Patent Application
20190390362
The invention relates to a process for the surface treatment of a metal wall which has the effect of reducing the formation of coke at the surface of this wall. More specifically, the invention relates to a process for the surface removal of carbides from a wall made of metal alloy, in particular by chemical treatment. The invention also relates to the use of a metal wall treated by the treatment process in a process for the treatment of hydrocarbons.
The walls of the reactors of some units of the petrochemical or chemical industry are sometimes subjected to very severe operating conditions which can result in phenomena of coking. For example, the manufacture of alkenes, monomers valued in the polymer industry, is obtained by cracking oil-derived hydrocarbons at temperatures of the order of 800 to 900° C. In this type of process, a mixture of hydrocarbons and steam is circulated at high speed in reactors, generally formed of metal tubes, often made of alloys rich in nickel and in chromium. The reactors are thus subjected to high temperatures and to complex aggressive atmospheres and the formation of carbon (coke) is observed at the surface of the walls of the tubes, this formation being catalysed by the iron and the nickel present in the metal alloy forming the walls. This deposition of coke can result in a fouling of the tubes, causing a head loss, a deterioration in the conductivity of the walls and a decrease in the yields. It is then necessary to shut down the unit in order to remove the coke formed, an operation which is harmful to the productivity of the unit.
This is the reason why numerous research studies have been carried out by manufacturers in order to limit the formation of coke.
These research studies have made possible the implementation of solutions comprising in particular:
The formation of a protective oxide layer can in particular be obtained by the use of appropriate alloys, for example rich in chromium or in aluminium, or by oxidation pretreatments.
Despite the existence of these solutions, there still exists a need for a simple and inexpensive treatment which makes it possible to reduce the formation of coke.
To this end, there is provided a process for the treatment of a wall made of Fe—Ni—Cr metal alloy of an industrial reactor which reduces the formation of coke on the said surface when it is subjected to operational conditions favourable to coking, the metal alloy comprising in particular, within its structure, carbides, some of which can show on the surface. For example, the metal alloy contains at least 5% by weight of iron, at least 18% by weight of chromium, at least 25% by weight of nickel and at least 0.05% by weight of carbon.
The term “operational conditions favourable to coking” is understood to mean conditions liable to bring about the formation of coke on the surface. The parameters influencing the coking comprise, for example, the temperature, the nature of the liquid or gaseous fluids circulating inside the reactor and in contact with the surface, the flow conditions of the fluids (turbulences).
According to the invention, the process comprises a chemical surface treatment stage, during which at least a part of the carbides initially present in the alloy, in particular at the surface, is removed by electrolytic dissolution.
Surprisingly, while the formation of coke is essentially catalysed by the presence of iron and of nickel, an improvement in the resistance to coking of a surface made of metal alloy is observed on using the treatment according to the invention. In other words, the formation of coke is reduced in comparison with an untreated surface.
Without wishing to be committed to a theory, the removal from the surface of at least a part of the carbides by electrolytic dissolution makes it possible to reduce the formation of coke.
Such a surface treatment stage exhibits the advantage of being easy to carry out and relatively inexpensive.
Advantageously, several chemical stages of treatment by electrolytic dissolution can be provided.
The process according to the invention can advantageously be carried out in order to treat a wall of a reactor after the manufacture of the reactor and before bringing it into service.
Advantageously, after the chemical treatment stage, a stage of oxidation of the wall can be envisaged, which makes it possible to further reduce the formation of coke. Without wishing to be committed to a theory, the surface treatment appears to promote the formation of a homogeneous oxidized layer and to thus reduce the formation of coke.
The invention is more particularly suited to treating a wall of a steam cracking reactor (furnace) or of any other plant in which there is observed the formation of coke catalysed by iron, nickel and optionally by other catalysing metal elements present in the metal alloy of which the reactor is composed.
The invention thus also relates to a process for the treatment of hydrocarbons under conditions capable of bringing about the formation of coke, characterized in that the hydrocarbons are brought into contact with a surface of a wall made of Fe—Ni—Cr metal alloy, the said surface of the metal wall being pretreated by a treatment process according to the invention so as to reduce the formation of a coke deposit. In particular, the metal alloy is preferably a metal alloy containing at least 5% by weight of iron, at least 18% by weight of chromium, at least 25% by weight of nickel and at least 0.05% by weight of carbon.
