HRP920433A2 - A process for regulating the tar content of anodes intended for the uction of aluminium by electrolysis - Google Patents

Description

The invention relates to a process for regulating the tar content in anodes intended for obtaining aluminum by electrolysis according to the Hall-Heroult process.

Description of the problem

Tkz. "Pre-baked" anodes are obtained by hot mixing of aggregates (most often coke) and binder (most often tar) and shaping of the resulting carbon dough by vibration hardening, pressing or vibration pressing and baking at elevated temperature (1100/1200°C).

The quality of the anodes has a decisive influence on the satisfactory performance and energy efficiency of electrolysis and represents one of the constant concerns of operators, which is particularly reflected in their efforts to optimize production in such a way as to try to achieve the highest density of the baked product. So, the main concern is to extend the duration of the anode cycle compared to the electrolytic tank and to reduce the number of blocks that must be made for equivalent carbon consumption.

The density of the baked product essentially depends on three parameters: the dry density of the raw product, the effectiveness of the binder during baking, and the volumetric changes that occur during baking.

These three parameters are interdependent and change depending on the binder content in the raw product.

Experience shows that for a given mixing of raw materials (coke, tar), production conditions (granulation, mixing, etc.) and shaping of the dough, the baked density depends on the binder content in such a way that it reaches its maximum at the content that allows obtaining the maximum dry density of the raw product.

The technical basis of the problem

The study of the structure of carbon anodes was greatly influenced by soil mechanics.

From the mechanics of multiphase media, we know that densification of soil with variable water content causes a dry density curve that has a maximum, and it is desirable that dry density be defined as the density of dry matter in the densified media.

As the densification pressure (static or dynamic pressure) increases, so does the maximum density, and the corresponding water content decreases.

This result is explained by the saturated water content, and at the given content and pressure, the density level decreases due to the grains of solid material that are in the liquid.

Taken as a whole, the curves created by changing the densification pressure lead to a curve called the saturation curve, which for a given water content determines the highest density that is likely to be reached (SUB-BARAO, PhD thesis, Grenoble 1972, "continuous densification and mechanics of materials ").

Use with carbon dough

Thickening of carbon products can be treated like soil (or similar to soil) if they behave similarly:

- gas medium: air

- liquid agent: carbon binder (tar, resins)

- solid agent: carbon material (coke, anthracite ...).

But there is a big difference in the porous structure of the solid aggregate.

This porosity actually changes depending on the densification pressure. The consequence of this phenomenon is that the uniformity of the saturation curve is lost. In this regard, for a given tar content and an increase in densification pressure, the saturation density of the product can be improved, but this phenomenon is still of less importance than the effect of the binder level.

Deficiencies in the state of the art

Until now, the highest density was most often attempted to be achieved with a raw condensed anode. However, this adjustment procedure leads to a regulated increase in tar content, so this deficiency was regulated either visually, by inspecting the anodes when they exit the thickening unit (greasy appearance) or by a priori regulation of the range of tar content (for content outside that range it was known that will come from "sticking" between the anodes in the baking oven).

In this connection, many procedures for regulation were proposed. In the example of the ALCOA patent (FR 2 436 763 corresponding to US 4 133 090) it is done by separating the raw anodes at the moment they are removed from the thickening trough. After that, the tar content is regulated, in order to minimize or optimize yielding in relation to the density of the anode. This means that after changing the production parameters (raw materials, granulation, etc.), this optimization cannot be easily performed until a full insight into the properties of baked anodes is obtained, which takes several days and does not correspond to the production of raw anodes. That's why in practice we have to worry about the average yielding, by means of which we get closer to the optimal result.

The task of the invention

The task of this invention is to provide a procedure for regulating the production of raw anodes (by regulating the tar content), which will ensure that we can increase the density of baked anodes as much as possible, without waiting for the results of the baking process.

At a certain and constant level of petroleum coke quality, the actual density of baked anodes essentially depends on the dry density of raw anodes. This fact leads to effective densification of the dry anode material.

This dry density is related to the raw apparent density of the anode and the tar content by the equation:

DS = crude DA x (100 - % tar)/100,

where DS stands for the dry density of the raw anodes, raw DA the actual density of the raw anode (which was measured directly), % tar stands for the tar content expressed as a percentage of the weight in the dough.

For given combinations of raw materials and production conditions (granulation, mixing, etc.) and dough shaping, the dry density of the raw anodes when determining the tar content goes through the optimum.

Seen in a simplified way and without a specific scale, Figures 1 to 3 show the way in which the properties of the anodes develop in relation to the tar content.

