ETG catalytic particle oxidiser (CPO) DPF system #1 diesel

TABLE 2
Test vehicles and engines.

Vehicle (test group) Vehicle Vehicle Vehicle Engine Engine model Engine Engine Engine type
type make model make displ rating
LHD1 (fuels) Load haul Wagner ST-3.5 Caterpillar 3126B DITA AA 7.243 l 200 hp Fully electronic controlled, turbo
dump charged, air to air after cooled
LHD2 (filters) Load haul Wagner ST-2D Deutz BF4M 1013 FC/MVS 4.764 l 151 hp Fully electronic controlled, turbo
dump charged, air to air after cooled

installed on vehicle LHD2 several months before study. The designated tank. Samples of each soy fuel blend were collected
other filter systems tested were fitted to vehicle LHD2 as from the designated fuel tank and sent out for analysis. Results of
temporary replacements for the AM DPF system specifically for the analysis are presented in Table 3. The producer of the soy
the purpose of conducting these tests. A muffler was fitted in biodiesel, reported the following properties for the neat soy
place of the AM DPF for the test which was conducted to biodiesel fuel: free glycerine 0.00135 per cent, total glycerine
establish a baseline case. The Deutz BF4M1013 FC/MVS engine 0.1995 per cent, flashpoint 153C, sulfur (by UV florescence)
used in LHD2 is designed with a closed loop crankcase breather 0.00014 per cent by mass, kinematic viscosity 4.20 mm2/s,
system which eliminated spurious DPM emissions. Cetane number 52.7.
The neat yellow grease biodiesel was blended with #1 diesel at
Tested fuel formulations the site using an electronic volumetric flow metre. Each fuel
blend was stored in a designated tank. The properties of the neat
As presented in Table 1 eight different fuel formulations were biodiesel reported on the certificate of analysis are following:
evaluated in this study. The LHD LHD1 powered by a Caterpillar free glycerine 0.01 per cent, total glycerine 0.096 per cent,
3126B DITA AA engine was the vehicle used for these tests. In flashpoint >130C, sulfur 0.004 per cent by mass, kinematic
order to control experimental variation, all the #1 diesel fuel used viscosity 4.81 mm2/s at 40C, Cetane number 53.9.
in this study was obtained from the same supplier and the same
process batch. The ultra-low sulfur (ULS) diesel fuel was received from a
Canadian supplier. After the test, a sample of the ULS diesel was
Two formulations of water-in-diesel fuel emulsions (PuriNOx)
collected from the auxiliary fuel tank and sent to a laboratory for
were tested in this study. The warm weather formulation contains
analysis. The results of the analysis on ultra-low sulfur diesel
approximately 77 per cent #2 diesel fuel, 20 per cent water, and
fuel are reported in Table 3.
three per cent proprietary emulsifying agent. This formulation is
suitable for fuelling vehicles when there is no risk of freezing.
The cold weather formulation contains approximately 86 per Tested aftertreatment systems
cent #2 diesel fuel, ten per cent water, two per cent methanol and The aftertreatment systems described in Table 4 were tested
two per cent of the proprietary emulsifying agent. It is using a LHD2 powered by a Deutz BF4M1013 FC/MVS engine.
formulated for colder climates. The major properties of cold- and
warm-weather formulations are summarised in Table 3.
ArvinMeritor (AM) DPF system
Neat soy biodiesel was tested as a 20 per cent and 50 per cent
blend with #1 diesel fuel. The neat biodiesel was blended in the The DPF system from ArvinMeritor (AM) uses an integrated fuel
desired volumetric fractions with #1 diesel at the site, using an burner to perform automatic regeneration of the filter element.
electronic volumetric flow metre. Each fuel blend was stored in a The computer controlled diesel fuel burner, integrated on the

264 Brisbane, QLD, 6 - 8 July 2005 Eighth International Mine Ventilation Congress
DIESEL PARTICULATE MATTER CONTROL TECHNOLOGIES AND MEASUREMENTS IN US MINES

TABLE 3
Fuel formulation properties.

Fuel formulations PuriNOx PuriNOx 20% soy 50% soy ULS fuel
Properties Method Units cold-weather warm-weather biodiesel biodiesel

Aromatics ASTM D1319 vol % 22.4 23.7 - - 26.4


Olefins ASTM D1319 vol % 2.3 2.4 - - 1.2
Saturates ASTM D1319 vol % 75.3 73.9 - - 72.4
Density @ 16C ASTM D4052 g/ml 0.853 0.866 0.836 0.854 0.850
Sulfur content ASTM D2622 ppm 300 279 205 129 4
Oxygen By diff % wt 7.8 15.3 4.4 7.4 1.3
Heat of combustion ASTM D240 BTU/lb 17 003 15 905 18 075 17 553 18 433
Flash point ASTM D93 C 47 - 68 70 64

TABLE 4
Aftertreatment systems.

