FR2966815A1 - METHOD OF PURIFYING CARBON NANOTUBES - Google Patents

Abstract

The present invention relates to a process for purifying carbon nanotubes and its applications, in particular for the manufacture of composites, electronic components and biomaterials. The method of the invention is characterized in that it comprises a step of treating a sample of carbon nanotubes in the presence of dihalogen gas and / or an interhalogen gas at 200-2000 ° C., without prior step of oxidation and without prior exfoliation of the carbonaceous multilayer shells in the sample. The present invention also relates to a device for purifying carbon nanotubes comprising: a system for introducing a dihalogenic gas and / or an interhalogen gas, and a thermal compartment in which temperatures of 200-2000 ° C can be implemented.

Description

TECHNICAL FIELD The present invention relates to a process for the purification of carbon nanotubes and its applications, in particular for the manufacture of composites, electronic components and biomaterials. BACKGROUND OF THE INVENTION The present invention also relates to a device for purifying carbon nanotubes. In the description below, references in brackets ([]) refer to the list of references after the examples. STATE OF THE ART Carbon nanotubes ("NTCs") are self-assembled nanostructures composed of rolled graphene sheets in rolls (Ijima, Nature, 1991, 354, 56-58 [ref. 1]). Nanostructures are called single-walled carbon nanotubes (SWNTs). If they are composed of a single cylindrical tube (lijima et al., Nature 1993, 363, 603-605 [ref 2] , and Bethune et al., Nature 1993, 363, 605-607 [ref. 3]). NTCs comprising two or more concentric tubes are called double-walled carbon nanotubes (DWNTs) and multi-walled carbon nanotubes (MWNTs, acronym for "multiwalled carbon nanotubes"). , respectively. Since their discovery in 1991, experimental and theoretical studies on carbon nanotubes, including single-wall nanotubes (SWNTs), have shown that they have physical properties superior to those of many materials. Their geometry is first and foremost remarkable. Their diameter is of the order of one nanometer (about 0.4 to 3 nm), while their length can reach ten millimeters. Nanotubes are also very good conductors of electrons and heat. They can thus be considered as molecular size threads. Their incorporation into electronic devices is one of the major challenges in this area. From a mechanical point of view, the NTCs have both excellent stiffness (measured by the Young's modulus), comparable to that of steel, while being extremely light. In addition, they have excellent properties of electrical and thermal conductivity which make it possible to consider using them as additives to transfer these properties to various materials, in particular macromolecular materials, such as polyamides, polycarbonates, polyesters, polystyrene, polyethylether ketones and polyethylene imine. Thus, nanotubes have mechanical properties that, coupled with their lightness, make them ideal objects to achieve high performance nano composites. For example, carbon nanotubes are commonly used in composite materials (Schaffer, MSP, Windle, AH, "Manufacturing and Characterization of Carbon Nanotube / Polyvinyl alcohol Composites", Adv Mater., 11, pp 937- 941 (1999) [ref 4]), the dihydroxy fuel cells (Ye, Y., Ahn, CC, Witham, C., Fultz, B., Liu, J., Rinzler, AG, Colbert, D., Smith, KA, Smalley, RE, "Hydrogen Absorption and 25 Cohesive Energy of Single-Walled Carbon Nanotubes", Physics App Lett., 74, pp 307-2309 (1999) [ref 5] Liu, C., Fan , YY, Liu, M., Gong, HT, Cheng, HM, Dresselhaus, MS, "Hydrogen Storage in Single-walled Carbon Nanotubes at Room Temperature", Science, 286, pp 1127-1129 (1999) [ref 6]; Kong, J., Chapline, MG, Dai, H., "Functionalized Carbon Nanotubes For Molecular Hydrogen Sensors", Adv Mater 13, 1384-1386 (2001) [ref 7]), supercapacitors (Aldissi, M. Schmitz, B. Lazaro, E. Bhamidipati, M. Dixon, B., "Conducting Polymer s ln Ultracapacitor Applications ", 56th Annu. Tech. Conf .-- Soc. Piast. Eng., (Vol 2), pp 1197-1201 (1998) [ref 8]; An, K.H .; Kim, W. S .; Park, Y. S .; Moon, J.-M .; Bae, D.