Figure 1 shows the measurement of the dry density of the raw anode depending on the binder content as a parameter during pressure forming;

Figure 2 shows the measurement of the basic parameters of the anodes in relation to the tar content at a certain level of densification pressure;

Figure 3 shows a diagram of the percentage principle for regulating the tar content according to the present invention and

Figure 4 shows the practical use of the invention.

The optimal tar content, which allows to obtain the highest real density of baked anodes, is shown in Fig. 2. We will see that it corresponds to the density that allows to obtain the highest dry density of the raw anode. Therefore, this method, which is the subject of this invention, allows to improve the actual density of the baked anodes only by taking into account the measurements made on the raw anodes, which are taken into account the measurements made on the raw anodes, which are taken from thickening phase, and which we can directly use for regulation.

Furthermore, this optimal density is related and obviously changes together with the production parameters (raw materials, granulation, etc.). Therefore, in all cases except for transition periods, the optimization of the dry density will lead to the optimization of the baked anode density.

The procedure, which is the subject of this invention, includes the following excellent sequence of operations:

Fill the carbon dough mixer with:

- on the one hand, with ground coke with a predetermined granulation that we keep constant; and the amount of coke, which is supplied, must also be constant;

- on the other hand, with tar, whose Bo % content in relation to coke is changed either manually (operator) or automatically which we can program or which is served by a microprocessor.

After leaving the mixer, the carbon paste is placed in a thickening device, and the thickened anodes are spread on a table with rollers. If we use an uncontrolled mixer, we fill it with carbon dough of constant composition. Therefore, any change in the amount of tar will negatively affect the overall charge of N anodes, which were produced during that charge. If the mixer is of continuous type, there is a shift between the moment when the tar content changes at the beginning of the mixer, and the moment of the appearance of the first thickened anode, which is made from a dough of a different composition. This shift is marked with "d" (in practice it can represent 3 to 6 anodes). That is why we program the computer in such a way that it takes this fact into account.

The actual raw density is determined by the weight and volume of the anode.

We measure weight with errors within 0.14. Experience shows that dimensions such as length and width can be treated as constants after densification, with similar precision as tar contents that do not deviate by more than 1% (absolute) from the mean value. Therefore, to determine the volume of the condensed anode, it is sufficient to measure its height H.

We enter into the computer: the measured weight, length and width (which is constant but, if necessary, subject to proofreading) and the measured height together with the value of the tar content.

The result is as follows: raw DA = anode weight P/ (H-ho) x L x 1 + Vo, where Vo denotes the volume of the "face" of the anode, or of the front part with cut edges and anode connections, and the height of the anode front. The dry density of raw anode is: DS = raw DA x (100-% tar)/100. We enter this data into the computer.

The computer also takes into account the initial Bo% tar content which is determined by production experience; the dry density of the anodes, which were made under these conditions, is denoted by γ(Bo).

Further procedure:

1. Increase the initial content of Bo tar to the value x (eg p.1 or 0.2%, absolute). The produced anodes, which have a new content of Bo+x tar, have a dry density x) which was measured as an average over n anodes (eg 5 to 20 and preferably 10), in order to equalize small fluctuations.

The addition of x can be positive or negative.

2. γ(Bo+x) is compared with the mean value of γ(Bo).

If γ(Bo+x) > γ(Bo), Bo+x is increased by the value of x with the same sign as the previous value.

If γ(Bo+x) > γ(Bo), Bo+x is increased by the value of x, but with a sign opposite to the sign of the previous value, etc. In doing so, each stage is based on a comparison of the density level, which was achieved in the current phase, with the density level from the previous phase.

This algorithm can be justified by the fact that we will be below the optimal value of Bm, if an increase in Bo (positive addition of x) results in an increase in the density of the anode, and very likely we will exceed the optimal value of Bm, if this increase in Bo leads to a decrease in density.

Analogously, if a decrease in Bo (negative addition of x) leads to a decrease in dry density, we will very likely be below the optimal value (content) of Bm, while we are above the optimum Bm, if this decrease in Bo leads to an increase in its density.

3. If any comparison between γ(Bo+nx) and γ(Bo+(n+1)x) leads to equality, we can give the operator or automaton the following instructions:

- or maintain the tar content (do not change it) at the level of Bo+(n+1)x;

- or change it to Bo+(n+1)x, in which x' is positive or negative, and it is preferable that this operation is performed with the addition of x' which is equal to x or less than x (eg x' = x/ 2), so that there is no excessive deviation from the optimum Bm, if we take into account that we are close to the value.