Filtration system Media type Catalyst DOC


ArvinMeritor (AM) with Pt DOC Ceramic, cordierite N/A Metal substrate platinum based catalyst
ArvinMeritor (AM) with Pd DOC Ceramic, cordierite N/A Metal substrate with
palladium based catalyst
Donaldson P604516 High temperature disposable filter element N/A N/A
Filter service High temperature disposable filter element N/A N/A
Clean air power/ETG CPO Deep bed fibre, quartz silica fibre Precious metal catalyst Ceramic monolith substrate
coated fibres with precious metal catalyst

inlet side of the DPF is used to raise the engine exhaust as 760C this filter element is finding use in some applications
temperature above 650C during the periodic regeneration of the for filtering relatively hot exhaust which is not cooled by a water
uncatalysed ceramic wallflow monolith filter element. The scrubber or other types of heat exchangers.
regeneration process can be initiated at predetermined time
intervals or can be triggered by predetermined levels of engine Filtration system with disposable filter elements
backpressure. The operation of this system is completely (DFE)s from filter services and testing
automated and transparent to the operator.
The AM DPF system was tested twice, in order to evaluate The filtration system used to evaluate Donaldson DFE was also
two different catalyst formulations for the DOC installed on the used to evaluate a prototype of the high temperature DFE from
exhaust side of the filter element. The first DOC had a Filter Services and Testing Corporation (Filter Services). That
platinum-based formulation wash-coated onto a substrate. DFE was similar in dimensions, design and physical properties to
Preliminary measurements showed that particular catalyst the Donaldson DFE. The filter material was constructed from
formulation tend to increase the NO2 emissions. The second high temperature fibreglass medium.
DOC, wash-coated with an alternative palladium (Pd) based
catalyst formulation was designed with intention of mitigating Fuelling procedures
the NO2 problem. Precautions were taken to avoid cross contamination of the fuels
from consecutive fuel tests. Between tests, the auxiliary tank was
Clean air power/ETG catalytic particulate oxidiser drained and rinsed with the next fuel formulation and the fuel
(CPO) system filter was replaced with a new one. After filling the emptied
auxiliary tank with the desired test fuel, the vehicle fuel system
A Clean Air Power/ETG CPO system Model CPO180 was was purged by operating the vehicle for two full cycles in the
installed temporarily on LHD2. The manufacturer states that the isolated zone prior to the start of the test. During this process the
catalytic particle oxidiser (CPO) uses filter cartridges made of excess fuel from the fuel system was diverted to the primary
catalysed silica fibres that are able to catalytically oxidise carbon vehicle fuel tank. After completing two purge cycles, the
particles at low temperatures without actually trapping them on a auxiliary tank was refilled with test fuel, the return line was
filter. The tested system consists of two units; a CPO followed by diverted back to the auxiliary tank, and the test run started.
a DOC, and was installed under the guidance of the distributor.
Isolated zone testing
Filtration system with disposable filter element (DFE)
from Donaldson (Model P604516) The objective of this study was to evaluate the performance of
selected control technologies using isolated zone testing. Isolated
One of the filtration systems tested during this study was a using zone tests are designed to be a compromise between the
high temperature disposable filter element (DFE) from genuineness of in situ measurements of concentrations and
Donaldson Company Inc. This pleated filter medium is highly corresponding exposures, and the repeatability and accuracy of
porous and creates relatively low exhaust backpressure when the emission measurements obtained under research laboratory
new. Since the filter material can withstand temperatures as high conditions. When carefully designed, isolated zone tests allow

Eighth International Mine Ventilation Congress Brisbane, QLD, 6 - 8 July 2005 265
S E MISCHLER, A D BUGARSKI and G H SCHNAKENBERG Jr

