FR2740771A1 - Manufacturing high thermal resistance carbon blocks - Google Patents

Abstract

Manufacturing carbon blocks and especially high thermic shock resistance anodes comprises: (i) crushing and grading carbon to obtain a mixture containing grains in size ranges, U- below 30 mu m, S- from 30 to 300 mu m and G- above 300 mu m, wherein the ratio G/S is more than 4; (ii) mixing the grains with a pitch-based binder at 130-180 deg C to form an homogeneous paste; (iii) compacting the paste and shaping it to form a block; and (iv) curing the block at above 900 deg C.

Description

HIGH THERMAL IMPACT RESISTANCE CARBON BLOCK
TECHNICAL AREA
The invention relates to a process for manufacturing carbon blocks with high thermal shock resistance and in particular anodes intended for the production of aluminum by igneous electrolysis of alumina dissolved in a bath of molten cryolite, according to the Hall-Héroult process.

PRIOR ART
Among refractory products, carbon products are distinguished by their thermal and electrical conduction properties. As a result, they are widely used in electrometallurgy; however, in most of their applications, for example as anode and cathode in igneous electrolysis cells, these carbonaceous products must in addition exhibit very good resistance to thermal shock.

Thus in the case of the production of aluminum by electrolysis of alumina in a molten cryolite bath, the new precooked anode is introduced into the electrolysis bath at a temperature close to 10000C and it is therefore very important to be assured its good resistance to thermal shock before it is put into service. Indeed, the carbonaceous anode is essentially a consumable product which is replaced on the electrolysis cell as it is consumed by combustion; however, a modern electrolysis plant producing 240,000 tons of aluminum per year simultaneously consumes 150,000 anodes, the unit weight of which is of the order of 1 ton. A deterioration in the quality of the anodes, resulting in the appearance and development of cracks following thermal shock, quickly leads to a loss of metal production, which leads to instabilities in the operation of the tanks due to the fall in the electrolysis bath of pieces of carbon detached from the cracked anodes. If one can consider as still acceptable an anodes rejection rate by thermal shock cracking not exceeding 1%, the operating cost for a series of electrolysis increases rapidly when this rate regularly exceeds 2 SO.

For those skilled in the art, it is unfortunately quite frequent to observe significant degradations and over long periods of the thermal shock resistance of prebaked anodes without being able to clearly determine the cause, whereas the conditions of manufacture and use of said anodes have not been modified.In this regard, it is necessary to recall for a good understanding of the invention that these prebaked carbonaceous anodes are obtained by mixing at a temperature generally between 1300C and 1800C of a mixture of an aggregate carbon based on ground coke and a coking binder, such as pitch, then shaping of the carbonaceous paste thus obtained by an appropriate means such as spinning, vibro-tamping, compacting ... and finally cooking for a period of time up to several weeks at a temperature of at least 9000C and generally between 1000 "and 1200" C.

For the operator, the search for the best possible quality of carbonaceous product is a permanent concern, and as regards anodes for electrolysis it is well known (EP0252859 = US 4770826) that obtaining a maximum density for the anode after baking, promotes regular operation and energy efficiency of electrolysis. At the same time, it results in a stabilized, well-distributed and ultimately slower consumption of the anode, hence a reduction in the number of anodes to be manufactured for the same production of metal. However, this search for a maximum density of the product after baking does not provide any solution to the problem of the resistance to thermal shock of these same carbonaceous products.

To date, in fact, a person skilled in the art does not know any intrinsic characteristic of the carbonaceous product after firing which is representative of its capacity to resist thermal shock, so that he does not know what are the manufacturing parameters on which he can act. effectively to limit, or even eliminate, anode cracking without losing the advantages associated with obtaining a high density of the prebaked carbonaceous anode.

