Gabriella Sipos 1, Michael Shaw 2, Ulrich Seydel 3, Gordon Parkinson 1, Anthony McKinnon 2, Peter Smith 4 and John Kildea 5

1 A. J. Parker Cooperative Research Centre for Hydrometallurgy,

School of Applied Chemistry, Curtin University of Technology,

GPO Box U 1987, Perth 6845, Western Australia, Australia

2 Alcoa of Australia Limited

3 Centre for Microscopy and Microanalysis UWA

4 CSIRO Division of Minerals

5 Nalco Australia Pty Ltd


Sodium oxalate is one of the many organics present in Bayer liquor. Due to its limited solubility, sodium oxalate can co-precipitate with alumina trihydrate during precipitation. This can have detrimental effects on the final product quality especially if it occurs in the initial stages of precipitation.

Quaternary amine type cationic surfactants can prevent sodium oxalate co-precipitation and increase the tolerable concentration of sodium oxalate in Bayer liquor. Their action is via the inhibition of nucleation or/and the inhibition of crystal growth. This paper presents work detailing the effect of quaternary amines on sodium oxalate crystal growth in Bayer liquor.

The results show that while quaternary amines inhibit crystal growth in Bayer process liquors, they have no effect on crystallization in synthetic liquor. It is postulated that the presence of certain organic molecules in the process liquor are required for quaternary amines to inhibit crystallization and therefore stabilize the liquor. Humic material obtained from Bayer liquor has been shown to be effective in this role. Commercial humic acids have been tested and compared with plant humates.

Adsorption isotherms of octyltrimethylammonium bromide (OTAB) and plant humates have been successfully measured in Bayer liquor. Investigations with OTAB and plant humates reveal a synergy between the two. From a mixture of quaternary amine and plant humates, an enhanced adsorption of both components was observed. Confocal laser scanning microscope (CLSM) images of plant humates adsorbed on sodium oxalate crystals are also presented.


octyltrimethylammonium bromide (OTAB), oxalate, Bayer, humic, adsorption, crystal growth, confocal laser scanning microscope (CLSM)



Gabriella Sipos, Michael Shaw, Ulrich Seydel, Gordon Parkinson, Anthony McKinnon, Peter Smith and John Kildea


Organics from bauxite can be decomposed to form soluble organic compounds in the Bayer process. One of these degradation products is sodium oxalate. The sodium oxalate may accumulate to concentrations above its solubility in the liquor due to liquor recycling. When the sodium oxalate concentration builds up to a critical supersaturation in the plant, it can co-precipitate with gibbsite. This co-precipitation seriously affects the aluminium trihydrate product quality. For this reason, most refineries apply oxalate removal in one way or another.

Concentrations of sodium oxalate can be maintained above its solubility due to the stabilizing influence of high molecular weight humics present in Bayer liquor (Lever, 1983).

The operational requirements for oxalate solubility are different in different areas of the Bayer circuit. In Alcoa refineries it is desirable to have a high dissolved oxalate concentration tolerated during the initial part of precipitation, whilst a low solubility is required during oxalate removal to make the removal as efficient as possible.

1.1 Methods for Sodium Oxalate Control in Bayer Liquors

It has been shown that the removal of a small fraction of the high molecular weight humates from the Bayer process liquors destabilizes the dissolved sodium oxalate and results in its spontaneous crystallization (Lever,1981). Several methods of overcoming the stabilizing effect of these high molecular weight humates have been patented. U.S Patent 3,337,305 (Byrns,1976) describes a method in which sodium oxalate is precipitated from solution by the addition of aqueous ammonia. In U.S. Patent 3,649,185 (Sato, 1972), sodium oxalate precipitation is affected by the addition of NaOH, thus decreasing the solubility of sodium oxalate. U.S.Patent 3,899,571 (Yamada, 1975) discloses a method in which seeding with activated crystals of sodium oxalate stimulates precipitation.

Some of the patents disclose methods which target directly the removal of high molecular weight organics responsible for the stabilising effect of sodium oxalate. U. S. Patent 3,457,032 (Breteque, 1969) describes a method involving passing Bayer liquor through a bed of a strongly basic anion exchange resin to eliminate organic impurities. U.S. Patent 4,275,042 (Lever, 1981) is about the removal of sodium oxalate from supersaturated solution of Bayer spent liquor by the treatment with cationic sequestrants and U. S. Patent 4,275,043 (Gnyra, 1981) refers to similar treatment through the use of activated carbon.