The process for the treatment of the hydrocarbons can be a cracking process, in which the hydrocarbons are brought into contact with the wall as a mixture with steam. Such a treatment is, for example, carried out in a steam cracking reactor.
Advantageously, the hydrocarbons can be brought into contact with the surface of the metal wall at a temperature of 800 to 900° C., in particular as a mixture with steam.
The treatment process according to the invention is intended for the treatments of Fe—Ni—Cr metal alloys containing in particular carbides within their structure. These carbides can show on the surface, in other words be in contact with the gaseous medium surrounding the alloy, and/or can be located in the immediate proximity of the surface, for example from a depth of 1 μm or more.
The presence of these carbides can be observed by observations with a scanning electron microscope of the surface of the wall and/or of sections of the wall.
Such carbides are formed by precipitation during the manufacture of the wall. They can also appear in part in operation. In a known way, carbides, which are particularly stable chemically, are formed from the carbon present in the metal alloy. These carbides can in particular be observed for a carbon content of the metal alloy of at least 0.05% by weight.
This type of metal alloy is suitable in particular for use at high temperature (heat resistant alloys).
Advantageously, the alloys treated are alloys exhibiting an Fe—Ni—Cr matrix, optionally an austenitic matrix, within which precipitate chromium carbides (CrxCy), indeed even niobium carbides (NbC), when niobium is present, and/or carbonitrides, when the alloy contains nitrogen, and/or other carbides optionally.
Such alloys thus comprise:
In addition, these alloys comprise carbon, in particular from 0.05% to 1% by weight of carbon, preferably from 0.08% to 0.6% by weight.
Advantageously, in these alloys, the nickel or the iron can be the predominant element. In general, the iron content is the remainder to 100% of the contents of the other elements present in the alloy.
The treated metal alloys can comprise other elements. They can in particular comprise one or more of the following elements:
The metal alloy used can preferably be suitable for centrifugal casting. It can in particular observe Standard EN 10295 relating to heat-resistant steel castings. This technique consists in pouring the liquid metal into a mould driven with a rotational movement around its main axis. Generally, the mould rotates at a speed such that it creates a mean acceleration of the order of several hundred and up to 1000 m/s2 or more, in some cases. The moulds can be made of sand or of metal die, fitted to machines having a horizontal, vertical or oblique axis. The parts obtained by centrifugation have very good physical and mechanical characteristics.
The treated wall can thus advantageously be produced by centrifugal casting.
More specifically, this stage is an electrochemical stage, in particular a selective dissolution electrochemical stage.
During this stage, at least a part of the carbides initially present in the alloy, in particular at the surface, namely the carbides showing on the surface and/or located in the immediate proximity of the surface, is removed by electrolytic dissolution.
In particular, this stage is advantageously carried out under conditions suitable for dissolving at least a part of these carbides over a depth of at least 10 μm (from the treated surface), preferably of at least 20 μm, more preferably of at least 30 μm, indeed even of at least 40 μm.
In particular, the electrolytic dissolution conditions can advantageously be suitable for dissolving one or more carbides chosen from chromium carbides, niobium carbides, when the alloy contains niobium, carbonitrides, when the alloy contains nitrogen, indeed even other carbides, preferably chromium carbides.
In an alternative form or in combination, several chemical surface treatment stages can be provided. Thus, the process according to the invention can comprise at least one other chemical treatment stage, during which at least a part of the carbides initially present in the alloy, in particular at the surface, and not dissolved during a preceding chemical treatment stage is removed by electrolytic dissolution.
Advantageously, it is possible to carry out a chemical treatment stage which is a stage of electrolytic dissolution of chromium carbides and another chemical treatment stage which is a stage of electrolytic dissolution of niobium carbides, when the alloy contains niobium, for example in the amounts mentioned above.
A washing stage can be provided between two successive chemical treatment stages under conditions capable of removing the traces of electrolytes from the treated surface. It can be one or more stages of rinsing the wall with water, preferably distilled water, optionally followed by one or more stages of rinsing with an alcohol, for example ethanol. This washing can be followed by a drying under conditions which make it possible to remove the rinsing fluid or fluids from the wall to be treated.
In a specific embodiment, the electrochemical dissolution of the chromium carbides is carried out and then the electrochemical dissolution of the niobium carbides is carried out.