the operation of vehicles under conditions and over duty cycles Ventilation
that closely mimic actual production duty cycles. In addition,
artifacts usually generated under laboratory conditions while The isolated zone was ventilated with fresh air from the near-by
attempting to simulate real-life conditions and processes do not portal. Since diesel-powered vehicles were not operated
compromise the results. Finally, isolated zone test measurements upstream of the test zone during the tests, the level of diesel
are not confounded by emissions from other diesel-powered contamination of the ventilation air upstream of the test zone was
expected to be very low. The intent was to maintain the same
vehicles as would occur in tests conducted in production areas.
ventilation rate (VR) of about 18.88 m3/s for all tests. This
The effects of each of the selected control technologies on amount of air was assumed to provide relatively stable air flow
DPM and gas concentrations in the mine air are determined from and good mixing of the pollutants.
the measurements taken while each test vehicle is operated
within the isolated zone with and without control technologies.
In general, corrections for the background concentrations of the
Sampling and measurements methodology
pollutants are made by subtracting the results of measurements
performed at the upstream end of the zone from the Sampling strategy
corresponding results obtained at the downstream end of the
The following procedure was established for sampling in the
isolated zone or on the vehicle. The efficiency of any technology
isolated zone:
tested is established by comparing the pollutant concentrations
determined for the case when the technology was used to the 1. The vehicle was driven to the fuelling station prior to the
corresponding concentrations determined for the case when the test and the auxiliary tank topped off with fuel. While the
same vehicle over the same duty cycle with a standard muffler vehicle was fuelled, the operator was briefed on the details
and fuel, ie the baseline configuration. of the test protocol and instructed on the duty cycles.
2. After refuelling was completed, the operator performed two
Test site warm-up cycles.
The isolated zone consisted of a 533 m ramp with a nine per cent 3. As the vehicle passed the fuelling station near the end of
rise towards the downstream end. The upstream end of the zone the second warm-up cycle, the auxiliary tank was topped up
was situated approximately 150 m from the portal and the entry and the vehicle proceeded to the upstream load/dump point
point of fresh air. The elevation of the portal is approximately where it was stopped. The particulate matter samplers and
1525 m above sea level. The average cross-sectional dimensions all real-time sampling instrumentation on the vehicle were
of the isolated zone opening were approximately 3.6 m by 2.7 m. turned on. At this time the test officially commenced.
The site selected for the isolated zone met the following 4. At approximately the same time, the upstream sampling
requirements: systems were turned on.
1. it is isolated from other parts of the mine where 5. After a short delay, sampling at the downstream station was
diesel-powered equipment is used; started. A real-time EC concentration instrument was used
2. it is ventilated with fresh air directly from the mine portal; to determine start time.
3. the quality and quantity of the air is not compromised by 6. The objective was to collect at least 30 g of elemental
portal traffic; carbon on the HV sampling filters used at the downstream
sampling station. The duration of a test was estimated from
4. the zone is sufficiently long and the opening is relatively the real-time measurements of particulate concentrations at
small to ensure thorough air mixing at the planned the downstream sampling station using the TEOM 1400a.
ventilation rates and uniform contaminant distribution The actual stop time was determined by using the real-time
across the drift at the downstream sampling station; trace of EC concentrations.
5. the ventilation controls allow for adjustment and 7. Shortly thereafter, when the vehicle reached the upstream
maintenance of relatively constant average air quantities load/dump point, the vehicle and the upstream sampling
during the tests; and stations were stopped and the test was terminated.
6. electric power to operate equipment and instrumentation is 8. The actual start and stop times and total sampling times
available at the downstream and upstream sampling were recorded.
stations.
The schematic of the isolated zone is shown in Figure 1. The Sampling methodology and instrumentation
two test LHD vehicles were operated over a simulated duty cycle A description of various equipment, instrumentation, and
between the upstream and downstream load/dump points that methods used in this study to collect particulate samples and
were approximately 305 m apart. For this study, three sampling directly measure concentrations of particulates and selected
stations were established in the isolated zone: the upstream gases is given below.
station, the downstream station, and the vehicle station. The
upstream sampling station was located approximately 91 m Standard sampling of DPM for carbon analysis
upstream of the upstream load/dump point. The downstream
sampling station was located approximately 137 m downstream A standard sampling method similar to the one used by the US
of downstream sampling station. The ventilation control doors Mine Safety and Health Administration (MSHA) for DPM
were located approximately 61 m downstream of the downstream compliance monitoring (66 Fed Reg 5706 and corrections 66 Fed
sampling station. Sufficient quantities of waste rock were Reg 35518 2001) was used to collect DPM samples at the
available on upstream and downstream load/dump points. downstream, upstream and vehicle sampling stations during the
fuel tests for carbon analysis. The samples at the upstream and
Duty cycles downstream stations were collected in triplicate. The 1.7 L/m
sampling flow rate used by this method is inadequate for
A defined, conservative, simple and repeatable duty cycle for the obtaining DPM samples in cases where the concentration of
LHD vehicles, schematically represented in Figure 1, was DPM is low and sampling times are short. Therefore, this method
developed for these tests. The test cycle simulated a production was not used to collect samples during the tests on the filtration
cycle typical for the LHD equipment. systems.

266 Brisbane, QLD, 6 - 8 July 2005 Eighth International Mine Ventilation Congress
DIESEL PARTICULATE MATTER CONTROL TECHNOLOGIES AND MEASUREMENTS IN US MINES

FIG 1 - LHDs in the isolated zone.

Eighth International Mine Ventilation Congress Brisbane, QLD, 6 - 8 July 2005 267
S E MISCHLER, A D BUGARSKI and G H SCHNAKENBERG Jr