Those skilled in the art only have, in order to check the resistance to thermal shock of carbonaceous products, analog methods, the representativeness of the real conditions of use of the carbonaceous products, and therefore the reliability, remain questionable. In addition, these control methods carried out on samples taken from batches of carbonaceous products after cooking, that is to say at the final stage of production, only allow the batches of products presumed to be defective to be eliminated by statistical control. , but without possible correction of upstream production since we do not know what the determining factors are. It should also be remembered that among these analog methods, the most frequently used is based on electrical conductivity and consists in heating by the Joule effect a sample of the product to a temperature supposed to correspond to that of its use, then in abruptly immersing it. ci in a fluid generally at room temperature.

Note also the studies by SATO (S.SATO et al- Carbone 1975- vol.13p.309-316) on the damage mechanisms of graphites by thermal shock. This method consists in cutting test pieces in the graphitic material to be tested and in heating it suddenly by discharging an electric arc at its center. There is then a crack in the material from the moment when the power in the arc reaches a certain level. On this principle FR 2635188 describes a method for controlling the resistance to thermal shock cracking of carbonaceous products. Although providing a better simulation of the thermal shock in the material to be tested, the methods of sudden heating by electric discharge or by induction do not make it possible to assess with sufficient reliability the future behavior of prebaked anodes during their immersion, due to the wide dispersion of the results of the cracking threshold measurements.

In summary, these analog control methods are of no help to those skilled in the art in determining on the carbon blocks and in particular the anodes an intrinsic characteristic of the material truly representative of its ability to withstand thermal shock and a fortiori for it. deduce the parameter (s) on which it can act in manufacturing to quickly correct any drift in this characteristic, without degrading at the same time other characteristics of the anode such as density.

PROBLEM
The manufacture of carbonaceous blocks and in particular of anodes intended for the production of aluminum by igneous electrolysis, systematically exhibiting good thermal shock resistance measured by a cracking reject rate not exceeding 1% (1 cracked electrode for 100 electrodes implemented) and this without altering the other characteristics remains a poorly resolved problem.

OBJECT OF THE INVENTION
The invention relates to a method of manufacturing carbon blocks and in particular anodes, which solves the problem of the thermal shock resistance of the anodes during their immersion in a molten electrolysis bath, without altering the other characteristics. The invention is based on the observation that by an appropriate choice of the particle sizes of the pulverulent carbonaceous materials, a significant improvement in the thermal shock resistance of the carbonaceous blocks is surprisingly obtained, controlled in their actual conditions of use and measured by the rate of scrap by cracking.

More precisely, the invention relates to a method of manufacturing carbon blocks and in particular anodes with high resistance to thermal shock, comprising the successive steps of: a) adjusting the particle size of a carbonaceous aggregate constituting the material
first dry, by appropriate grinding and grading; b) mixing at a temperature generally between 1300C and 1800C
crushed aggregate with a predetermined amount of binder based on
pitch to form a homogeneous paste; c) densification by compacting and shaping of said paste to
constitute the carbonaceous block in the raw state; d) firing of the raw carbonaceous block at a generally higher temperature
to 9000, characterized in that the adjustment of the particle size of said carbonaceous aggregate by grinding and classification is carried out according to 3 theoretical particle size fractions thus defined
Ultra-fine, or UF, grain diameter <30 µm
Sand, or S, with a grain diameter between 30 pm and 300 pm
Grain, or Gr, of grain diameter> 300 µm.

and that the weight ratio of the Gr and S fractions is adjusted such that
Gr / S is at least equal to 4 and preferably between 6 and 15.

During numerous tests to check the resistance to thermal shock of precooked carbonaceous blocks placed under industrial conditions of use, the Applicant has been able to observe behavior that is appreciably different from the products which are nevertheless manufactured and used under quite similar conditions. If the initial hypothesis aimed at relating these variations in thermal shock resistance to the change in origin of certain raw materials such as coke, could not be confirmed, the applicant was able to observe surprisingly that by increasing the weight ratio Gr / If it was possible to very significantly reduce the rate of rejection of carbonaceous blocks by cracking following thermal shock, the immersion of the anode in the bath. Usually this weight ratio Gr / S is systematically less than 4 and generally set between 1 and 3 (Zabreznik p.527- publication LIGHT METAL 24 to 26 February 1987, or Mannweiler & BR <
Keller- JOM - February 1994- fig.5- p.20).