J .; Lim, S. C .; Lee, Y. S .; Lee, Y. H. "Electrochemical Properties Of High-Power Supercapacitors Using Single-walled Carbon Nanotube Electrodes", Adv. Funct. Mater. 11, pp 387-392 (2001) [ref 9]), catalysis (Yu, R., Chen, L., Liu, Q., Lin, J., Tan, K.-L., Ng, SC, Chan, HSO, Xu, G.-Q., Hor, TSA "Platinum Deposition on Carbon Nanotubes Via Chemical Modification", Chem Mater 10, pp 718-722 (1998) [ref 10] Planeix, JM, Coustel, N, Rooster, B. Brotons, V. Kumbhar, PS, Dutartre, R. Geneste, P. Bernier, P. Ajayan, PM, "Application of Carbon Nanotubes as Supports_in Heterogeneous Catalysis," J. Am. 116, pp. 7935-7936 (1994) [ref 11]) and nano-sized electronic components or systems (Tans, SJ, Verschueren, ARM, Dekker, C., Boom-Temperature Transistor Based On). Single Carbon Nanotube ", Nature 393, pp 49-52 (1998) [ref 12], Bachtold, A. Hadley, P. Nakanishi, T. Dekker, C.," Logic Circuits With Carbon Nanotube Transistors ". 294 pp. 1317-1320 (2001) [ref 13]).

However very few devices based on carbon nanotubes have emerged so far. Indeed, one of the major obstacles to their use remains the presence of impurities in the samples. The methods of production of carbon nanotubes have largely developed in recent years. The samples thus produced, however, remain heterogeneous both in terms of the presence of impurities (metal catalyst residues and carbonaceous particles other than nanotubes) than surface heterogeneities (defects, functions). These heterogeneities are a hindrance to the development of devices and new high-performance materials for applications. Synthetic methods producing nanotubes in large quantities, excluding all the processes leading to nanotubes (oriented or not) on substrates, can be grouped into two categories: "low temperature" methods (CVD). chemical vapor deposition ") and" high temperature "methods (laser ablation, electric arc). For the first category, the carbon input is provided by a gaseous mixture containing molecules of the alkane, alkene, alcohol type, etc., brought to a temperature between 500 and 800 ° C. The preparation of carbon nanotubes by decomposition of hydrocarbons or oxygen compounds in the presence of supported transition metals has been reported in the literature. The most studied route is the catalytic decomposition of methane mainly on iron-type catalyst (Muradov et al., J. Hydrogen Energy, 1993, 18: 211-215 [ref 14]), or nickel (Ishihara et al. Chem. Lett., 1995, 2: 93-94 [ref. 15]). For the second, the nanotubes are produced from graphite which is sublimed at temperatures up to 6000 ° C. The arcing method uses two graphite electrodes between which an electric arc is established, the anode is consumed to form a plasma whose temperature can reach 6000 ° C (Ijima, [ref 1]); Ebbeson, T.W., Ajayan, M.P. "Large Scale Synthesis Of Carbon Nanotubes", Nature 358, pp 220-222 (1992) [ref 16]). Laser ablation involves exposing a graphite target to laser radiation, high energy, pulsed or continuous. Graphite is either vaporized or expelled into small fragments of a few atoms (Saito et al., Chem Phys Lett, 1995, 236: 419-426 [ref 17]); Thess, A .; Lee, R .; Nikolaev, P .; Dai, H .; Petit, P .; Robert, J .; Xu, C .; Lee, Y. H .; Kim, S. G .; Rinzler, A. G .; Colbert, D.T .; Scuseria, G. E .; Tomanek, D .; Fischer, J. E. Smalley, R. E., "Crystalline Ropes Of Metallic Carbon Nanotubes". Science 273, pp 483-487 (1996) [ref 18]). In both cases (low and high temperature methods), the preparation of single-walled nanotubes requires the presence of metal catalysts. In addition, nanotubes from synthetic methods performed at high temperature have the advantage of having a narrow diameter distribution and few defects on their walls. Therefore, the development of purification procedures for this type of samples, particularly those synthesized by electric arc, is of major interest. In this case, the catalysts used are most often a nickel-yttrium mixture in a 4: 1 atomic ratio. Although efforts have been made to intensify the production of these materials, all currently known synthetic methods result in presence of large amounts of impurities in the product. For example, carbon-coated metal residues typically represent 20-30% by weight of carbon nanotubes produced by the HiPco method ("High-Pressure CO Conversion") (Nikolaev et al., Chemical Physics Letters, 1999, 313, 91-97 [ref 19]), and about 60% of carbonaceous material (ie, non-nanotube) is formed in the