At the downstream and upstream sampling stations, a model temperatures are controlled to minimise the effects of ambient
0523-101Q high volume rotary vane pump from Gast Company, temperatures. To prevent condensation and ensure that the
Benton Harbor, Michigan and a manifold with a bank of critical sample filter always collects particulates under similar
orifices with a nominal flow rate of 1.7 L/m were used instead of conditions, the TEOM intake is heated to maintain the sampling
a MSA Elf Model pump to maintain and control the flow rate. At stream at 50C.
this flow rate, only aerosols with D50 smaller than 0.820 m During this study, the flow rates on both the upstream and
reached the collection filter (Olson, 2001). downstream TEOM was set at 1.7 L/m. A cyclone and impactor
were used as preclassifiers to the TEOM, allowing only particles
High volume (HV) method for sampling DPM for carbon with an average aerodynamic diameter (D50) smaller than
analysis 0.820 m to reach the collection filter. The average ambient
concentrations of TPM were recorded and saved every ten
During all tests conducted in this study, a HV sampling method seconds. The reported average concentrations for a test were
was used to collect three DPM samples at the downstream and obtained from the difference in filter masses recorded at the same
two DPM samples at the upstream sampling stations. This high start and stop times used by the particulate samples for carbon
volume sampling train, described elsewhere (Bugarski et al, analysis.
2004), was used to enhance the collection of samples for carbon
analysis by increasing the sampling flow rate and decreasing the
Measurement of concentration of O2, CO, NO and NO2
collection area of the filter. The sampling flow rate was increased
by merging flows from five preclassifiers, each consisting of a using an industrial scientific iTX multi-gas monitor
10 mm Dorr-Oliver cyclone followed by a US Bureau of Mines The ambient concentrations of carbon monoxide (CO), nitrogen
(USBM) single stage diesel impactor, into a single stream. dioxide (NO2) and nitric oxide (NO) were measured at all three
A flow rate of approximately 2.0 L/m was maintained through sampling locations using Industrial Scientific iTX multi-gas
each cyclone and impactor pair. At this sampling flow rate only monitors. One iTX multi-gas monitor was dedicated to each of
particles with geometric mean smaller then 0.775 m were the sampling locations for the duration of the isolated zone
deposited on the filters (Olson 2001). All five preclassifiers were testing. The iTX measured ambient concentrations every ten
attached to a symmetrical plenum that distributed a total flow seconds and stored them in its memory.
rate of approximately 10.0 L/m uniformly among the five The iTX is a diffusion gas monitor based on electrochemical
streams. Each of the preclassifier assemblies was connected to cell technology implemented in such a way as to measure
the plenum chamber by a 0.91 m long section of conductive concentrations independent of atmospheric pressure. The
tubing. The outlet of the plenum was directly connected to a instrument continuously monitors and simultaneously displays
stainless steel 25 mm diameter filter holder containing two
all gases sampled. The concentration data were downloaded to a
stacked 25 mm tissue quartz fibre filters (Tissuequartz 2500QAT,
laptop PC at the end of each test day.
Pall Corporation, Ann Arbor, Michigan).
The iTX gas monitors were calibrated with certified
The total sampling mass flow rates were maintained using a
concentrations of Industrial Scientific calibration gases prior to
Model HFC 302 mass flow controller from Teledyne, Hampton
VA in the each of the three sampling lines from the 25 mm and upon completion of the isolated zone testing. Additionally,
diameter filter holder. The total volumetric flow rates through each each iTX was checked daily using the iTX DS1000 Docking
of the sampling streams were measured periodically by inserting a Station. The iTX DS1000 Docking Station is an automated
Gilibrator II bubble flow metre inline, between filter holders and instrument management system, which consists of a master
mass flow controllers. The volumetric flow rates measured during control and PC interface station. The Docking Station provides
this study were all corrected to ambient conditions. A Model automatic calibration and instrument diagnostics as well as
0523-101Q high volume rotary vane pump from Gast Company maintenance of the instrument database records.
was used to draw the sample though the filter.
Measurements of ventilation rate
DPM concentration measurements with a tapered Air velocities in the isolated zone were measured continuously
element oscillating microbalance (TEOM) series 1400a during the tests in the approximate centre of the drift at the
ambient particulate monitor downstream sampling station using an Anemosonic UA6 digital
Two TEOM Series 1400a ambient particulate monitors from ultrasonic anemometer. The anemometer sensor was located in
Rupprecht and Patashnick Co were used to continuously measure the centre of the steel grid supporting the DPM samplers. The
concentrations of total particulate matter (TPM) with a D50 data was logged and stored at ten-second intervals.
smaller than 0.820 m. One TEOM was located at the upstream The air velocities were also measured periodically using a
station, the other at the downstream sampling station. vane anemometer to conduct a moving traverse across the entry.
The TEOM draws air through a filter at a constant flow rate, The average air velocity for a test was determined by
while continuously measuring the mass accumulating on the averaging all of the downstream air velocity readings over the
filter and calculating near real-time mass concentrations. The period for downstream DPM sampling.
sample stream is drawn through this filter from a hollow tapered
element that is connected to the suction side of the sampling Analysis of the samples
system. The vibration frequency of the tapered element decreases
as particulate mass collected on the filter increases. By
frequently measuring the tapered element frequency, the TEOM Analysis of the DPM samples collected using
calculates the increase in mass of the sample that has standard and high volume sampling method for
accumulated on the filter. The concentration of TPM can be elemental carbon content
calculated by dividing the accumulated mass by the volume of Samples collected on quartz fibre filters, using the standard and
airflow across the filter during the period over which the the high volume sampling procedures, were analysed by the
frequency change is measured. National Institute for Occupational Safety and Health, Pittsburgh
The flow through the instrument is maintained at a constant Research Laboratory (NIOSH PRL) analytical laboratory for
rate by a mass flow controller. The flow is corrected for elemental carbon content using the NIOSH Analytical Method
temperature and barometric pressure. The internal instrument 5040 (NIOSH, 1999; Birch and Cary, 1996).

268 Brisbane, QLD, 6 - 8 July 2005 Eighth International Mine Ventilation Congress
DIESEL PARTICULATE MATTER CONTROL TECHNOLOGIES AND MEASUREMENTS IN US MINES

NIOSH Analytical Method 5040 analyses for OC and EC in two [C ]


