Twenty years of gas chromatographs at QLD underground coal mines – Part one

Source: Queensland Mines and Energy

The Warden’s Inquiry into the 1986 Moura No. 4 Mine Disaster called for mines rescue stations to be equipped with gas chromatographs (GC’s) “to accurately and expeditiously determine the explosibility of a mine air sample”.
The 1975 Warden’s Inquiry into the Kianga No. 1 explosion recommended;
“All coal mines have available at short notice the means of analysing the air samples obtained while dealing with an out-break of fire below ground. This end may be accomplished by either mobile laboratories or laboratories established in each mining locality.”
During a coal fire, advanced spontaneous combustion event or post explosion the need to perform a complete analysis by GC is not only critical but the only option for an accurate assessment of the flammability status of the underground environment.
The use of a GC expands analytical capabilities to include gases crucial in the interpretation of spontaneous combustion events, particularly ethylene and hydrogen. The GC provides a complete analysis of the gases expected underground and is the only reliable technique employed by mines for the measurement of hydrogen, nitrogen, ethylene and ethane.
During the Moura No. 4 Inquiry the ability required to maintain and operate a gas chromatograph was debated as there were differences in opinion as to the skills required. The Moura No.4 Warden’s recommendation did not eventuate exactly as written, instead a more effective approach involving the installation of GCs at each underground coal mine, rather than at rescue stations, linked back to a “master” GC.
It was possible to exceed the Warden’s recommendation because of the emergence of onboard communication capabilities in GCs. Using this communication capability, Simtars developed a scheme called Camgas (Computer Assisted Mine Gas Analysis System). This allowed mine site personnel to access expert gas chemists based at the Simtars gas analysis laboratory at Redbank. The gas chemists were able to optimise the operation of the GC, provide initial training in its use and then oversee GC operation at each mine site remotely, thereby reducing the expertise required at the mine site and assisting in the ongoing operability at the mine site of this complex and often temperamental piece of analytical equipment.  Effectively Camgas provided each mine site with an expert gas chemist 24/7 at each mine site.
This system overcame problems previously experienced where mine sites purchased a GC, installed it on site and waited until it was needed in an emergency. Unfortunately there could be no guarantee that when this need eventuated there would be a competent operator available or even that the equipment would function properly since it may not have been maintained between infrequent usage.
The first Camgas system was installed at Cook Colliery in 1989. Since then there have been quantum leaps in GC analysis in both hardware and software. These improvements have seen the GC now being used proactively rather than just reactively after a major incident.

Basic principles of gas chromatography
A gas chromatograph is essentially a long column through which the gas sample must travel prior to entering a detector. A representative portion of the gas sample is injected into the instrument and “pushed” through the separating column to the detector under the influence of an inert “carrier gas”.
The separating column is the heart of the instrument and must be able to completely separate the component gases so that the subsequent measurement can be truly interference free. It is in this column that intended interactions between the gas components in the sample and the packing of the column result in the different sample gases exiting the column at characteristic and predictable times. The resulting separation of sample components allows the chromatographer to measure each gas concentration sequentially using a detector placed at the column exit.
The role of the detector is to determine the amount of component eluting from the column. The concentration of the component present is proportional to the magnitude of the detector’s response. Generally the detector has little or no ability to determine the identity of a particular chemical as it passes over it. Instead the identity is determined by the time the component takes to travel through the column, known as retention time.

The Camgas evolution
The first Camgas GCs used were Perkin Elmer 8500 gas chromatographs as these were the only instruments at the time with the communications capability to realise the virtual chemist concept. Simtars and Perkin Elmer partnered to develop a configuration of the 8500 GC suitable for coal mine gas analysis.
These GCs were fitted with dual separating columns operating in series. The columns used were 1/8 inch outer diameter packed molecular sieve and porous polymer columns. These columns could be more than five metres long depending on separation requirements. Successful analysis relied on the columns being perfectly matched and the development of methods in which the order of flow through the columns was switched at exactly the right time. It was a difficult and time consuming task to achieve matched columns with minimal analysis times. Unfortunately the required timing of this switching varied as the separation efficiency of the columns degraded with time and then needed to be reset following regeneration of the columns.
The time for analysis using these systems could be up to 30 minutes depending on the efficiency of the columns.
These systems also used two detectors connected in series. The eluting components firstly passed over the non destructive thermal conductivity detector (TCD) or hot wire detector. These detectors were not particularly sensitive and in order to measure even 50 ppm of hydrogen required the use of argon as a carrier gas. For this mine gas analysis application the TCD was used to measure hydrogen, oxygen and nitrogen. The outlet from the TCD detector was then directed to a methanised flame ionisation detector (FID). The methaniser was used to convert the already separated carbon monoxide and carbon dioxide to methane enabling improved sensitivity and detection using the FID. Use of the FID added to the intricacy of the instrument with fuel (hydrogen) and air gas supplies needing to be tuned to enable easy lighting of the flame but not compromise sensitivity. The efficiency of the methaniser was also influenced by the hydrogen flow and the catalyst utilised in its operation could be poisoned. This meant that there was a need to monitor the efficiency of the methaniser to ensure that sensitivity requirements were being maintained.
Although communication with a computer was possible with the Perkin Elmer 8500 GC, all of the processing was done onboard using proprietary operating systems and software. For checking by Simtars’ gas chemists, the data and method files needed to be downloaded to a computer connected to the GC at the remote site transferred via modem to another computer and downloaded to a Perkin Elmer 8500 at the Simtars laboratory at Redbank for processing. Any changes made to the methodology were then saved and uploaded to the computer at Redbank from which the mine site could connect to via modem and retrieve the modified method and then upload to their GC.
This meant that remote assistance could only be provided from a laboratory equipped with the same type of GC.
The long analysis time for samples with this initial system did not lend itself to the analysis of large numbers of samples. Furthermore, the sensitivity of these earlier instruments was not necessarily good enough to detect the early onset of spontaneous combustion. This resulted in several mines adopting a philosophy that GCs were there with the primary purpose of being used in a mine emergency.
The second generation Camgas system utilised a Perkin Elmer Autosystem GC. The first of these systems was installed at North Goonyella in 1994. A significant improvement over the previous model was that the Autosystem was computer controlled using Microsoft Windows based software and all data processing was done using the controlling computer.  This improved remote support capabilities significantly. It also opened up the use of remote login software packages that enabled a gas chemist to take control of the processing computer remotely and demonstrate to users onsite how to rectify problems thus improving the skill level of the users. Remote support was now able to be provided from any computer that had the remote login software installed while the option to transfer method and data files to Simtars was still available for routine checking.
The operators of this system required more intense training to incorporate the new controlling software. It meant that those assigned to be GC operators needed a new skill set. This was at a time, unlike now, where familiarity with computers was not widespread.
The column switching process was further complicated in this generation of GC but reduced analysis time down to approximately 14 minutes.