electric arc method. Both metal and carbon impurities can compromise the performance of carbon nanotubes in many applications. Most current methods for purifying carbon nanotubes are based on at least one of the following two steps: (1) removing the carbon layer that encapsulates the metal catalysts by oxidation with O 2, CO 2 or steam; water at temperatures of 300 to 800 ° C (for example, Chiang et al., J. Phys Chem 8 2001, 105, 1157-1161 [ref 20], Chiang et al., J. Phys Chem 8 2001, 105, 8297-8301 [ref. 21]), or by wet chemical oxidation with oxidants, including HNO3, H2O2 or KMnO4, during which sonication is frequently employed (e.g., Liu et al. Science, 1998, 280, 1253-1256 [ref 22]), and (2) using centrifugation or filtration to separate carbon nanotubes from soluble impurities (e.g., Bandow et al., J. Phys. Chem 8, 1997, 101, 8839-8842 [ref 23]).

Most often, the purification procedures generally proposed follow at least three steps. The first consists of an oxidation of the sample in order to weaken the protective carbon shells of metals to make the latter more accessible to chemical treatments. The oxidation is generally carried out under a gaseous flow containing oxygen, for example air, at temperatures below 500 ° C. If this treatment makes it possible to oxidize the less protected metal particles, it also leads to an attack on the walls of the nanotubes. Thus oxygen functions can be grafted onto the nanotubes, which creates defects in their structure. The second step is to remove the metal oxides that can be solubilized for example in acidic solutions. Heat treatment at high temperature (1100-1600 ° C) is then necessary to eliminate the functions introduced during steps 1 and 2 and restore the crystallinity of the nanotubes. Although these methods provide better samples, they have a number of limitations. They are highly sample consuming: on average 90% of the sample is lost at the end of the three stages (see Cho et al [ref 48]). These multi-step procedures are moreover relatively heavy to implement. The experimental conditions used at each stage must be optimized to improve the final yield. As a result, the number of experimental parameters is high, which makes it difficult to optimize the procedure. Moreover, they contain in all cases an initial oxidation step which inexorably attacks the walls of the nanotubes. Part of the nanotubes is thus lost and unwanted functions are introduced onto the walls of the remaining part. These functions play the role of defects in the structure of the nanotubes. The final high temperature treatment is not totally effective in eliminating these functions. Even though the oxidation of metals occurs, it only leads to their partial elimination. Indeed, certain functions protected by the multi-layer hulls remain inaccessible to the treatments. Therefore, an efficient purification process on an industrial scale to remove these impurities is essential, since most applications of carbon nanotubes require highly purified carbon nanotubes. There is therefore a real need for a method for purifying carbon nanotubes to overcome these defects, disadvantages and obstacles of the prior art, given the fact that many applications based on carbon nanotubes require purity NTCs. high. DESCRIPTION OF THE INVENTION The present invention is specifically intended to meet this need by providing a process for purifying carbon nanotubes comprising a step of treating the nanotubes in the presence of dihalogen and / or interhalogen gas. By "dihalogen gas" is meant herein a gas of the general formula X2 wherein X is F, Cl, Br or I. By "interhalogen gas" is meant herein a gas of the general formula XYn wherein X and Y each represents a different halogen atom, n represents 1, 3, 5 or 7 and X is the least electronegative of the two halogens. It may be for example interhalogen diatomic gas (such as CIF, BrCl, ICI or IBr for example), tetraatomic (such as CIF3 or BrF3 for example), hexaatomic (such as CIF5, BrF5 or IF5 for example) , or octaatomic (such as IF7). An important aspect of the process of the invention is that it does not require a prior oxidation step such as those used in the processes of the prior art. Typically, in the processes of the prior art, this oxidation step aims at (i) converting metallic catalytic impurities into metal oxides or metal hydroxides, and / or (ii) removing amorphous carbon which can be present in the carbon nanotube