different stages. In the first stage, the OC evolves as the instrument Control Technology Effect for C i [%] = 1 i ,VR 100 (3)
[C i ]BL
ramps the oven temperature up over four progressively higher
temperature steps in a pure helium (He) atmosphere. The EC does
where:
not evolve in the pure He atmosphere. The evolved OC is oxidised
to carbon dioxide (CO2), reduced to methane (CH4), and finally [Ci,VR] is the ventilation-adjusted net concentration of the
measured using a flame ionisation detector (FID). monitored pollutant for the control technology test
In the second stage, the oven temperature is reduced to [Ci]BL is the net concentration of the monitored pollutant for
approximately 600C. The EC is evolved as the instrument steps the baseline test
the oven temperature back up to approximately 900C in a
He/oxygen (O2) atmosphere. Because of the O2, EC is oxidised RESULTS AND DISCUSSION
to CO2, which is consequently reduced to CH4 and measured
with the FID. The NIOSH Analytical Method 5040 also corrects Results from this study are reported in three sections. The first
for the pyrolysis of OC and carbonates. section discusses results from the ventilation measurements. The
second section discusses the results from the ten tests evaluating
Calculation of the effects of control technologies the effects of selected fuel formulations on concentrations and
characteristics of particulate matter and selected gases. The third
The effects of each of the tested control technologies were section is dedicated to results of filter tests conducted to establish
determined by comparing the results of tests with and without the effects of selected diesel particulate filter (DPF) systems
control technologies. The effects of a specific control technology and disposable filtration elements on concentrations and
were quantified by comparing the contaminant concentrations characteristics of the same pollutants
observed for the test when vehicle was operated with a control
technology to those observed for the baseline test. The test of Ventilation air flow rates for the study
vehicle LHD1 using #1 diesel and standard muffler was used to
establish a baseline for the fuel tests. The test with vehicle Air velocities were continuously measured in the centre of the
LHD2, using #1 diesel fuel with a muffler, was used to establish drift during each of the tests at the downstream and upstream
a baseline for filter tests. sampling stations. Ventilation air volumes were calculated by
multiplying corrected flow rates by the corresponding
The concentrations of selected pollutants at the upstream and cross-sectional area of the drift. The average ventilation rates for
downstream sampling stations were determined using various downstream sampling station are presented in Table 5 for the fuel
sampling and measurements methods. Owing to the nature of the tests and Table 6 for the filter tests.
test environment and methods, small variations in operating
conditions were possible between tests, including variations in
the ventilation rate, air temperature, etc. The data show that TABLE 5
ventilation rate differences between tests could be the largest Average ventilation rates for the downstream sampling station
during the fuel tests.
source of uncertainty for the concentration measurements. This
factor of uncertainty can be eliminated however, by correcting all Fuel tests (LHD1) Average ventilation
measured concentrations for ventilation rate. The calculations for rates (m3/sec)
these corrections are presented below. #1 Diesel/Muffler 19.28
For the purpose of these analyses, the tests conducted in this PuriNOx Cold-weather/Muffler 19.39
study were divided into two groups, as shown in Table 1 and
PuriNOx Warm-weather/Muffler 19.59
described earlier. In order to allow a direct comparison of the
results between the tests within each test group, the measured 20% Soy Biodiesel blend/Muffler 19.25
concentrations (ci) were corrected to the average ventilation rate 50% Soy Biodiesel blend/Muffler 19.16
for the baseline tests for each of the two groups (VRbaseline LHD1 50% Soy Biodiesel blend/DOC 19.84
and VRbaseline LHD2) using Equation 1.
20% Yellow grease biodiesel blend/Muffler 19.60
m3 50% Yellow grease biodiesel blend/Muffler 18.99
VRi [ ]
g g s ULS Diesel/Muffler 19.19
ci ,VR [ 3 ] = ci [ 3 ] (1)
m m m3 ULS Diesel/DOC 19.44
VRbaseline [ ]
s

The net contribution of the tested vehicle/technology


configuration to the air concentrations of pollutants (Ci, VR) was TABLE 6
calculated by subtracting the concentrations measured at the Average ventilation air flow rates for the downstream sampling
upstream sampling station (ci,VR,UP) from the ventilation adjusted station during the filter tests.
concentrations at the downstream sampling station (ci, VR,DOWN): Filter tests (LHD2) Average ventilation rates
(m3/sec)
g g g
C i ,VR [ ] = ci ,VR , DOWN [ 3 ] ci ,UP [ 3 ] (2) Muffler 19.29
m3 m m
AM Pd DOC 19.27
In cases where the analysis of the upstream data showed that Donaldson 19.36
the upstream concentrations of the measured pollutant were Filter Service 19.57
below the detection limit of the method or instrumentation, the
background concentrations were assumed to be negligible.
The corrected concentrations were then used to calculate Figure 2 shows the example of ventilation rate measured at the
the relative effects of the tested control technologies on the downstream station for the baseline fuel test. The ventilation
concentrations of the monitored pollutants: rates were strongly influenced by movement of the vehicle inside

Eighth International Mine Ventilation Congress Brisbane, QLD, 6 - 8 July 2005 269
S E MISCHLER, A D BUGARSKI and G H SCHNAKENBERG Jr

the isolated zone during the tests. However, the average air flows mine air. Test results with seven alternative fuel formulations were
among the tests were nonetheless very consistent, as indicated in compared with those of #1 diesel, selected as baseline fuel. The
Table 5 and Table 6. objective of the two other tests was to evaluate the effects of a
The average airflow rates presented in Table 5 were used to DOC when used with alternative fuels. The effects of the vehicle
adjust all measured concentrations of particulate matter and equipped with a DOC and fuelled with 50 per cent soy biodiesel
gases to those that would prevail if the average ventilation rates blend or ultra-low sulfur (ULS) diesel were compared with those
for those tests were equal to the average baseline fuel test of the same vehicle fitted with muffler and fuelled with #1 diesel.
ventilation rate of 19.28 m3/s. The average air flow rates
presented in Table 6 were used to adjust all measured Effects of fuel formulations on aerosols
concentrations of particulate matter and gases to the ventilation
rate of the baseline filter test, 19.29 m3/s. The adjustment of Table 7 summarises the results of the elemental carbon and
measured concentrations to a common ventilation rate allowed a TEOM analyses on the effects of fuel formulations on aerosol
direct comparison of the results obtained from different tests. concentrations. Two sets of results are shown for EC analyses.
One set of results is from analyses performed on samples
Effects of fuel formulations collected using a high volume (HV) sampling method. Another
set is from analyses performed on the samples collected by the
Ten fuel tests were conducted during this study. The objective of standard sampling method (SSM). The results are presented as
eight of these tests was to assess the effects of selected fuel the ventilation rate-adjusted average net contributions of the test
formulations on the concentrations of DPM and selected gases in vehicle to the downstream mass concentrations.
24.00

22.00
Ventilation Rate [m /s]
3

20.00

18.00

16.00
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Time [s]

FIG 2 - Ventilation rates measured during the fuel tests at the downstream sampling station.