The chemical stage is carried out by placing the wall to be treated at the anode of an electrolysis cell, the cathode being formed of a conductive part (for example made of metal or graphite), and by applying an electric potential through the electrolysis cell.
The chemical treatment stage can be carried out in an electrolysis cell comprising an aqueous solution of an alkali metal hydroxide or an aqueous sulfuric acid solution.
A procedure for the dissolution of chromium carbides is, for example, described in U.S. Pat. No. 4,851,093 A.
The electrolytic solution can thus comprise an aqueous solution of a soluble metal hydroxide. This metal can be an alkali metal, such as Na, K or Li, for example Na.
The electrolytic solution can comprise from 100 to 200 g/l of alkali metal hydroxide, preferably from 120 to 150 g/l.
Preferably, the chloride content of the solution is less than 10 ppm by weight.
A procedure for the dissolution of niobium carbides is, for example, described in “Anode dissolution characteristics of titanium, niobium and chromium carbides”, 1971, V. Cihal, A. Desestret, M. Froment and G. H. Wagner.
The electrolytic solution can thus be an aqueous sulfuric acid solution, the sulfuric acid concentration of which can be from 1 to 10 mol·l−1, preferably from 2 to 9 mol·l−1. However, the invention is not limited to these specific conditions: a person skilled in the art is in a position to determine other suitable concentrations of sulfuric acid, indeed even to use other suitable electrolytic solutions.
The difference in electric potential applied to the electrolysis cell can be from 4 to 8 volts or from 3 to volts, indeed even from 3 to 5 volts. It may be preferable to avoid greater differences in potentials in order not to generate too much heat.
The current flow passing through the electrolysis cell can vary according to the surface area to be treated. The current density can typically be from 5 A/in2 (7750 A/m2) to 10 A/in2 (15 500 A/m2) of surface area of wall to be treated.
The duration of the treatment can be variable, for example from 4 to 50 hours or from 2 to 50 hours, for example from 2 to 30 hours, depending on the amount of carbides and/or on the depth of wall which it is desired to treat.
The temperature of the electrolytic solution can vary from ambient temperature up to approximately 85° C. However, it is preferable for the temperature of the solution to be kept below 60° C.
When it is present, this stage is preferably carried out after the chemical treatment stage described above. This impact surface treatment is obtained by hammering the surface by projection of particles under conditions suitable for obtaining a permanent plastic deformation of the surface, in particular under conditions suitable for obtaining covering of the carbides initially present at the surface by permanent plastic deformation of the surface.
The carbides initially present at the surface can show on the surface and/or be located in the immediate proximity of the surface, in particular located at a depth of 1 μm and more from the surface.
These projections of particles create, at the surface, a small impression or crater. This surface is thus plastically deformed in tension or in compression during the impact.
Usually, the objective of this type of impact surface treatment is to compress the substance under the impacted surface: this compressed substance tends to regain its initial volume, resulting in high residual compressive stresses. This makes it possible to significantly increase the lifetime of a part made of alloy as virtually all of the fatigue and corrosion failures under tension are initiated at the surface of such parts.
In the present invention, the impacts caused by the projectiles will cover this surface with a uniform layer in compression.
While the chemical treatment stage brings about the formation of cavities, the permanent plastic deformation obtained by the implementation of the mechanical treatment stage makes it possible to fill these cavities in again, at least partially. In addition, covering of the carbides initially present at the surface—in other words covering of the carbides showing on the surface and/or located in the immediate vicinity of the surface before the mechanical treatment—which have not been dissolved by the chemical treatment stage is also observed. Surprisingly, this modification of the surface, which limits the access to the carbides trapped inside the metal alloy, also makes it possible to reduce the formation of coke.
Depending on the form of the particles and on the projection power, such a surface treatment can be denoted by the terms “shot peening” (use of beads), “sand blasting”, “alumina blasting” (use of corundum particles) or “shot blasting”.
The particles can be diverse in nature (inorganic, metallic, and the like) and of varied shapes (spherical or angular) and dimensions. The particles can thus be chosen from aluminium oxide (for example white or brown corundum) particles, metal particles, beads made of material which is inert under the operational conditions of use of the wall made of metal alloy, for example made of glass or of aluminium oxide, or nesosilicate particles.