It is in fact known that the pulverulent dry materials for preparing the carbonaceous paste by mixing with pitch and then densifying it by compacting, are classified into 3 theoretical particle size fractions Gr, S and
UF having specific and complementary functions during the production of the raw carbonaceous block which prove to be decisive for the properties of the final product.

Thus the grains Gr, by their compact stacking, form a framework with
low thermal contraction.

The ultra-fine UF form with the liquid pitch after mixing a
plastic mixture that does not flow.

As for the sand S, it plays an intermediate role.

These points being clarified, it appears that a person skilled in the art has no reason to adjust these weight proportions outside the limits usually recommended (1 <Gr / S <3) because nothing in the prior art suggests that there is any correlation between this Gr / S ratio and the rate of rejection by cracking of the carbonaceous blocks and in particular of the anodes.

In addition, a notable modification of the relative weight proportions of Gr, S and UF can alter their complementarity relationships and create heterogeneities in the carbonaceous paste and then in the block, or even a degradation of the raw and after baking densities, even if the ratio overall weight dry matter (Gr + S + UF) / pitch is unchanged in the carbonaceous paste.

Finally, those skilled in the art are hampered in their initiatives to change the weight proportions of the 3 size fractions Gr, S and UF by economic operating constraints which we will examine below.

Nevertheless, by deliberately departing from the usual framework for the use of powdered carbonaceous raw materials, the applicant has demonstrated (see Table 1 below) that for a Gr / S 2 4 ratio, the rate of scrap by cracking of the anodes n 'did not exceed 1% whereas for values 6 <Gr / S <15 the reject rate was zero and that without degradation of the density characteristics. On the other hand, for Gr / S> 15, a significant decrease in raw and after cooking densities begins to be recorded.

TABLE 1
(For each Gr / S immersion test under real conditions on 50 anodes)

<tb> Gr / S <SEP> Nb <SEP> anode <SEP>% <SEP> waste <SEP> density <SEP> raw <SEP> density <SEP> after <SEP> firing
<tb><SEP> scrap <SEP> ~ 0.002 <SEP> +0.002 <SEP>
<tb><SEP> 1.5 <SEP> 3 <SEP> 6 <SEP> 1.640 <SEP> 1.590
<tb><SEP> 2.5 <SEP> 2 <SEP> 4 <SEP> 1.64 <SEP> 1.595
<tb><SEP> 4 <SEP> 1 <SEP> 2 <SEP> 1,640 <SEP> 1,595
<tb><SEP> 6 <SEP> 0 <SEP> 0 <SEP> 1.641 <SEP> 1.60
<tb><SEP> 10 <SEP> 0 <SEP> 0 <SEP> 1.642 <SEP> 1.595
<tb><SEP> 15 <SEP> 0 <SEP> 0 <SEP> 1,640 <SEP> 1,590
<tb><SEP> 30 <SEP> 0 <SEP> 0 <SEP> 1.626 <SEP> 1.578
<tb>
These new Gr / S adjustments in unusual particle size ranges were furthermore made without economically penalizing the industrial process, that is to say by taking into account the constraints inherent in the basic process for manufacturing the anodes. These constraints, which are well known to those skilled in the art, relate to the obligation to initially use 4 industrial particle size fractions as pulverulent dry material, namely:
TG (very coarse) 1.5 mm <recycled grains <15 mm obtained by grinding
inherent manufacturing wastes such as spent anode butts.

G (coarse) 1.5 mm <coke grains <5 mm formed by the slice
greater particle size of the coke grains after sieving to 1.5 mm.

M (medium) 0 <coke and recycled grains <1.5 mm consisting of
particle size fractions of less than 1.5 mm of coke grains and
recycled.

F (fines) - fines <0.2 mm constituted by the fraction of the means M refined
by grinding. This fine fraction must contain a sufficient proportion
ultra-fine UF <0.03 mm (according to the theoretical ternary formulation
previously defined). The theoretical particle size fractions and
industrial therefore overlap according to the known diagram of Figure 1
single (in Annex).