There are already two patents registered about the stabilising effect of synthetic organics. Cationic surfactants can be used to stabilise sodium oxalate further in Bayer process liquors. U. S. Patent 5,385,586 (Farquharson, 1995) refers a method for inhibiting the precipitation of sodium oxalate by the addition of an alkylammonium salt. AU. Patent AU9733241 (Farquharson,1998) describes a process using a mixture of surfactants to achieve high dissolved oxalate concentrations.


1.2 Synthetic stabilizers

Quaternary amine salts have been found to be very powerful oxalate stabilizers in Bayer liquors (Farquharson,1995). From the wide range of quaternary amine salts tested, an appreciation of the relationship of compound structure to liquor oxalate stabilization capacity has emerged. This structure-activity relationship has greatly assisted new product development, but the mechanism by which these additives work is poorly understood (Farquharson,1996).

In the work described in this paper, the effects of quaternary amines, octyltrimethylammonium bromide (OTAB), dodecyltrimethylammonium bromide (DTAB) and hexadecylpyridinium chloride (HPC) on sodium oxalate crystal growth have been investigated. To clarify the connection between the inhibitory effect of quaternary amines and the mechanism of their adsorption onto the oxalate surface, adsorption isotherms have been measured. To gain a mechanistic understanding of the stabilizing effect of quaternary amines in plant liquors, interactions with the high molecular weight humates have also been examined.

1.3 The Role of High Molecular Weight Organics

Our results confirm that high molecular weight plant humates have a significant impact on oxalate behaviour in the Bayer process. Their dramatic effect on oxalate morphology has been shown previously (Brown,1988). Due to the importance of high molecular weight humates in helping to stabilize oxalate, adsorption isotherms on needle-like sodium oxalate have been measured. The effect of plant humates on oxalate morphology, the distribution of humates adsorbed on oxalate crystals, the inhibition of oxalate crystal growth and their interaction with quaternary amines have been investigated.


2.1 Materials

Chemicals: Octadecyltrimethylammonium bromide 98% (OTAB), dodecyltrimethyl-ammonium bromide (DTAB), hexadecylpyridinium chloride (HPC) were obtained from Fluka Chemical Company. Humic acid and sodium humate were obtained from Fluka Chemical Company and Aldrich Scientific Company respectively.

All other chemicals used were of analytical grade.

Preparation of plant humic extract: Sodium oxalate obtained from Alcoa of Australia’s Pinjara refinery was collected, and the fraction of high molecular weight humic type material present in or on the solid oxalate was isolated. The concentrations of plant humates dosed into synthetic liquor are expressed as the mass of sodium oxalate from which the humic content is isolated (eg. 1g/L humic solution represents the humic isolated from 1 gram of sodium oxalate and dosed into 1L of synthetic liquor).

Adsorbent: Clean needle-like sodium oxalate was provided by Alcoa of Australia Limited. The surface area was determined by the BET method, and was found to be 1 m2/g. The average crystal size is 30 mm with an aspect ratio of 10:1.

2.2 Analysis

Oxalate concentrations of liquors were determined using the method developed by Grocott (Grocott,1988).

Quaternary amine concentrations were determined using a method developed recently (Hind,1997). The method involves extraction of the iodide ionpair of the quaternary amine into dichloromethene. These solutions are then analysed for quaternary amine using capillary gas chromatography. Typical measured concentrations were between 0-8x10-5 mol/L (0-30ppm).

Humic concentrations in synthetic liquor were determined using a UV-VIS spectrophotometer. Absorbances of the humic liquor solutions were determined using a 1 cm or 10 cm pathlength cuvette to record the spectra in the range of 290-560 nm.

CMCs (critical micelle concentrations) of plant humates and OTAB were obtained using surface tension measurements carried out using an Analite Surface Tension Meter (Selby Scientific) and Marienfeld plates. The CMCs were determined from the plot of the surface tension of liquor versus the surfactant concentration. The CMC was found to be 2.25 g/L for the high molecular weight humates and 3.8x10-6 mol/L for OTAB in synthetic liquor.

