In the previous blog about mineralogical monitoring for the aluminum industry, we discussed the added value of accurate on-line and at-line mineralogical monitoring for bauxite ore. The advantages of X-ray diffraction (XRD) and near-infrared spectroscopy (NIR) that provide information about the mineralogical composition and important process parameters of bauxite were discussed. This information is not only important for bauxite mining but also for efficient downstream processing in the alumina refinery. 

Step II. Alumina refining 

Process 

Alumina refineries process bauxite ore to produce alumina, which is then used to extract aluminium metal. Alumina (aluminium oxide) is a white granular material.  

Figure 1. Alumina refining process (image courtesy of Australian Aluminium Council Ltd)

The process to produce alumina from bauxite ore is called the “Bayer process”, developed by Carl Josef Bayer in 1888 (figure 1). It consists of four steps: digestion, clarification, precipitation and calcination.

After milling, bauxite is mixed with caustic soda (sodium hydroxide) under high temperature and pressure. Alumina dissolves from the aluminium bearing phases (excluding clays). Undissolved impurities settle down as a fine red mud, which after few recycling steps, is discarded as waste.

The solution of alumina in caustic soda (liquor) goes through further clarification, filtration and precipitation steps. Alumina crystals are recovered from the caustic solution by mechanically stirring the solution in open-top tanks. The precipitated material (called hydrate) is washed and dried at temperatures exceeding 1000°C. The dry white anhydrous aluminium oxide powder (alumina) is cooled and conveyed to storage. Alumina powder is further used to extract metallic aluminium using electrolytic baths. Caustic soda is recovered and returned to the start of the process and used again.

Let’s discuss the effect of mineralogy on the above process. Temperature required for the digestion of diaspore α-AlO(OH) and boehmite γ-AlO(OH) (both called monohydrate alumina phases, MHA) is higher than for gibbsite γ-Al(OH)3 (called trihydrate alumina, THA). Therefore, the temperature for effective digestion of bauxite depends on the ratio between the different aluminium containing mineral phases.

In addition, the consumption of caustic soda per ton of bauxite depends on the amount of silica impurities: clays and quartz. Under certain conditions, these minerals react with caustic soda and consume part of the reagents from the process. Low-temperature digestion suffers from the reagent loss to clays only, but during high-temperature digestion, both quartz and clay minerals react with caustic soda, increasing reagent consumption and costs.

Therefore, the knowledge about the mineralogical composition of bauxite is an important factor that defines the efficiency of the Bayer process.

Analysis of alumina for quality control (QC) and quality assurance (QA) 

Mineralogical monitoring not only adds value for the analysis of the raw material bauxite but also for the quality control of the final product from the Bayer process, alumina. XRD is the only suitable tool to distinguish between the different modifications of alumina (eg. α-Al2O3, γ-Al2O3) which defines the quality of dry alumina powder, as well as particle size and impurities.

The knowledge of the different modifications is important to predict and optimize the behavior during the smelting process. γ-Al2O3 is desired for the electrolysis since it dissolves more easily during the smelting process than α-Al2O3. For that reason, the ratio of the different sub-α and α-Al2O3 modifications of alumina must be monitored. The analysis of α-Al2O3 can be done with a classical straight-line calibration of the α-Al2O3 peaks or using full pattern fitting methods. The advantage of using the full information of the XRD pattern is the simultaneous quantification of all sub-α-Al2O3 modifications. Even a small fraction of 0.5 wt.% α-Al2O3 can be detected and quantified. Figure 2 shows a measurement of dry alumina powder using X-ray diffraction. The majority of the sample consists of g-alumina, with only 0.5% of a-alumina. 

Figure 2. Automatic γ-Al2O3 and α-Al2O3 quantification of dry alumina using Aeris Minerals, measurement time is 10 minutes.

Impact of particle size 

Particle size impacts directly on the rate of dissolution of the alumina in the cryolite bath and is therefore another important variable. Furthermore, fines are an issue from both health and safety as well as a product transport point of view, so particle size distribution needs to be carefully controlled. The ideal particle size distribution is defined between 45μm and 150μm to prevent problems with dissolution in the cryolite bath and the accumulation of fines during processing that cause conveying and process instabilities and health issues. 

Figure 3 Typical display of on-line alumina process data

The application of on-line particle size analysis allows aluminium processors to operate more efficiently and to produce a more consistent product for downstream unit operations. Economic benefits in the form of reduced waste, reduced energy consumption, reduced manpower and increased throughput are achieved. The availability of industrially relevant systems, at a cost that can be recovered in a relatively short time, makes on-line analysis an increasingly attractive option.

Red mud – monitoring of waste products using XRD 

Red mud is the bauxite residue generated during the refinement of bauxite into alumina using the Bayer process. It is composed of various oxide compounds, including iron oxides which give its red color.

This image has an empty alt attribute; its file name is Mine-1.jpg

Depending on the bauxite grade about 1.3 to 2.2. tons of red mud are produced per ton of alumina. The treatment, long-term storage and re-use of the red mud is one of the challenges of alumina refining. The adequate analytical tools for the control of mineralogy, chemistry and physical properties of red mud are essential for the correct, environmentally friendly management of the waste product.

The complete mineralogical composition of the red mud sample determined by XRD is shown in Figure 4. The main part of the sample consists of the insoluble impurities of bauxite (titanium and iron oxides) and products of chemical reactions, which occur during the digestion steps. However, in this example, valuable aluminium bearing phases such as gibbsite, boehmite and diaspore still comprise about 25% of the red mud sample. This indicates that the processing, in particular the digestion, was not executed to the maximum level of efficiency.  

Figure 4. Automatic quantitative mineralogical analyses of red mud using Aeris Minerals. Measurement time is 15 minutes.

Comprehensive analysis of bauxite and red mud mineralogy along with the analysis of digestion conditions should be performed to understand the root cause for the lost efficiency.  

Value of analytical monitoring of mineralogy and particle size for the Bayer process 

Alumina refining is a complex process the efficiency of which is directly related to the mineralogy of the bauxite ore. Fast and accurate mineralogical information throughout the entire process, speed of the XRD analyses, increased safety for the operators and the possibility to automate makes XRD a reliable and economic alternative compared to the traditional analytical methods such as wet chemistry and does justify the initial investment. 

To learn more about the benefits of alumina refinery process control by XRD, watch our webinars on-demand discussing evaluation of XRD technique for this application and bauxite case study using Aeris Minerals

In our next blog, we will discuss the added value of mineralogical analyses for the next step in the bauxite-to-aluminum process: the conversion of refined alumina to aluminum metal.