Thus the grain Gr is brought by TG, G and a part of M. The sand S is brought by a part of M and F and it is the same for the UFs.

The increase in Gr / S with a view to improving the thermal shock resistance of the anodes obviously leads to increasing the Gr fraction and / or to reducing the S fraction, but the new adjustment of the dry matter formulation besides the need to keep a certain percentage of ultra-fines must respect a certain number of constraints which are as follows:
The 4 industrial fractions TG / G / M / F exist and are inherent to
processes for the production of electrolysis anodes (recycling of anode butts).

All of these 4 industrial fractions must be consumed for
obvious economic reasons after their reclassification according to the 3 fractions
theoretical particle size Gr / S / UF and adjustment of the proportions
respective weight.

The raw and after cooking densities must remain as high as possible,
while preferably keeping the pitch / dry matter ratio unchanged,
since the total amount of dry matter Gr + S + UF remains constant.

The particle size distribution of coke as a raw material is imposed by
constraints of the supplier process.

The fixed percentage of TG must be sufficient to allow the recycling of
all of the waste inherent in manufacturing, in particular cigarette ends
anode. It must therefore be greater than 10% and preferably compns
between 20 and 30% of the total load of dry carbonaceous matter.

These details being given, the range of the respective weight proportions of the 3 theoretical particle size fractions Gr, S and UF making it possible to increase the Gr / S without deterioration of the density characteristics are as follows:
60% <Gr <90% including in particular 20 to 30% of TG in the form
of recycled grains
0.5% <S <15%
5% <UF <25%
On a practical level, the invention will be better understood by the detailed description of its implementation, that is to say by the description of the means for increasing the Gr / S ratio up to at least 4 and preferably between 6 and 15 for a precise formulation of dry matter and with the help of FIG. 1 representative of the particle size distributions of the 4 industrial fractions TG, G, M and F and of the 3 reconstituted theoretical fractions Gr, S and UF.

DETAILED IMPLEMENTATION OF THE INVENTION
We first act on Gr to maximize it and then on S to minimize it.

a) To maximize Gr, it is necessary to keep their particle size in the 2 fractions
coarser industrial, TG and G of the carbon aggregate. So for
a classic formulation where all the recycled grains, for example
26%, are consumed in the form of TG, we obtain grain sizes
typical of the different industrial fractions such as those indicated
in Table 2 below with TG = 26% - G = 21% - M + F = 53% (knowing that
the fines F will subsequently be obtained by grinding specific to the
less than an aliquot of M).

TABLE 2

<tb><SEP> TG <SEP> 26.0% <SEP> G <SEP> 21.0 <SEP>% <SEP> M + F <SEP> 53.0 <SEP>% <SEP> Total <SEP > calculated <SEP> Total <SEP> cumulative
<tb> Mesh <SEP> Tyler <SEP> 1x% <SEP> 2x% <SEP> 1x% <SEP> 2x% <SEP> 1x% <SEP> 2x% <SEP>% <SEP>% <SEP><SEP> Gr / S <SEP> 6.75
<tb><SEP> RT <SEP> 3/8 <SEP> 45.0 <SEP> 11.7 <SEP> 5.9 <SEP> 1.2 <SEP> 12.9
<tb><SEP> RT <SEP> 4 <SEP> 30.9 <SEP> 8.0 <SEP> 26.2 <SEP> 5.5 <SEP> 13.5 <SEP> 26.4
<tb><SEP> RT <SEP> 10 <SEP> 18.7 <SEP> 4.9 <SEP> 48.1 <SEP> 10.1 <SEP> 4.2 <SEP> 2.2 <SEP> 17.2 <SEP> 43.6 <SEP> Gr <SEP> 87.1
<tb><SEP> RT <SEP> 20 <SEP> 3.7 <SEP> 1.0 <SEP> 11.7 <SEP> 2.5 <SEP> 28.5 <SEP> 15.1 <SEP> 18.6 <SEP> 62.2
<tb><SEP> RT <SEP> 48 <SEP> 0.8 <SEP> 0.2 <SEP><SEP> 5.8 <SEP> 1.2 <SEP> 44.3 <SEP> 23.5 <SEP> 24.9 <SEP> 87.1
<tb><SEP> RT <SEP> 100 <SEP> 0.2 <SEP> 0.05 <SEP> 1.5 <SEP> 0.3 <SEP> 15.4 <SEP> 8.2 <SEP> 8.55 <SEP> 95.65
<tb><SEP> RT <SEP> 200 <SEP> 0.2 <SEP> 0.05 <SEP> 0.5 <SEP> 0.1 <SEP> 4.9 <SEP> 2.6 <SEP> 2.75 <SEP> 98.4 <SEP> S <SEP> 12.9
<tb><SEP>> 30 m <SEP> 0.5 <SEP> 0.1 <SEP> 0.3 <SEP> 0.1 <SEP> 2.7 <SEP> 1.6 <SEP> 100, 0
<tb><SEP><<SEP> 30 m <SEP> 0.0 <SEP> 0.0 <SEP> 0.0 <SEP> 0.0 <SEP> 0.0 <SEP> 0.0 <SEP > 0.0 <SEP> 100.0 <SEP> UF <SEP> 0.0
<tb><SEP> Total <SEP> 100 <SEP> 26.0 <SEP> 100 <SEP> 21 <SEP> 100.0 <SEP> 53.0 <SEP> 100
<tb> lx percentage of rejection per 100 g, at the mesh considered expressed in mesh.