TABLE 7
Average net contributions and calculated change relative to baseline of elemental carbon (EC) concentrations and TEOM determined
total particulate matter (TPM). (Common ventilation rate = 19.28 m3/s.)

Fuel tests (LHD LHD1) EC EC TPM


NIOSH 5040 HV NIOSH 5040 SSM TEOM
Net contributions % Change Net contributions % Change Net contributions % Change
to conc relative to to conc relative to to conc relative to
(g/m3) baseline (g/m3) baseline (g/m3) baseline
#1 Diesel/Muffler 275 - 262 - 388 -
PuriNOx Cold-Weather/Muffler 90 67 99 62 207 46
PuriNOx Warm-Weather/Muffler 40 85 40 85 129 66
20% Soy Biodiesel/Muffler 141 49 166 32 244 37
50% Soy Biodiesel/Muffler 93 66 108 48 192 30
50% Soy Biodiesel/DOC 87 68 103 61 - -
20% YG Biodiesel/Muffler 184 33 177 30 278 27
50% YG Biodiesel/Muffler 120 56 121 44 212 46
ULS Diesel/Muffler 250 9 243 3 405 -3
ULS Diesel/DOC 188 32 187 32 278 27

270 Brisbane, QLD, 6 - 8 July 2005 Eighth International Mine Ventilation Congress
DIESEL PARTICULATE MATTER CONTROL TECHNOLOGIES AND MEASUREMENTS IN US MINES

Effects of fuels on the concentration of elemental changes between the reported TEOM filter masses at the start
carbon and end of the sampling period.
TPM concentrations at the upstream sampling station were
The average net contributions of a tested configuration to the EC found to be very low. In several instances, due to external
concentration at the downstream station during fuel formulation vibration and shocks imposed on the instrument, the readings
tests are shown in Table 7. EC concentrations measured at the were found to be negative. Therefore, upstream concentrations
upstream sampling station were below the limits of detection were assumed to be negligible and the net vehicle contributions
(0.9 g/sample for HV and 1.6 g/sample for SKC) and thus to TPM concentrations were assumed to be equal to downstream
were not used in the calculations. concentrations.
Results shown in Table 7 were used to calculate the relative TPM concentrations observed at the downstream sampling
effects of alternative fuel formulations and DOC on EC and TPM station during the baseline test as well as tests with cold- and
concentrations. Relative effects were calculated by comparing warm-weather water emulsions are shown in Figure 3. The peak
the net contributions from the each of the fuel tests to those from concentrations and cumulative net contributions to TPM
the baseline test (a muffler with #1 diesel fuel). Positive values concentrations, measured by TEOM, were substantially lower in
indicate reductions in concentrations. cases where water emulsions were used. The relative reductions
Results of the EC analysis on the HV samples show that in net TPM concentrations were calculated to be 46 and 66 per
the cold- and warm-weather water-fuel emulsions reduced cent for cold- and warm-weather water emulsions, respectively.
concentrations of EC by 70 and 85 per cent, respectively. The EC The net contributions of the TPM concentrations, measured by
analysis performed on SSM samples showed fairly comparable TEOM, were found to be substantially reduced when #1 diesel
results with a 62 per cent and 85 per cent reduction in net EC was substituted with soy and yellow grease biodiesel blends. The
contribution for the two water-fuel blends, respectively. data in Table 7 shows that the reductions were found to be slightly
The 20 per cent soy biodiesel and 80 per cent #1 diesel fuel higher for soy than for yellow grease blends. Due to problems with
blend reduced EC concentrations by 49 per cent (HV) and 37 per the instrument at the downstream sampling station, the results are
cent (SSM). An increase in the soy biodiesel fraction to 50 per not available for the test during which the vehicle was equipped
cent resulted in further reductions in the concentrations of EC with a DOC and operated with 50 per cent soy biodiesel blend.
(71 per cent and 59 per cent). The results for the yellow grease
biodiesel blends indicated that EC reductions are slightly less Effects of fuel formulations on gas phase species
pronounced than for the corresponding soy biodiesel blends. During fuel tests, nitric oxide (NO), nitrogen dioxide (NO2),
Results indicated very little difference in the effects of carbon monoxide (CO), and carbon dioxide (CO2) concentrations
ultra-low sulfur (ULS) diesel fuel and baseline #1 fuel on EC were measured continuously at the upstream, downstream, and
concentrations. When ULS fuel was used however, with DOC in vehicle sampling stations. During these tests, the concentrations of
the exhaust system, there was about a 30 per cent reduction in NO, NO2, and CO at the upstream sampling station were found to
EC concentrations. It is important to note that the sulfur content be under the detection limit of the iTX multigas instrument (0.1
of #1 diesel fuel was approximately 255 ppm and the sulfur ppm, 0.1 ppm, and 1 ppm, respectively). Therefore, the net
content of ULS diesel was 10 ppm. contributions of the vehicles to those gases were assumed to be
equal to their concentrations at the downstream and vehicle
Effects of fuels on the concentration of total particulate locations. Continuous measurements with the RKI Eagle CO2
matter (TPM) measured using the TEOM 1400a monitor at the upstream sampling station showed that the
background concentration of CO2 remained relatively constant
The ventilation-adjusted average concentrations of TPM and the throughout the tests, averaging 321 ppm when corrected for local
calculated reductions relative to baseline are presented in Table atmospheric pressure. This value was used during the analysis as
7. The average concentrations of TPM were calculated as the average background concentration of CO2 for all tests.