sample. Typically, in the processes of the prior art, this oxidation step is carried out by treating the nanotubes in an oxidizing atmosphere, preferably comprising dioxygen or carbon dioxide, preferably at a temperature> 100.degree. C., for example from 200 to 500 ° C. Thus, in one aspect, the method for purifying carbon nanotubes according to the invention comprises a step of treating the nanotubes in the presence of dihalogen and / or interhalogen gas, without prior oxidation step, in particular aimed at (i) converting metallic catalytic impurities to metal oxides or metal hydroxides, and / or (ii) removing the amorphous carbon that may be present in the nanotube sample as carbon oxides. Whatever the embodiment, in the nanotube treatment step, the temperature will depend on the catalytic impurity to be extracted (nature of the metal or metals) and the dihalogen and / or interhalogen gas used. For example, the temperature may be between -20 ° C and 2000 ° C, for example between 0 and 2000 ° C, for example between 100 and 2000 ° C, for example between 200 and 2000 ° C, for example between 300 and 2000 ° C, for example between 700 and 2000 ° C, for example between 800 and 2000 ° C, for example between 850 and 2000 ° C, for example between 900 and 2000 ° C, for example between 950 and 2000 ° C, by example between 1000 and 2000 ° C, for example between 1050 and 2000 ° C. The lower temperature limit may be lower than -20 ° C, depending on the catalytic impurities to be extracted and the dihalogen and / or interhalogen gas used. In this case, the treatment conditions may be adapted to favor the purification process. For example, the treatment time of the nanotubes in the presence of dihalogen and / or interhalogen gas may be longer. Similarly, the upper temperature limit that may be used may be greater than 2000 ° C. This depends in particular on the limits of the thermal compartment (e.g., oven) used.

For example, for iron-based catalytic impurities (e.g., single-wall carbon nanotubes synthesized by the HiPCo method, acronym for "High Pressure carbon monoxide"), a temperature around 200 to 300 ° C may for example be used. Of course, the treatment of carbon nanotubes containing ferric catalytic impurities (such as HiPCo nanotubes) may be carried out at higher temperatures, for example in the temperature ranges given above. For example, the temperature may be between 200 and 2000 ° C, for example between 300 and 2000 ° C. According to one embodiment, the step of treating the nanotubes is carried out at a high temperature, for example at a temperature> 100.degree. The temperature may be, for example between 200 and 2000 ° C, for example between 300 and 2000 ° C, for example between 700 and 2000 ° C, for example between 800 and 2000 ° C, for example between 850 and 2000 ° C, for example between 900 and 2000 ° C, for example between 950 and 2000 ° C, for example between 1000 and 2000 ° C, for example between 1050 and 2000 ° C. As discussed above, the upper temperature limit that may be used may be greater than 2000 ° C. This depends in particular on the limits of the thermal compartment (e.g., oven) used. According to one embodiment, for the process according to the invention, the treatment temperature of the nanotubes can be> 850 ° C. and can go up to the upper limit of temperature accessible by the thermal compartment in which the nanotubes of the nanotubes are contained. carbon to be purified. Thus, in one embodiment, in the nanotube treatment step, the temperature may be between 850 ° C <T <2000 ° C, preferably 900 ° C <T <2000 ° C, for example 900 ° C <_T <_1800 ° C, for example 2966815 900 ° C <_T <_1700 ° C, for example 900 ° C <_T <_1600 ° C, for example 900 ° C <_T <_ 1500 ° C, for example 900 ° C <= 1400 ° C, for example 900 ° C <1300 ° C, for example 900 ° C <1200 ° C, for example 900 ° C <1100 ° C. The process eliminates catalytic impurities polluting the carbon nanotube samples in a single chemical treatment. Carbon nanotube samples contain non-carbon impurities that inhibit the use of nanotubes for many applications. Their presence is indeed undesirable because they can induce significant changes or even mask the chemical and physical properties of the nanotubes. It is for example reported that these particles can catalyze certain parasitic reactions. Furthermore, these catalytic impurities themselves have chemical and physical properties different from those of nanotubes.