2x percentage of rejection at the mesh considered of the particle size fraction.

It is by the other mode of action consisting in minimizing S that we can
definitively set the Gr / S ratio in the desired range and produce the
quantity of ultra-fines necessary to obtain a high density of the anode.

b) To minimize S, 2 methods are applicable which are consistent with
the constraint of consuming all of each of the 4 fractions
industrial. These 2 methods are described successively.

1st method
It consists in increasing the UF fraction to the detriment of the S fraction by intensive grinding of at least one aliquot F of the industrial fraction M.

Usually this ground aliquot F is intended to enrich the 2 ternary fractions S and UF. According to the invention, by further grinding in a ball mill, the UF fraction is increased while reducing S. Thus, in its usual operation, a ball mill produces around 35% of ultra-fine UF and therefore by approximately difference. 65% of S. Finer grinding therefore consists in reducing S.

Table 3 below shows the impact of the degree of fineness of the industrial fraction M after grinding, on Gr, S and therefore on the Gr / S ratio and this for different percentages of ultra-fine UF in the final reconstituted formulation of the Gr, S and UF mixture, percentages which must obviously remain between 5 and 25% so as not to subsequently alter other properties of the finished product.
F1, F2, F3, F4 increasing, measured by the percentage of ultra-fine UF (<30 μm) relative to the passing through a 200 mesh (<75 μm) mesh sieve and such that: F1 = 54S, F2 = 64S, F3 = 74S, F4 = 100%, for 3 distinct ultra-fine contents, for example 6%, 9%, and 12%, we see in the final reconstituted formulation of the mixture Gr, S, UF deduced from a table of particle size classification not shown but similar to
Table 2, that for each of the 4 degrees of fineness: - the fact of finer grinding, F2 compared to F1, F3 compared to F2 and F4 by
compared to F3, allows to increase Gr / S - this is only in the case of a ground powder with 100% UF (F4)
that Gr / S also increases with the rate of ultra-fines that we impose in
the final Gr, S, UF mixture.

