Frother Measurement and Control

A significant issue within coal washeries is that of over frothing; such problems can occur with pumps, sumps and thickeners when the level of frother in the process water rises too high. The industry has long needed the means to monitor frother levels within the plant water in order to minimise frothing problems within the washery.

The object of this project was to develop a unit capable of routinely monitoring the frother concentration levels in plant water and be accurate enough to allow plant operators to easily determine if these concentration levels were increasing. They could alter addition rates and thereby keep frother levels within the safe operating zone. In fact, the ability to monitor frother levels has a two fold application, firstly, to ensure that concentration levels do not rise to a point where process units are adversely affected, and secondly, to allow plant operators to keep dosing rates of frother at optimal levels, taking into consideration the variable feed and  assumed percentage of fines to flotation.

The most common frother used in Australia with coal preparation plants is MIBC, methyl isobutyl carbinol, a secondary alcohol. Its formula is C₆H₁₄O and has a flash point of 42^oC. As an alcohol it should respond to alcohol detectors, commonly used by law enforcement agencies to test drivers, i.e. ‘breathalyser’ units. The similarities are obvious; the breathalyser units are used to estimate the level of ethanol in a person’s blood stream by determining the percentage of alcohol in a person’s breath and using known relationships between blood alcohol levels and alcohol levels in the breath, determine a persons’ blood alcohol level.

There are a number of different alcohol meters available on the market. Two types were selected for initial testing as the most prospective, and then further tested to determine which was the most suitable for achieving the projects objectives. The types are based on

• Semi-Conductor unit
• Fuel Cell unit

Initial testing showed that the semi-conductor unit gave the best response. The fuel cell unit with the existing factory setup was insensitive to MIBC, even though it is set up to respond to ethanol. The secondary alcohol nature of MIBC seems to limit its response.

Initially, a laboratory system was setup to test the concept.
Based on the test results it was decided to construct a unit to be taken onto a washery site.  A test program was undertaken to characterize the operation of this unit. Solutions ranging from 0 ppm to 100 ppm were tested.

A calibration chart was developed for the plant unit. Figure A shows the chart developed based upon all the analysis.

 Fig A. Calibration chart developed under laboratory conditions

The site selected to undertake the test program was Bloomfield Coal Preparation Plant, NSW. It was selected for the following reasons

• Uses MIBC as frother
• Has some over frothing issues
• Processes a number of seams, varying the MIBC dose rate to accommodate the characteristics of each seam
• Good access to the water sampling point

Bloomfield CPP uses a Jameson Flotation Cell to process its fine coal. MIBC is dosed at various rates, depending on the coal type being processed. The Jameson Cell also has froth washing.

A series of tests were undertaken to determine if the analyzer response was the same as that determined in the laboratory. Test solutions were run through the unit and sensor outputs monitored. The test solutions were at various concentrations of MIBC in water prepared at CSIRO Newcastle, the concentrations were 0 ppm (blank), 10 ppm and 20 ppm.

Figure B shows the performance of the unit on 25/1/06. During this test both plant water and calibration solutions were processed. The figure shows that the unit gave a stable response for the plant water over almost a three hour period, noting that the MIBC level in the plant water rose from 5 ppm to 8 ppm over that period.

 Figure B. Response of analyser for plant water and calibration solutions

Subsequent tests showed that the sensor is sensitive enough to repond to changes in the dosing rate of MIBC over a short period of time.

In order to confirm the MIBC concentrations indicated by the breathalyzer reading, samples were taken during some of the test runs for later analysis in the CSIRO laboratories. Unfortunately, problems associated with extended storage period of samples and probable absorption of MIBC onto ultrafine coal and clays in the plant water samples meant that these samples could not be used to corroborate MIBC determinations based on sensor readings.

As to the performance of the system, the following points are worth making

• The sensitivity and reproducibility are sufficient to monitor MIBC concentrations in plant water. It is accurate to at least 1-2 ppm
• It can be operated in a semi-continuous mode; this would provide washery operators with warning of increasing frother, MIBC, concentrations.
• If the washery water could be filtered without removing frother from the solution, then any part of the washery system could be monitored for frother (MIBC) concentrations.
• The system is simple and can be developed to be robust for washery operations.

 

With regard to developing a commercial unit for the industry, negotiations are currently underway with a supplier. A number of issues need to be addressed, such frequency of calibration and stability over prolonged use. It is expected that all these issues will be dealt with quickly. All the components used in our prototype are low cost, the sensors used are a mature technology with low unit costs and readily available, therefore the cost of units will be low.

Finally, the coal industry now has the opportunity to precisely control the flotation system by monitoring the frother concentrations in the re-circulating plant water, and control the concentrations so as to alleviate any over frothing problems within the plant, such as in the pumps, sumps and thickener. This system may also assist in monitoring frother concentrations in the flotation system itself which together with the monitoring of the plant water should result in a more effective control of the flotation system.