For example, the signal of the metals contained in the catalytic impurities can very strongly disturb the response of the carbon nanotubes to an electrical or magnetic stress. By "catalytic impurity" is meant herein the non-carbon impurities contained in carbon nanotubes from metal catalysts that can be used in the synthesis of nanotubes, including SWNTs. The catalytic impurities targeted by the process according to the invention are any impurity originating from metal catalysts that can be used in the synthesis of carbon nanotubes, in particular SWNTs. For example, they may be catalytic metal-based impurities, such as transition metals, lanthanides or alkaline earths. By way of example, without being limiting, mention may be made of chromium (Cr), molybdenum (Mo), tungsten (W), iron (Fe), cobalt (Co), nickel (Ni) , ruthenium (Ru), rhodium (Rh), palladium (Pd), osmium (Os), iridium (Ir), platinum (Pt), vanadium (V), yttrium (Y) ), magnesium (Mg), titanium (Ti), gold (Au), 2966815 II zirconium (Zr), lanthanides such as lanthanum (La) or holmium (Ho), and their alloys, or a mixture of these. Most often, it is catalytic impurities based on nickel, yttrium, iron, tungsten, molybdenum, palladium or cobalt, or a mixture thereof. The catalytic impurities may be in the form of pure metals, metal oxides, metal carbides, metal nitrates or other compounds containing the metal element. Preferably, the treatment of the nanotubes is not preceded by a step of exfoliation of the multilayer shells empty or filled with catalytic impurities 1 o in the sample, in particular carbonaceous multilayer shells. Typically, in the state of the art, this prior exfoliation step of the sample aims at removing at least a portion of the multilayer shells from the carbonaceous impurities contained in the samples of carbon nanotubes, in order to make accessible to the treatments these metal impurities previously encapsulated in these multilayer shells. In the state of the art, this exfoliation was considered necessary to allow the elimination of the metal impurities encapsulated by the multilayer hulls of the carbonaceous impurities of the samples of carbon nanotubes. Generally, such exfoliation is achieved by intercalating an element such as potassium between the carbonaceous layers. Removal of exfoliated carbon species is generally accomplished by chemical transformation methods (eg, wet oxidation and dissolution, filtration (eg, cross-flow or microfiltration methods), 25 and chromatographic methods (eg, exclusion chromatography) The major disadvantage of these methods is that they lead to the destruction or loss of a portion of the sample. Exfoliation is not necessary or even unnecessary.