TABLE 3

<tb><SEP> Degree <SEP> of <SEP> fineness <SEP> * <SEP> F1 <SEP> F2 <SEP> F3 <SEP> F4 <SEP>
<tb><SEP>%<SEP> UF <SEP> in <SEP> PT200 <SEP> 54 <SEP> 64 <SEP> 74 <SEP> 100
<tb><SEP> 6% UFin <SEP> Gr <SEP> 73.7 <SEP> 77.7 <SEP> 79.9 <SEP> 82.5
<tb><SEP> form.reconst. <SEP> S <SEP> 20.3
<tb><SEP> Gr / S <SEP> 3.63 <SEP> 4.77 <SEP> 5.67 <SEP> 7.17
<tb><SEP> 9% <SEP> of UF <SEP> in <SEP> Gr <SEP> 67.0 <SEP> 73.0 <SEP> 76.3 <SEP> 80.2
<tb><SEP> form.reconst. <SEP> S <SEP> 24 <SEP> 18.0 <SEP> 14.7 <SEP> 10.8
<tb><SEP> Gr / S <SEP> 2.79 <SEP> 4.06 <SEP> 5.19 <SEP> 7.43
<tb><SEP> 12% <SEP> of UF <SEP> in <SEP> Gr <SEP> 60.4 <SEP> 68.3 <SEP> 72.8 <SEP> 77.9
<tb><SEP> form. <SEP> reconst. <SEP> S <SEP> 27.6 <SEP> 19.7 <SEP> 15.2 <SEP> 10.1
<tb> Gr / S <SEP> 2.19 <SEP> 3.47 <SEP> 4.78 <SEP> 7.71
<tb>
* M fraction crushed after sieving to 1.5 mm.

2nd method
It consists in reducing the fraction S while avoiding sending for grinding the upper particle size range of M between 0.3 mm and 1.5 mm which, if reference is made to FIG. 1, constitutes a fraction of Gr. This amounts to reducing the particle size distribution of the average industrial fraction M by making it very close to that of the sand fraction S. For this, it is sufficient to replace the 1.5 mm sieve by a sieve of the closest possible mesh size. 0.3 mm and carry out intensive grinding on at least one aliquot and preferably all of said industrial fraction M passing through this new sieve. If the quantity of ultra-fines targeted in the final reconstituted formulation of the mixture Gr , S and UF, between 5 and 25%, cannot be reached from the single crushed industrial fraction M, an additional supply will be made from the industrial fractions of crushed G or TG grains.

As previously, Table 4 shows the impact on the Gr / S ratio of the degree of fineness of the new industrial fraction M sieved at 0.3 mm and then ground to different percentages of ultrafine F 1 = 54S , F2 = 64S, F3 = 745 and F4 = 1005; and this for 3 different percentages of ultra-fine UF in the final reconstituted formulation of the mixture
Gr, S and UF namely 6%, 9% and 12%.

TABLE 4

<SEP> Degree <SEP> of <SEP> fineness <SEP> * <SEP> F1 <SEP> F2 <SEP> F3 <SEP> F4
<tb>% <SEP> UF <SEP> in <SEP> P <SEP> 200 <SEP> 54 <SEP> 64 <SEP> 74 <SEP> 100
<tb> 6% <SEP> of UF <SEP> in <SEP> formula <SEP> Gr <SEP> 82.7 ** <SEP> 87.10 <SEP> 87.1 <SEP> 87.1
<tb> reconstituted <SEP> S <SEP> 11.3 <SEP> 6.9 <SEP> 6.9 <SEP> 6.9
<tb><SEP> GrlS <SEP> 7.32 <SEP> 12.62 <SEP> 12.62 <SEP> 12.62
<tb> 9% <SEP> of UF <SEP> in <SEP> formula <SEP> Gr <SEP> 73.9 ** <SEP> 81.7 ** <SEP> 86.0 ** <SEP> 87.1
<tb> reconstructed <SEP> S <SEP> 17.1 <SEP> 9.3 <SEP> s <SEP><SEP> 3.9
<tb><SEP> Gr / S <SEP> 4.32 <SEP> 8.78 <SEP> 1 <SEP> 7.2 <SEP> 22.33
<tb> 12% <SEP> of UF <SEP> in <SEP> formula <SEP> Gr <SEP> 65.3 ** <SEP> 75.7 ** <SEP> 81 <SEP>, 4t * <MS> 87.1
<tb> reconstituted <SEP> S <SEP> 22.7 <SEP> 12.3 <SEP> 6.6 <SEP> 0.9
<tb><SEP> Gr / S <SEP> 2.88 <SEP> 6.15 <SEP> 12.33 <SEP> 96.7
<tb> * M fraction crushed after sieving to 0.3 mm.

