Mercury removal from Bayer refinary condensate

Mullett, M 2006, Mercury removal from Bayer refinary condensate, Masters by Research, Applied Sciences, RMIT University.

Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
mullett.pdf Thesis application/pdf 5.05MB
Title Mercury removal from Bayer refinary condensate
Author(s) Mullett, M
Year 2006
Abstract The alumina refining industry is one of a number of global industries that produce process waters containing mercury. The research program detailed in this dissertation focuses on processes for removing mercury from Bayer digestion condensate streams.

Bayer refinery digestion condensate is chemically unique because it exhibits high pH and is chemically reducing, thermodynamically favouring the presence of elemental mercury. The temperature of condensate is typically 95 oC, with flow rates up to 490 kL/hour. On the basis of these unique conditions, a literature review of mercury removal processes was conducted.

Of the many processes reviewed, three broad process options were selected for further examination. These were:

• Adsorption onto activated carbon or other suitable materials

• Sparging mercury into a gas stream

• Complexation or amalgamation onto modified silica substrates using noble metals or sulphur functional groups as impregnates A variety of activated carbon types were tested using batch, up-flow column or filtration tests.

These carbon types included a granular virgin activated carbon (VAC); granular and extrudate sulphur impregnated activated carbon (SIAC); an extruded silver impregnated activated carbon (AgAC) and a combination of two powdered virgin activated carbons (PAC).

The adsorption mechanism for both the VAC and the SIACs was physical and as such the mercury capacity of the carbon was determined to be a function of the concentration of mercury in solution. The sulphur impregnated into the SIAC did not increase the mercury capacity
because of sulphur dissolution into the condensate. The removal efficiency and mercury loading demonstrated by these carbon types was poor and did not justify further investigation.

This was primarily because the adsorption and chemisorption processes were interfered with by the digestion condensate matrix.

The PACs were tested in batch mode and were also incorporated into a filter bed, through which condensate was passed. The PACs demonstrated a superior mercury loading and a higher removal efficiency compared to the VAC and SIACs. The volume of condensate required to achieve mercury breakthrough was not determined and it is recommended that further tests be conducted to determine this. In addition to the high removal efficiency
demonstrated by the PACs, the short residence times of <30 seconds associated with the filtration procedure, justifies further investigation as a mercury removal process option.

The AgAC demonstrated the highest mercury loading of the granular and extruded carbon types. This may be attributed to the chemisorption or amalgamation of elemental mercury and it is recommended that mercury removal via amalgamation be investigated using a high surface area form of suitable metals such as silver and tin.
A sulphur impregnated silicate material was also tested using batch and up-flow column tests. As was the case for the SIACs, the impregnated silicate material suffered from sulphur dissolution in the condensate. This material is therefore not recommended for further investigation.

Each of the impregnated mesoporous silica based materials, except for the gold impregnated silica, demonstrated poor stability in condensate and should not be considered for further assessment or implementation. Although the gold impregnated silica was stable, and demonstrated effective mercury removal, it would be too expensive to justify for use on an industrial scale unless a process for regenerating the gold was developed. It did, however, further demonstrate the opportunity for tests involving mercury amalgamation.

3 Gas sparging was the most viable opportunity with ~80% mercury removal demonstrated. The system involves steam sparging to volatilise mercury from the solution phase to a vapour phase, from which the mercury is compressed and re-condensed for collection. It was estimated that the system would cost approximately 4 million dollars to install with annual operating costs of approximately $500 000.
Degree Masters by Research
Institution RMIT University
School, Department or Centre Applied Sciences
Keyword(s) alumina
Version Filter Type
Access Statistics: 417 Abstract Views, 2484 File Downloads  -  Detailed Statistics
Created: Fri, 26 Nov 2010, 16:10:12 EST
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us