Thus, according to any one of the embodiments, the process for purifying carbon nanotubes according to the invention may comprise a step of treating a sample of nanotubes in the presence of dihalogen and / or interhalogen gas, without exfoliation. preliminary multilayer hulls empty or filled with catalytic impurities in the sample, in particular carbonaceous multilayer hulls. For example, the process may comprise a step of treating the nanotubes in the presence of dihalogen and / or interhalogen gas, between -20 ° C. and 2000 ° C., for example between 0 and 2000 ° C., for example between 100 and 2000. ° C, for example between 200 and 2000 ° C, for example between 300 and 2000 ° C, for example between 700 and 2000 ° C, for example between 800 and 2000 ° C, for example between 850 and 2000 ° C, for example between 900 and 2000 ° C, for example between 950 and 2000 ° C, for example between 1000 and 2000 ° C, for example between 1050 and 2000 ° C, without prior exfoliation of the multilayer shells empty or filled with catalytic impurities in the sample, including carbonaceous multilayer shells. According to any one of the embodiments, the nanotube treatment step can be carried out at high temperature, and the method can comprise a nanotube treatment step in the presence of dihalogen and / or interhalogen gas, for example a temperature> _100 ° C. For example, the temperature may be between 200 and 2000 ° C, for example between 300 and 2000 ° C, for example between 700 and 2000 ° C, for example between 800 and 2000 ° C, for example between 850 and 2000 ° C, for example between 900 and 2000 ° C, for example between 950 and 2000 ° C, for example between 1000 and 2000 ° C, for example between 1050 and 2000 ° C. As discussed above, the upper temperature limit that may be used may be greater than 2000 ° C. This depends in particular on the limits of the thermal compartment (e.g., oven) used. For example, the nanotube treatment step may be carried out at 200 ° C. <300 ° C., 300 ° C. <2000 ° C., preferably 850 ° C. <2000 ° C., preferably 900 ° C. ° C <_T <_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ According to any of the embodiments, the nanotube treatment can be carried out in a continuous stream of dihalogenated and / or halogenated gas or in a static system. According to any of the previous embodiments, the dihalogen gas may be F2, C12, Br2 or 12. For example, it may be C12. The dihalogen gas can come from a commercial source, for example in the form of cylinders of F2 or C12 gas. For bromine, the Br 2 vapors can be obtained from liquid bromine which is commercially available, optionally by moderately heating to increase the flow of Br 2 gas at the desired flow rate. Similarly, the vapors of 12 can be obtained from solid diiod that is commercially available and sublimates spontaneously at 20-25 ° C. It can optionally be moderately heated to increase the flow of 12 gases at the desired rate. Alternatively, the dihalogen gas can be prepared "in situ" in the nanotube purification device. Any method compatible with a continuous flow device comprising a thermal compartment (e.g., furnace) for the treatment of nanotubes can be used. For example, the following methods may be mentioned: Difluor: Electrolysis at 100 ° C of a molten electrolytic bath composed of an anhydrous liquid mixture of potassium fluoride (KF) and hydrogen fluoride (HF). Dichlore: a. Addition of a concentrated aqueous solution of HCl to KMnO4 crystals (see Example 1). b. Electrolysis of NaCl For example, it may be an electrolysis on platinum anode and nickel cathode, molten sodium chloride. It is also possible to mention the techniques of the chlorine industry (ie, the industry which produces chlorine (C12) and a sodium hydroxide (NaOH) or potassium hydroxide (KOH), by electrolysis of a salt solution) . The main techniques used to produce chlor-alkali are electrolysis in mercury, diaphragm or membrane cells, which primarily use sodium chloride (NaCl) as a feedstock or, to a lesser extent, chloride. of potassium to produce potassium hydroxide. For example, the electrolysis of sodium chloride dissolved in water, for example on graphite anode and an iron cathode, or on graphite anode and a mercury cathode, may be used. c. Addition of a concentrated aqueous solution of hydrochloric acid on bleach. Dibrome: a. By the action of chlorine gas on a bromide For