In particular, the nesosilicate (garnet) particles exhibit a general formula AmBn(SiO4)t, where A is a transition metal or an alkaline earth metal and B is a transition metal or a rare earth metal. In particular, A can be chosen from Mg, Ca and Mn and B can be chosen from Y, Ce and La.
The particles can exhibit a mean diameter of 100 to 500 μm.
Use may be made, by way of example, of glass beads with a mean diameter of 100 to 200 μm, of aluminium oxide particles with a mean diameter of 250 to 500 μm.
The particles can be projected by a gaseous fluid, for example air, argon or other, under a pressure of 200 to 400 kPa (2 to 4 bars), preferably of 250 to 350 kPa (2.5 to 3.5 bars).
For glass beads, a pressure of 300 to 350 kPa can be used. For aluminium oxide particles, pressures of 270 to 320 kPa can be used.
The projection distance can be from 5 to 25 cm, for example from 10 to 20 cm.
The projection time can be from 0.2 to 3 minutes, preferably from 0.5 to 2 minutes (in particular for a surface area of a few cm2).
Use may be made, by way of example, of a plant which can contain approximately 40 litres of particles, projected using a nozzle with a diameter of 7 to 8 mm, with compressed air under a pressure of 2.5 to 3.5 bars.
Other projection conditions can nevertheless be envisaged depending on the particles used in order to obtain the plastic deformation of the surface as described in the subject-matter of claim 5.
Advantageously, the impact surface treatment stage can be carried out under conditions suitable for obtaining covering of the carbides and/or closing of the cavities over a depth of at least 20 μm, preferably over a depth of at least 30 μm.
This mechanical stage is preferably carried out “under cold conditions”, in other words at ambient temperature, namely a temperature ranging from 18 to 35° C.
Oxidation Stage
This stage is carried out after the chemical treatment stage, optionally after the mechanical surface treatment stage. It is carried out under conditions which make it possible to form a layer of oxide(s) on the treated surface of the wall, in particular a layer containing one or more chromium oxides.
The oxidation conditions can be those generally used to form a layer of oxide(s) on this type of alloy and known from the prior art.
By way of example, the oxidation can be carried out at a temperature of 800 to 1100° C., under partial molecular oxygen pressure of 10−6 atm to 0.2 atm, for a period of time of 30 min to 5 h.
The invention is now described by means of examples and with reference to the appended non-limiting drawings, in which:
The substance to be dissolved has to be located on the anode 2 (+ terminal). The space between the two electrodes 2, 3 is, for example, approximately 1 cm. For the cathode (the − terminal), a simple metal plate can be used. The electrolyte 4 will, for example, be a sodium hydroxide solution.
Samples made of metal alloy of HP modified 25-35 type and of 35-45 type were tested. These alloys are composed of an Fe—Ni—Cr austenitic matrix within which niobium carbides (NbC) and chromium carbides (Cr7C3) precipitate. The characteristics of the metal alloys of the samples used are given in Table 1 below.
The samples used are plaques with dimensions of 8×30 mm (samples C1 to C5) and 8×25 mm (samples C6 to C9) and with a thickness of 2 mm obtained by electrical discharge machining to the core of 5 cm portions of new steam cracking tubes, with an initial thickness of 8 mm. The initial surface state is a crude machining state.
The tubes from which the tested samples result were manufactured by centrifugal casting.
Each tested sample was polished by means of SiC-based abrasive papers in the following order of fineness: 600, 800, 1200 and 2400.
Scanning electron microscopy (SEM) for observation of the surfaces and of sections. The SEMs used are the Philips XL 20 SEM and the Zeiss Supra 55 VP SEM.
Ion beam cutting: cross sections are produced by ion beam cutting with a beam of defocused ions. This technique uses accelerated argon ions to tear off the material, thus making possible a very fine and contamination-free surface polishing. The samples are adhesively bonded to titanium masks using a “silver lacquer” formed of fine silver platelets in suspension in a solvent.
In this example, the sample is subjected to an electrolytic dissolution chemical treatment.
The sample to be tested is placed at the anode of an electrolysis cell, such as described in
An electrolytic solution is prepared by dissolving, with mechanical stirring, 135 g of NaOH (in the form of pellets) in 1 l of distilled water and then the electrolysis cell is filled with the solution obtained. The chloride content of the solution is less than 10 ppm by weight.
A potential difference is applied between the anode (sample) and the cathode.