Confocal Laser Scanning Microscopy. Crystals were examined using a Bio-Rad MRC-1000 confocal laser scanning microscope (CLSM). The CLSM was equipped with a Krypton - Argon laser. Images were collected using the 488 nm line of the laser and a 515 nm barrier emission filter. Either a 4 x lens (plan apo, Nikon) with a numerical aperture (NA) of 0.25 or a 20 x lens with a NA of 0.75 was used. The pixel sizes were 3.9 Ám2 for the 4x lens and 0.8 Ám2 for the 4x lens. Optical sections (Z-series) were collected with a stepping size of 10 Ám for the 4 x lens and 2 Ám for the 20 x lens. Z-series were projected using the software "Confocal Assistant" (by T. Breijle). The maximum algorithm was employed resulting in extended focus images.

2.3 Conditions

Synthetic liquor used contained the main inorganic components of plant liquor (hydroxide, carbonate, chloride, sulphate, alumina, silica and phosphate) with a low A/TC (0.3) to ensure the liquors were stable with respect to gibbsite precipitation. The liquors were equilibrated with sodium oxalate prior to use.

Temperature. All of the experiments were carried out at 60 oC.

2.4 Procedures

Crystallization tests developed by Alcoa of Australia Limited were used. These tests involved equilibrating oxalate seed crystals in oxalate saturated synthetic liquor containing the test organics of interest (ie humates, quaternary amine or humates and quaternary amine). Once equilibrated, the solution was supersaturated with sodium oxalate and crystallization proceeded. After a one hour crystallization time, the soluble oxalate concentration was measured. The difference between the oxalate concentration at the start and end of crystallization gave the yield. The stronger the crystal growth inhibition, the lower the yield.

An adsorption isotherm procedure for OTAB and plant humates was developed. The synthetic liquor was saturated with sodium oxalate at 60 oC and filtered. Liquor (10.0 mL) in glass culture tubes was conditioned with 5 g/L needle-like sodium oxalate for one hour at 60oC in a rotating water bath. The liquor was dosed with different volumes of a stock solution of the adsorbate. After 20 hours adsorption time, which was found to be sufficient to reach equilibrium, the tubes were centrifuged at 60 oC for 10 minutes and the liquor sampled into another tube for analysis. After decanting the liquor, the remaining solid in the original tube was dissolved with water and the adsorbate content determined.


3.1 Inhibitory Effect of Quaternary Amines under Plant Conditions

Crystallization tests were carried out to establish the effect of quaternary amines on sodium oxalate crystal growth using plant liquor. Figure 1 shows the test results at different doses of the DTAB, OTAB and HPC.

Figure 1

Oxalate yields measured in plant liquors dosed with different concentrations


Oxalate yields in the presence of quaternary amines are decreased due to the inhibitory effects of these chemicals. The crystal growth inhibition is proportional to the concentration of amines. Quaternary amines are powerful crystal growth inhibitors in plant liquor with OTAB being the strongest inhibitor completely preventing yield at a concentration of 1x10-5 mol/L.

3.2 Inhibitory Effect of Quaternary Amines in Synthetic Liquor

To compare the action of quaternary amines in plant liquor the crystallization tests were repeated using synthetic liquor. It was found that all the above quaternary amines had no effect on oxalate yield in synthetic liquor. Yields in the presence of quaternary amines did not differ from the control yield.

3.3 Effect of Quaternary Amines with Plant Humates Present

The above results show that quaternary amines do not inhibit oxalate crystal growth in synthetic liquor, but do in plant liquor. The reason that quaternary amines inhibit crystal growth in plant liquor is probably due to synergy with some organic compounds found in plant liquor. As it is known that plant humates affect oxalate crystallization they were investigated as possible compounds. The effect of the presence of quaternary amines on oxalate yield was measured in synthetic liquors dosed with plant humates. Figure 2 shows the effects of quaternary amines on oxalate yield in the presence of plant humic materials in synthetic liquor.