700

600
TPM Mass Concentration [g/cm3]

500

400

300

200

100

0
0 1000 2000 3000 4000 5000 6000

Time [s]

#1 Diesel / Muffler Downstream PuriNOx Cold-Weather / Muffler Downstream

PuriNOx Warm-Weather / Muffler Downstream


3
FIG 3 - The effects of water emulsions on mass TPM concentrations measured by TEOM. (Common ventilation rate = 19.28 m /s.)

Eighth International Mine Ventilation Congress Brisbane, QLD, 6 - 8 July 2005 271
S E MISCHLER, A D BUGARSKI and G H SCHNAKENBERG Jr

Gas data was adjusted to the common ventilation rate of the EC concentrations by 92 per cent. The disposable filter element
baseline fuel test. The ventilation adjusted net contributions of from Filter Services reduced the contribution by 70 per cent.
fuel formulations and diesel oxidation catalyst to average and
maximum peak concentrations of NO, NO2, CO, and CO2 are Effects of filtration systems on the total particulate
summarised for downstream and vehicle sampling stations in matter mass concentration measured using TEOM
Table 8. 1400a
NO, NO2, and CO results collected at the downstream
Continuous measurements of TPM mass concentrations from
sampling station during the tests on PuriNOx water emulsions,
the TEOM 1400a instruments at the downstream and upstream
yellow grease biodiesel blends, and ULS diesel are not available
sampling stations are shown in Figure 4. TPM concentrations
due to problems in data acquisition. The results of measurements
at the upstream sampling station were much lower than
conducted at the vehicle sampling station with PuriNOx
the downstream concentrations and thus were not used in
warm-weather water emulsion and 20 per cent yellow grease
the calculations. The average concentrations of TPM were
biodiesel blends are also not available due to problems
calculated using the differences between the reported TEOM
encountered during the data recording process. An examination
filter masses at the start and the end of the sampling period.
of the measurements obtained at the downstream location
The ventilation-adjusted averages are presented in Table 9.
showed little difference in the NO, NO2, and CO concentrations
between the baseline and the fuel formulations tested. Generally, results obtained from TEOM measurements are in
agreement with the gravimetric results. TEOM results show that
Effects of filtration systems downstream TPM concentrations were reduced from those
observed during the baseline test by 72 per cent in the case of
The ventilation rate-adjusted concentrations of EC and TPM AM DPF system with Pd catalysed DOC, by 76 per cent in the
measured at the downstream station during the filter tests, are case of Donaldson and by 65 per cent in the case of Filter
summarised in Table 9. Services disposable filter elements (DFE)s.
Throughout the tests, continuous monitoring with the TEOM
Effects of filtration systems on aerosols instrument revealed differences in the effects that the tested
filtration systems had on TPM concentrations at the downstream
Effects of filtration systems on elemental carbon mass sampling station, as shown in Figure 4. The mass concentrations
concentration of TPM during the test with the Donaldson DFE steadily
declined throughout the length of the test. It can be hypothesised
The results presented in Table 9 show that two systems, the AM that the relatively clean filtration element from Donaldson,
DPF system with a palladium (Pd) catalysed DOC and the having operated for only a couple of hours prior to the test,
Donaldson DFE, reduced the net contribution of ambient vehicle gained efficiency while accumulating DPM throughout test.

TABLE 8
Net contribution of fuel formulations to average and peak NO, NO2, CO and CO2 concentrations at the downstream sampling station.
(Common ventilation rate = 19.28 m3/s.)

Fuel tests (LHD1) Net contributions to concentrations


NO Avg NO Max NO2 Avg NO2 Max CO Avg CO Max CO2 Avg CO2 Max
(ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm)
#1 Diesel/Muffler 3.7 13.0 0.1 0.4 2.2 10.0 808 2258
PuriNOx Cold-weather/Muffler - - - - - - 816 2270
PuriNOx Warm-weather/Muffler - - - - - - 778 2320
20% Soy biodiesel blend/Muffler 4.2 13.0 0.1 0.6 1.8 7.0 844 2307
50% Soy biodiesel blend/Muffler 4.1 12.9 0.1 0.6 1.6 7.0 852 2349
50% Soy biodiesel blend/DOC 4.4 13.4 0.3 1.1 0.0 0.0 859 2269
20% YG Biodiesel blend/Muffler - - - - - - 843 2269
50% YG Biodiesel blend/Muffler - - - - - - 861 2301
ULS Diesel/Muffler - - - - - - 877 2300
ULS Diesel/DOC 3.5 11.1 0.2 0.9 0.0 0.0 855 2303

TABLE 9
Average net contributions and calculated change relative to baseline of elemental carbon (EC) concentrations and TEOM determined total
particulate matter (TPM). (Common ventilation rate = 19.29 m3/s.)
Filter tests (LHD LHD2) EC NIOSH 5040 TPM TEOM
Net contributions to % Change relative to Net contributions to % Change relative to
concentrations baseline concentrations baseline
(g/m3) (g/m3)
Muffler 105 - 180 -
AM Pd DOC 8 92 51 72
Donaldson 9 92 43 76
Filter service 31 70 62 65
AM Pd DOC regeneration - - 183 -

272 Brisbane, QLD, 6 - 8 July 2005 Eighth International Mine Ventilation Congress
DIESEL PARTICULATE MATTER CONTROL TECHNOLOGIES AND MEASUREMENTS IN US MINES

350.0

300.0

TPM Mass Concentration [g/cm3]


250.0

200.0

150.0

100.0

50.0

0.0
0 1000 2000 3000 4000 5000 6000 7000 8000

Time [s]

Muffler Downstream AM Pd DOC Downstream

Donaldson Downstream Filter Systems Downstream

FIG 4 - Filtration system effects on TPM mass concentrations measured throughout the tests by TEOM.
(Common ventilation rate = 19.29 m3/s.)