Two series of five and four tests were carried out on HP 25-35 alloys, which were all polished before being placed in the solution. The conditions used for each test are collated in Table 2.
Under the conditions tested, observation with an SEM of the surface of the samples C1 to C4 shows a dissolution of the chromium carbides Cr7C3 but no dissolution of the niobium carbides NbC. The austenitic matrix remains intact.
It will be noted that the current intensity does not appear to influence the depth of dissolution of the chromium carbides, unlike the duration of the dissolution.
Sections of the different samples were observed with an SEM.
It is thus possible to influence the depth of the carbides attacked by modifying the electrolysis conditions.
Thus, the presence of niobium carbides (NbC), of chromium carbides (Cr7C3) and of cavities is noted in
A polished HP 25-35 alloy sample is subjected to shot peening in a sleeve sandblasting chamber. The parameters used are as follows:
A sample M1 is obtained.
A polished HP 25-35 alloy sample is subjected to alumina blasting in a sleeve sandblasting chamber. The parameters used are as follows:
A sample M3 is obtained.
The sample C4 of Example 1 is subjected to the same shot peening treatment as that described in Example 2. A sample CM4 is obtained.
The sample C4 of Example 1 is subjected to the same alumina blasting treatment as that described in Example 3. A sample CM5 is obtained.
Coking tests were carried out on the samples C4, CM4 and CM5 prepared in Examples 1, 4 and 5, and also on a reference sample simply polished. The samples were brought to high temperature in the presence of a mixture of light hydrocarbons and of steam (similar to industrial conditions). They were thus subjected to conditions favouring the formation of coke.
Each sample was subjected to the following conditions:
The samples were observed with an SEM.
The samples which have been subjected to a chemical treatment exhibit overall less coke than the reference sample. Coke is still observed over approximately 10% of the surface of the sample.
A notable reduction in the amount of coke is also observed for the samples which have been subjected to a mechanical treatment after the chemical treatment (samples CM4 and CM5), as may be made out in
In this example, the sample is subjected to an electrolytic dissolution chemical treatment in order to remove the niobium carbides.
The sample to be tested is placed at the anode of an electrolysis cell of the same type as that represented in
A 7.2 mol·l−1 electrolytic solution of sulfuric acid (H2SO4) is prepared, with which the electrolysis cell is filled.
A first test was carried out on an HP 25-35 alloy with dimensions of 8×25 mm and with a thickness of 2 mm, which was polished before being placed in the sulfuric acid solution.
A potential difference of the order of 0.8 V is applied between the anode (sample) and the cathode for 2 hours. The sample is subsequently rinsed with distilled water and then with ethanol, dried and stored in a case sheltered from scratches and from the air in a desiccator.
A second test was carried out under the same electrolysis conditions on a sample with the same dimensions and of the same alloy subjected beforehand to an electrolytic dissolution of the chromium carbides. The latter is carried out with a current density of 5 A·in−2 (0.775 A·cm−2) for 2 hours in a NaOH solution (135 g in the form of pellets in 1 l of water). The sample obtained is subsequently rinsed with distilled water and then with ethanol and dried before being introduced into the sulfuric acid solution for the dissolution of the niobium carbides.
An examination of the surface state by backscattered electron scanning electron microscopy (mode of imaging sensitive to chemical contrast) shows that there was no dissolution of the niobium carbides in the case of the first test.
On the contrary, a dissolution of all the carbides (chromium and niobium) is observed by examination of the surface state of the sample subjected beforehand to a dissolution of the chromium carbides. The absence of niobium carbides (NbC) at the surface was confirmed with an SEM by an EDX (Energy Dispersive X-ray) analysis. The chemical distribution of the niobium over the surface analysed shows a few places locally rich in Nb but no niobium carbide left.
Unlike a simple dissolution in sulfuric acid, the successive electrolytic decomposition of the chromium carbides and of the niobium carbides thus makes it possible to dissolve the NbCs at the surface. Without wishing to be committed to a theory, the electrolytic dissolution of the M23C6/M7C3 might come partially “to lay bare” the NbCs and to increase the free surface area in contact with the electrolyte of the second dissolution.
| Number | Date | Country | Kind |
|---|---|---|---|
| 1662911 | Dec 2016 | FR | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2017/083566 | 12/19/2017 | WO | 00 |
