1
24-025
Determination of Incombustible Content with
Portable Spectrometers Using Chemometric Modelling:
Preliminary Results
Sebastian D’Hyon
Simtars, Redbank, QLD, Australia
Shevaune Zeng
Simtars, Redbank, QLD, Australia
Amoussou Coffi Adoko
Nazarbayev University, Astana, Kazakhstan
Eleonora Widzyk-Capehart
Simtars, Redbank, QLD, Australia
Gareth Kennedy
Simtars, Redbank, QLD, Australia
ABSTRACT
Inertization of coal dust with stone dust is a common con-
trol implemented within underground coal mines to miti-
gate the hazard of a coal dust explosion. To achieve adequate
inertization, the total incombustible content (TIC) of the
coal/stone dust mixture is required to meet certain percent-
ages depending on the risk and relevant legislation. Current
methods for determining the TIC of a mixture involves
laboratory analysis, colorimetric comparison, or the use
of portable instruments. This paper provides preliminary
analysis of determination of a sample’s TIC in real-time
with chemometric modelling, using portable instruments
employing near-infrared (NIR).
INTRODUCTION
The development of portable instrumentation for road-
way dust testing in underground coal mines is not a recent
concept. The National Institute for Occupational Safety
and Health (NIOSH) had developed a commercial prod-
uct called the Coal Dust Explosibility Meter (CDEM) in
the early 2000’s, which was in commercial use by [1]. The
CDEM instrument worked on the broadband infrared
reflectivity (IR) of a sample and was calibrated on Pittsburgh
coal and stone dust samples with up to 80% total incom-
bustible content (TIC). The device uses a photodiode to
evaluate the concentration of total incombustibles present
in a sample. An evaluation of the CDEM for Australian
mining operations recommended its implementation [2].
However, in the study, the comparison of the CDEM was
only investigated against laboratory results and the analy-
sis did not consider natural incombustible content of the
coal samples tested within Australia. In addition, the device
would have had to be modified to allow analysis in accor-
dance with Australian legislative requirements. In 2015,
the device was evaluated in the field for use in Australian
coal mines [3]. Several limitations were observed, includ-
ing poor reference sample design for Australian coal mines,
unfriendly calibration methodology, lack of certification for
intrinsic safety and sensitivity of the measurement to mois-
ture within the sample requiring coal sample to be dried.
This led to the recommendation not to adopt the NIOSH
device in its existing state as a tool for rapid analysis of
stone dust compliance in Australia.
For Queensland, Australia, the Recognised Standard 5
(RS5) stipulates the requirements for incombustible con-
tent (IC) per zone and how the testing methodologies are
to be conducted [4]. RS5 allows for a portable instrument
to be used for the analysis of roadway dust samples if it is
as accurate as laboratory analysis however, RS5 does not
stipulate the accuracy required for laboratories undertaking
the analysis. This poses a challenge with how accuracy is
reported not only between laboratories but also due to the
24-025
Determination of Incombustible Content with
Portable Spectrometers Using Chemometric Modelling:
Preliminary Results
Sebastian D’Hyon
Simtars, Redbank, QLD, Australia
Shevaune Zeng
Simtars, Redbank, QLD, Australia
Amoussou Coffi Adoko
Nazarbayev University, Astana, Kazakhstan
Eleonora Widzyk-Capehart
Simtars, Redbank, QLD, Australia
Gareth Kennedy
Simtars, Redbank, QLD, Australia
ABSTRACT
Inertization of coal dust with stone dust is a common con-
trol implemented within underground coal mines to miti-
gate the hazard of a coal dust explosion. To achieve adequate
inertization, the total incombustible content (TIC) of the
coal/stone dust mixture is required to meet certain percent-
ages depending on the risk and relevant legislation. Current
methods for determining the TIC of a mixture involves
laboratory analysis, colorimetric comparison, or the use
of portable instruments. This paper provides preliminary
analysis of determination of a sample’s TIC in real-time
with chemometric modelling, using portable instruments
employing near-infrared (NIR).
INTRODUCTION
The development of portable instrumentation for road-
way dust testing in underground coal mines is not a recent
concept. The National Institute for Occupational Safety
and Health (NIOSH) had developed a commercial prod-
uct called the Coal Dust Explosibility Meter (CDEM) in
the early 2000’s, which was in commercial use by [1]. The
CDEM instrument worked on the broadband infrared
reflectivity (IR) of a sample and was calibrated on Pittsburgh
coal and stone dust samples with up to 80% total incom-
bustible content (TIC). The device uses a photodiode to
evaluate the concentration of total incombustibles present
in a sample. An evaluation of the CDEM for Australian
mining operations recommended its implementation [2].
However, in the study, the comparison of the CDEM was
only investigated against laboratory results and the analy-
sis did not consider natural incombustible content of the
coal samples tested within Australia. In addition, the device
would have had to be modified to allow analysis in accor-
dance with Australian legislative requirements. In 2015,
the device was evaluated in the field for use in Australian
coal mines [3]. Several limitations were observed, includ-
ing poor reference sample design for Australian coal mines,
unfriendly calibration methodology, lack of certification for
intrinsic safety and sensitivity of the measurement to mois-
ture within the sample requiring coal sample to be dried.
This led to the recommendation not to adopt the NIOSH
device in its existing state as a tool for rapid analysis of
stone dust compliance in Australia.
For Queensland, Australia, the Recognised Standard 5
(RS5) stipulates the requirements for incombustible con-
tent (IC) per zone and how the testing methodologies are
to be conducted [4]. RS5 allows for a portable instrument
to be used for the analysis of roadway dust samples if it is
as accurate as laboratory analysis however, RS5 does not
stipulate the accuracy required for laboratories undertaking
the analysis. This poses a challenge with how accuracy is
reported not only between laboratories but also due to the