Figure 2

Oxalate yields measured in the presence of plant humates at different concentrations of DTAB,

OTAB and HPC in synthetic liquor

These results show that quaternary amines at concentrations of 1.5 and 3.0x10-5 mol/L in synthetic liquor alone have no effect on oxalate yield. Humic material on its own affects oxalate yield, with the 5 g/L concentration giving an oxalate yield which is of 25% of that found in pure synthetic liquor. Combining quaternary amines with plant humates results in increased inhibition compared to humates on their own. This effect is dependent on both the humic and the quaternary amine concentration, with maximum inhibition occurring at higher quaternary amine and humic concentrations. These results show that there is a synergistic interaction between plant humates and quaternary amines: together they make a strong oxalate crystal growth inhibitor.

3.4 Effect of Commercial Humics

Commercial (non-Bayer) humics were tested and had an impact on oxalate yield, but the synergistic effect with quaternary amines only appeared after the degradation of these materials at high temperature (95 oC) in liquor. The chemical nature of the degradation on the commercial humics has not been identified. It is postulated however, that the effect of the degradation is to produce lower molecular weight humic material which may contain more carboxylate, hydroxyl functional groups.

3.5 Adsorption Isotherm of OTAB

Further detailed investigations were carried out using OTAB, which was found to be the strongest crystal growth inhibitor of the compounds tested in plant liquor.

Based on crystallization test results in synthetic liquors, it was expected that OTAB would not show any adsorption on the surface of sodium oxalate, because it did not show any inhibitory effect on oxalate crystallization in pure synthetic liquor. However, an adsorption isotherm of OTAB in synthetic liquor was successfully generated on needle-like sodium oxalate (Figure 3). The isotherm was found to conform to Langmuir type adsorption behaviour.

Figure 3

Adsorption isotherm of OTAB on needle-like sodium oxalate measured in synthetic liquor at 60oC in the presence of 5 g/L solid oxalate. The line represents the Langmuir function calculated with the parameters determined

This behaviour indicates a high affinity of adsorbate towards the surface. The equilibrium constant or adsorption affinity (K) and the adsorption capacity (S(max)) were obtained by plotting S-1 (the reciprocal surface concentration) versus x-1 (the reciprocal equilibrium concentration) of the OTAB.

K, adsorption affinity: 8.4 x 106 L/mol

S(max), adsorption capacity: 5.6 *10-6 mol/m2

Figure 3 shows that the Langmuir function calculated with the determined parameters correlates well with the experimental data. Calculated from the adsorption capacity, the area occupied by one molecule at monolayer coverage corresponds to ca 30 ┼2/molecule. This value is consistent with that found in the literature for similar compounds (Rubingh, 1991) and suggests that molecules adsorb perpendicularly to the surface.

3.6 Adsorption Isotherm of Plant Humates

Plant humates were shown to have an impact on oxalate yield without the presence of quaternary amines (Figure 2). Thus, it was expected that significant adsorption of plant humates would occur on the oxalate surface. Figure 4 shows the adsorption isotherm for plant humates measured at different wavelengths.

Figure 4

Adsorption isotherm of plant humates on needle-like sodium oxalate measured

at different wavelengths in synthetic liquor

Different apparent extents of adsorption were detected when the solution concentration was measured by using UV-VIS spectrophotometry at different wavelengths. The higher the wavelength, the higher the apparent level of the adsorbed amount of plant humates. Plant humates are mixtures of compounds, and some of the components probably adsorb preferentially onto the oxalate surface. The absorbance of humates is dictated by size and degree of functionalities. It is assumed that the adsorption measured at higher wavelengths represents the behaviour of the higher molecular weight components of the mixture. This would explain why different extents of adsorption are measured at different wavelengths

In the adsorption isotherms, a "knee" in the curve was observed close to the measured CMC for humates in synthetic liquor. This indicates that adsorption above this value probably consists of micelles of humates rather than single species adsorption.

3.7 Adsorption from Mixtures of OTAB and Plant Humates

Both plant humic extract and OTAB have been shown to individually adsorb onto sodium oxalate from synthetic liquor. However, the stabilizing effect of OTAB on oxalate crystallization was observed only when both humic material and OTAB were present together. The simultaneous adsorption of both OTAB and plant humates from the same solution were measured (constant plant humate concentration, varying OTAB concentrations). It was found that these materials enhanced the adsorption of each other. Figure 5 shows the amount of plant humates adsorbed on needle-like solid oxalate (measured at 290 nm) as a function of the amount of OTAB adsorbed onto the same oxalate surface at the same time in synthetic liquor. The dotted line indicates the adsorption capacity for OTAB measured in the absence of plant humates.