4. The maximum setting limit is for changes in Bo.

This limit Bo ± x (fig. 3) can be fixed at n additions of x. Each phase can be, for example, 0.1 or 0.2% of tar (relative to the absolute value), and x can be dixed, for example, at ± 0.5 or 0.6% ( absolutely).

5. Similarly, in order to avoid predominantly large fluctuations around the optimal value of Bm, after a certain settling time, we can re-determine the scale of the initial value Bo of the tar content and give a new value, which is essentially equal to the optimal value of Bm, which can be derived from the actual fluctuation curve dry density versus tar content and which is shown in Fig. 4 and which the computer can automatically achieve.

As shown before, all of these operations can be performed:

- or manually: in this case, the operator reads the data provided by the computer and acts in such a way as to achieve the addition x in the indicated direction and within the limits, which are determined for the total fluctuations

X = Σx;

- or automatically, and the calculation data can be entered into a programmed automatic unit or into an automatic unit controlled by a microprocessor, which provides data and/or preparation for displaying the development of various parameters.

Fig. 4 shows an experimental curve showing the full range from 13.4 to 14.5% tar content.

It will be found that the density values, measured for each value of tar content, regroup on the dispersion line, with an amplitude of about +0.002 to the absolute value of the dry density.

Example of performance

On the line, which produces anodes intended for electrolysis tanks, Bo is fixed at 13.4% tar, x at 0.1% and X at ± 0.6%. It was found that the regulator optimizes the tar content at about 13.6%, which corresponds to the highest density (dry density) at 1.416 and, in other words, a raw density of 1.638, which represents an extremely high value, which ensures baked anodes of very high quality.

The use of the invention is not limited to the production of anodes:

any carbon block, obtained by forming a carbon dough by vibration curing, pressing or vibration pressing, can be made with maximum dry density if the tar content is regulated according to this invention.

Claims (9)

1. Procedure for regulating the content of tar in previously baked anodes, which are intended for obtaining aluminum by electrolysis, to the optimal value Bm, which corresponds to the maximum dry density, while these anodes are made by hot mixing of ground carbon aggregate and baking the tar, which is introduced in a controlled ratio, and which are then shaped by thickening and final baking at an elevated temperature, indicated by the fact that the following stages are repeated in the specified order: 1.) and priorities, the initial value of Bo tar is fixed as a percentage by weight in the carbon dough; the mean dry density γ(Bo) of a number of n anodes, which are obtained from a carbon paste containing Bo% tar, is obtained by measuring the height H after densification and their weight P (length and width are treated as constant within the limits of the procedure); of permanent tar is changed for a positive or negative addition x, which is added to the value (Bo+x)%; the mean dry density γ(Bo+x) of the number of n anodes is obtained from the carbon paste which, according to the calculation, contains (Bo+x)% tar. 2.) the computer compares γ(Bo+x) with γ(Bo). If γ(Bo+x) > γ(Bo), Bo+x is incremented by the value of x with the same sign as the previous increment. If γ(Bo+x) < γ(Bo), Bo+x is incremented by the value of x with the opposite sign of the previous increment, etc. 2. The method according to claim 1, characterized by the fact that the current content of Bo+(n+1)x remains unchanged, if equality is established in one of the comparisons of γ(Bo+nx) and γ(Bo+ (n+1)x). 3. The method according to claim 1, indicated by the fact that Bo+(n+1)x is increased by the value x', which can be positive or negative, and x' is equal to or less than x, if when comparing γ(Bo +nx) and γ(Bo+(n+1)x) establishes equality. 4. The procedure according to claim 1, indicated by the fact that the maximum limit ± X is set for Bo changes. 5. The method according to one of claims 1 to 4, indicated by the fact that the actual curve of dry density changes depending on the tar content and based on this the optimal value of Bm corresponding to the maximum dry density of the anodes is established. 6. The method according to claim 1, characterized by the fact that the initial value Bo of the tar content after a certain settling time is re-directed to a new value, which is equal or almost equal to the optimal value Bm. 7. The method according to any one of claims 1 to 6, characterized in that the absolute value of the additive x is fixed between 0.1 and 0.2% and that the maximum change in tar content is Bo±0.6%. 8. The method according to any one of claims 1 to 7, characterized in that the parameters Bo, γ, H, P, X and x are entered into the computer, which informs the operator of the sign of the addition x that is added to each stage of the process. 9. The method according to any one of claims 1 to 7, characterized in that the parameters Bo, γ, H, P, X and x are entered into a computer which is connected to an automatic unit that can be programmed or controlled by a microprocessor that determines the addition of x for each stage of the procedure.