** means that coarse grain G had to be crushed to achieve the desired level of ultrafine.

It can be seen that with one exception, values greater than 4 are obtained with the very possibility of clearly exceeding the preferential range 6 to 15, in particular in the case of a 100% UF powder, where again , the Gr / S increases with the level of ultra-fines, unlike what is observed with the other powders F1, F2 and F3.

Those skilled in the art therefore have 2 methods available to minimize S and adjust
Gr / S. He will preferably use the 1st method if he has sufficient grinding means, otherwise he will opt for the 2nd method which requires only a small over-equipment in sieving means due to the displacement of the cut-off threshold from M towards the fines, which obviously does not exclude the combination of the 2 methods.

Moreover, by setting Gr / S in the preferential range 6 to 15, it is ensured to avoid any degradation of the other characteristics of the anodes and in particular of the densities in the uncured state and after baking, while maintaining the standard conditions of development of anodes at all subsequent stages of manufacture.

These known conditions are only briefly recalled:
Mixing in the presence of a predetermined quantity of pitch,
preferably in the vicinity of 1500C, the crushed coke-based aggregate of which
the particle size reconstituted according to the ternary mixture Gr / S / UF satisfies the
relation G <Gr / S <UF and formation of a carbonaceous paste.

Densification by compacting and shaping of said paste for
constitute the green anode.

Baking preferably between 1000 "C and 1200" C, preferably in an oven with
chambers, from the raw anode.

Claims (1)

CLAIMS A method of manufacturing carbon blocks and in particular anodes with high thermal shock resistance comprising the following successive steps: a) adjustment of the particle size of a carbonaceous aggregate, by grinding and particle size classification, b) mixing at a temperature generally included between 1300C and 1800C of the crushed aggregate with a predetermined amount of pitch binder to form a homogeneous paste, c) densification by compacting and shaping of said paste to constitute the carbonaceous block in the raw state, d) Firing of the raw carbonaceous block at a temperature generally greater than 9000C. characterized in that the adjustment of the particle size of said aggregate carbonated by grinding and classification is carried out according to 3 fractions: Ultra-fine, or UF, grain diameter <30 µm Sand, or S, with a grain diameter between 30 pm and 300 pm Grain, or Gr, of grain diameter> 300 µm and that the weight ratio of the Gr and S fractions is adjusted so that Gr / S is greater than or equal to 4. 2 Method according to claim 1 characterized in that the Gr / S ratio is preferably set between 6 and 15. 3 Method according to claims 1 or 2 characterized in that the respective weight proportions of the 3 grain size fractions theoretical Gr, S and UF are as follows: 60% <Gr <90% 0.5% <S <15% 5% s <UF <25 7O 4 Process according to any one of Claims 1 to 3, characterized in that we maximize the value of Gr by keeping their particle size at 2 coarser industrial fractions TG and G of the carbon aggregate. 5 A method according to any one of claims 1 to 4 characterized in that the value of S is minimized by intensive grinding of at least one aliquot of the average industrial fraction M of the carbon aggregate of grain size between 0 and 1.5 mm, so as to increase the UF weight fraction to the detriment of S in the ternary distribution Gr / S / UF. 6 A method according to claim 5 characterized in that the degree of fineness of the so-called intensive grinding, measured by the percentage of ultra-fine UF in the particle size fraction passing through the 200 mesh sieve, is included preferably between 54% and 10070 .. 7 A method according to any one of claims 1 to 4 characterized in that we minimize the value of S by reducing the spreading particle size of the average industrial fraction M to make it very close to that of the sand fraction S and that by displacement of the threshold of usual cut from 1.5 mm to 0.3 mm and by grinding intensive on at least one aliquot and preferably on the whole of said industrial fraction M. 8 A method according to claim 7 characterized in that the percentage weight of ultra-fine UF in the industrial fraction M after grinding intensive is preferably between 54 and 100% of the passing through the sieve of 200 mesh.