Figure 5

A plot of the amount of plant humates adsorbed on needle-like solid oxalate as a function of the amount of OTAB adsorbed onto the same oxalate surface in synthetic liquor

The OTAB adsorption isotherm showed that a plateau occurred at 5.6 mmol/m2 (Figure 3). Figure 5 shows that significantly more than 5.6 mmol/m2 can be adsorbed when the plant humic material is present. Similarly, the amount of plant humates adsorbed is significantly greater (5x) in the presence of OTAB. This suggests a mechanism of co-adsorption, whereby the OTAB and humates interact together at the oxalate surface. It is possible that this interaction is an electrostatic one. The overall result is a more hydrophobic (surface active) molecule which has a greater propensity to adsorb and block growing sites on the oxalate crystal.

The enhanced adsorption of plant humates in the presence of OTAB may be the clue to the stabilizing effect of quaternary amines measured in plant liquors. Quaternary amines probably do not play an active role in the stabilization of sodium oxalate by poisoning the oxalate seeds, because they were shown to have no effect on oxalate crystallization without plant organics present in synthetic liquor. Rather, their action is to interact with plant organics and to attract excess organics to adsorb on the oxalate surface.

3.8 CLSM Images of Adsorbed Plant Humates

Confocal Laser Scanning Microscopy is commonly used to image biological samples, in which either naturally occurring or added fluorescent molecules are excited by the powerful monochromatic laser light, and the emitted fluorescence light is detected. Applying this technique to our samples gives us the possibility to track the humates adsorbed on the surface of needle-like oxalate or incorporated in oxalate crystals produced from solutions containing humic material. With this microscope, we can see the humic material only, because it fluoresces under the microscope, and we can not see the sodium oxalate crystals. The digital imaging capability of CLSM allows the distribution of the humates to be seen inside the crystals by mapping layers at a 10 mm depth.

Figures 6a and 6b show the adsorbed plant humates on needle-like sodium oxalate. The brighter image in the presence of OTAB (6b) indicates an enhanced adsorption of humates on the surface. The adsorption of humates occurs all around the crystal (6c), and patches of plant humates can be recognised on the surface (6d).

Figure 6

Confocal images of plant humates (a),(c) and (d) adsorbed on needle-like sodium oxalate from 10 g/L liquor solution, (b) adsorbed in the presence of 9.0 x 10-5 mol/L OTAB

Usually, the needle-like sodium oxalate is produced under plant conditions. However, from synthetic liquor containing only humic material, sodium oxalate precipitates out in a ball-like morphology applying very low supersaturation and room temperature (Figure 7a). The ball-like oxalate crystals become doughnut-like in the presence of OTAB (7b). Mapping the crystals layer by layer, the humates show a very even distribution inside the oxalate crystals.

Figure 7

Confocal images of plant humates (a) incorporated in oxalate crystals produced from 10 g/L liquor solution, (b) incorporated in oxalate crystals produced from 10 g/L liquor solution in the presence of 6.0 x 10-5 mol/L OTAB



Quaternary amines have been shown to be powerful crystal growth inhibitors in Bayer process liquors. OTAB was shown to be the strongest inhibitor of those investigated. Whilst strong inhibitors in plant liquors, none of these chemicals showed inhibitory effect in synthetic liquor.

Plant humates have been measured to be strong oxalate poisons in the absence of quaternary amines in synthetic liquor, however, an increased inhibitory effect (stabilizing effect) is observed in the presence of quaternary amines. It was also shown that this stabilizing effect increased with increasing concentrations of both quaternary amine and plant humate.

Commercial humic acids did inhibit crystal growth of sodium oxalate and, surprisingly, no synergy with quaternary amines was found. After degradation of these materials at high temperature in synthetic Bayer liquor, the degraded humates did show synergy with quaternary amines.

Adsorption isotherms of plant humates and OTAB have been measured individually. OTAB adsorption follows a Langmuir behaviour on needle-like sodium oxalate with 8.4x106 L/mol adsorption affinity and 5.6 x10-6 mol/m2 adsorption capacity.