TABLE 10
Net filtration system contributions of NO, NO2, CO and CO2 concentrations at the downstream sampling station.
(Common ventilation rate = 19.29 m3/s.)
Filter tests (LHD2) Net contributions to concentrations
NO Avg NO Max NO2 Avg NO2 Max CO Avg CO Max CO2 Avg CO2 Max
(ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm)
Muffler 3.5 12.0 0.1 0.6 0.6 4.0 572 1574
AM Pt DOC - 8.0 - 1.9 - 0.0 - -
AM Pd DOC 3.1 11.0 0.2 0.8 0.0 0.0 487 1335
AM Pd DOC Regeneration - 11.0 - 1.1 - 0.0 - -
Donaldson 3.6 13.0 0.0 0.2 0.7 4.0 551 1658
Filter Services 4.1 13.2 0.0 0.2 0.7 4.1 590 1783
CAP/ETG CPO - 9.0 - 2.9 - - - -

Effects of filtration systems on gas phase species REFERENCES


30 CFR 57.5060. Diesel particulate matter exposure of underground
Nitrogen dioxide (NO2)
metal and nonmetal miners. Limit on concentration of diesel
NO2 results, presented in Table 10, show that the test vehicle, particulate matter, Code of Federal Regulations, Washington, DC,
equipped with the AM DPF and a Pt DOC, had average and USA (Government Printing Office, Office of the Federal Register).
maximum peak NO2 concentrations approximately three times 30 CFR 57.5065. Diesel particulate matter exposure of underground
higher (0.3 versus 0.1 and 1.9 versus 0.6 ppm, respectively) than metal and nonmetal miners. Fueling practices, Code of Federal
for the baseline test. A substantially smaller increase in NO2 Regulations, Washington, DC, USA (Government Printing Office,
Office of the Federal Register).
concentration (0.8 versus 0.6 ppm) was observed when the
vehicle was fitted with the AM DPF and a Pd DOC. Birch, M E and Cary, R A, 1996. Elemental carbon-based method for
monitoring occupational exposures to particulate diesel exhaust,
It is important to note that the 19.29 m3/s average ventilation Aerosol Science and Technology, 25:221-241.
rate maintained during these tests was significantly higher than Bugarski, A D and Schnakenberg, H G, 2001. Field evaluation of diesel
the MSHA established 5.3 m3/s ventilation rate for the Deutz particulate filters: size selective measurements of aerosols in mine air
BF41013FC engine. Despite this relatively high air quantity and engine exhaust, in Proceedings Mining Diesel Emissions
however, at the vehicle sampling location the peak NO2 Conference MDEC 2001, Markam, Ontario, November.
concentration for the AM DPF having a Pt DOC exceeded Bugarski, A D and Schnakenberg, H G, 2002. Evaluation of diesel
5 ppm, the ACGIH Ceiling Limit for NO2 currently enforced in particulate filter systems at INCO Stobie Mine, in Proceedings
underground metal and non-metal mines (30 CFR 57.5001). Mining Diesel Emissions Conference MDEC 2002, Markam,
Ontario, October.
Similarly elevated NO2 concentrations were recorded during
Bugarski, A D, Schnakenberg, G H, Noll, J D, Mischler, S, Crum, M and
the initial test stages on the CAP/ETG CPO system, resulting in a Anderson, R, 2005. Evaluation of diesel particulate filter systems and
premature termination to prevent overexposure of the vehicle biodiesel blends in an underground mine, SME Transactions, 318.
operator.

Eighth International Mine Ventilation Congress Brisbane, QLD, 6 - 8 July 2005 273
S E MISCHLER, A D BUGARSKI and G H SCHNAKENBERG Jr

McGinn, S, Grenier, M, Bugarski, A D, Schnakenberg, G H and Petrie, Olson, B, 2001. Particle calibration of the disposable personal diesel
D, 2002. Performance evaluation of diesel particulate filter aerosol sampler (PDAS), Particle Calibration Laboratory, University
technology in the underground environment, in Proceedings The of Minnesota, Report submitted to National Institute of Occupational
North American/Ninth US Mine Ventilation Symposium, Queens Safety and Health.
University in Kingston, Ontario, Canada, June. Schnakenberg, G and Bugarski, A, 2002. Review of technology available
MSHA, 2004. Diesel particulate matter control technology. Available to the underground mining industry for control of diesel emissions,
from: <http://www.msha.gov/01-995/Coal/DPM-FilterEfflist.pdf>. US Department of Health and Human Services Information Circular
NIOSH (National Institute for Safety and Health), 1999. Elemental IC 9262. Available from:
carbon (diesel particulate), Method 5040, Issue 3 (Interim), in <http://www.cdc.gov/niosh/mining/pubs/pdfs/ic9462.pdf>.
NIOSH Manual of Analytical Methods, Fourth revised edition,
Cincinnati, OH, USA. Available from:
<http://www.cdc.gov/niosh/nmam/ pdfs/5040f3.pdf>.

274 Brisbane, QLD, 6 - 8 July 2005 Eighth International Mine Ventilation Congress