Adsorption isotherms of plant humates show a continuous increase of humates adsorbed on the oxalate surface with a knee occurring at the CMC measured for plant humates. Confocal images show that humates adsorb all around the crystal, and patches of humates can be recognised on the surface at concentrations above the CMC. The adsorption of plant humates is thought to be molecular weight dependent as demonstrated by UV-VIS, and reflects the heterogeneous nature of the humic material. Despite the fact that both the OTAB and humic material adsorb, only the humic material was shown to be an effective crystal growth inhibitor.

The simultaneous adsorption of OTAB and plant humates from the same solution have shown an enhanced adsorption of both components. Significantly larger amount of both plant humates and OTAB can adsorb on the oxalate surface from the mixture than from solutions of the individual components. The enhanced adsorption of humates can be seen on the confocal images in the presence of OTAB. The increased adsorption of both OTAB and plant humates is probably due to an electrostatic interaction between OTAB and plant humates occurring on the oxalate surface. This interaction promotes excess humates to adsorb on the surface, and to enhance the inhibitory effect in the presence of quaternary amines.


The authors are grateful to the many people who have contributed to the work reported in this paper. Particular thanks to the staff of the Research and Development Department of Alcoa of Australia Limited. Financial support from Nalco Australia Pty Ltd, Alcoa of Australia Pty Ltd and Minerals and Energy Research Institute of Western Australia (MERIWA) is gratefully acknowledged.


Breteque, Pierre de La. (1969). Process for Purifying Solutions Containing Aluminates. U. S. Patent 3,457,032 1969 July 22.

Brown, N. (1988). Crystallization of Sodium Oxalate on Metal Surfaces. Journal of Crystal Growth Vol.87. pp.287-294.

Byrns, A. C. (1976). Purifying Caustic Aluminate Solutions with Ammonia" U. S. Patent 3,337,305, 1976 August 22.

Farquharson, G.J., Kildea, J.D., Gross, A. E., Grocott, S. C.(1995). Liquor Oxalate Stabilizers. U. S. Patent 5,385,586, 1995 January 31.

Farquharson, G.J.,Kildea, J.D. (1998).Bayer Liquor Oxalate Stabilizer. AU. Patent AU9733241, 1998 February 12.

Farquharson, G.J., Gotsis, S., Kildea, J.D., Grocott, S. C. and Gross, A. E. (1995).Development of an effective liquor oxalate stabilizer. Light Metals pp95-101.

Farquharson, G.J, Gotsis, S., Kildea, J.D. and Grocott, S.C. (1996). The role of quaternary ammonium compounds in stabilizing sodium oxalate in Bayer liquors. Proceedings of the Fourth Alumina Quality Workshop, Darwin.

Gnyra, B. (1981). Removal of Oxalate from Bayer Process Liquors. U. S. Patent 4,275,043, 1981 June 23.

Grocott, S.C. (1988). Bayer liquor impurities: measurement of organic carbon, oxalate and carbonate extraction from bauxite digestion. Light Metals pp.833-841.

Hind, A.R., Bhargava, S.K. and Grocott, S.C. (1997). Quantitation of alkyltrimethylammonium bromides in Bayer process liquors by gas chromatography and gas chromatography-mass spectrometry. Journal of Chromatography A, vol.765, pp.287-293.

Lever, G. (1981). Removal of Oxalate from Bayer Process Liquors. U.S. Patent 4,275,042, 1981 June 23.

Lever, G. (1983). Some aspects of the chemistry of bauxite organic matter on the Bayer process: the sodium oxalate-humate interaction. Travaux Vol. 13. No.18. pp.335-347.

Yamada, K. et al. (1975). Method for the Removal of Organic Substances from Sodium Aluminate Solution. U. S. Patent 3,899,571, 1975 August 12.

Rubingh, D. N. and Holland, P. M. (1991). Cationic Surfactants: Physical chemistry Surfactant Science Series: Volume 37, pp110-112. New York and Basel, Marcel Dekker Inc.

Sato, C. et al. (1972). Method for Removing Impurities in the Bayer Process. U. S. Patent 3,649,185, 1972 March 14.