example, KBr (or NaBr or NH 4 Br) can be placed in a vessel, and gaseous dichlor gas generated elsewhere is circulated therein. Liquid and vapor dibrom is obtained according to: a. 2 KBr (s) + C12 (g) -> 2 KCl (s) + Br2 (1) b. From a bromide and manganese dioxide For example, a stoichiometric mixture of KBr (or NaBr) and MnO 2 can be produced in a two-necked flask equipped with a preferably non-isobaric dropping funnel containing Sulfuric acid, a column head with a thermometric hold, and a straight condenser with a bent elongate and an ice-bath-cooled collecting flask (if an isobaric ampule is used, the bromine formed escapes through the isobaric tube before to go to the top of the column). The reaction is carried out according to: 2 KBr (s) + 3 H2SO4 (1) + MnO2 (s) -> MnSO4 (aq) + 2 KHSO4 (aq) + 2 H2O (1) + Br2 (1) It is possible to heat moderately the flask to distil the bromine vapor. Diiode: Action of an oxidant, for example hydrogen peroxide, on iodide ions H2O2 (aq) + 21- (aq) + 2 H3O + (aq) = 12 (aq) + 4 H2O (i) Inter gas -halogen can come from a commercial source. Alternatively, it can be prepared "in situ" in the nanotube purification device by any method available to those skilled in the art. Interhalogens are obtained by reaction between two different halogens. CIF for example can be obtained by thermally activated reaction of C12 with F2. The table below summarizes the various known interhalogens: ## STR1 ## ## Greenwood, Norman N .; Earnshaw, A. (1997), Chemistry of the Elements (2nd ed.), Oxford: Butterworth-Heinemann (ref [47]). The carbon nanotubes targeted by the present process may be of the single-wall or multi-wall type. The term "carbon nanotubes" herein has the conventional meaning given to this term in the art, namely a self-assembled nanostructure composed of one or more sheets of graphene wound in a cylinder (s), depending on whether it is single-walled or multi-walled nanotubes. Thus, the nanotubes are in the form of concentric hollow tubes, possibly closed at their ends by carbon pentagons characteristic of fullerenes. The multi-walled nanotubes may be in the form of graphene planes arranged in concentric cylinders ("Russian doll" model), or in the form of a single plane of graphene wound on itself, like a sheet of paper (model " parchment "). The carbon nanotubes may be in the form of bundles or bundles, or in individualized form for example according to the method of Penicaud et al. (see FR 2 864 454 [ref 49], WO 2005/073127 [ref 50], and references 51 (Pénicaud et al.) and 52 (Vigolo et al.)), without however including nanostructures of microfiber type or carbon nanofibres which, unlike carbon nanotubes, do not have a central channel (carbon microfibers or nanofibers are typically hollow or non-hollow shapes with a diameter generally of 50 nm whose arrangement of graphene planes is not cylindrical long distance). Preferably, the nanotubes are single-walled carbon nanotubes. It is known that single-wall carbon nanotubes have properties superior to those of multiwall carbon nanotubes, which is one of the reasons why they are ideal candidates for their use in many applications. Indeed, the quasi-one-dimensional nature of single-walled nanotubes, lost when the nanotubes have several walls, is particularly at the origin of the superiority of their properties (mechanical and conductivity for example) compared to those of the latter. Single-walled carbon nanotubes therefore have a greater potential for application than multi-walled carbon nanotubes, placing them at the forefront of their interest. By way of example, the single-wall carbon nanotubes that can be used in the present process may have diameters of 0.4 to 2-3 nm. By way of example, the multi-walled carbon nanotubes that can be used in the present process may have diameters ranging from 2 to 300 nm or more. By way of example, the carbon nanotubes usable in the present process may have a length ranging from 10 nm to 1 millimeter (mm), 1 centimeter (cm), 3 cm, 5 cm or more than 5 cm. By way of example, their length / diameter ratio may advantageously be greater than 10 and most often greater than 1000. By way of example, their specific surface area is, for example, between 20 and 1000 m 2 / g and their apparent density. may especially be between 0.5 and 1.5 g / cm3 and more preferably between 1.2 and 1.4 g / cm3. By "bulk density" is meant herein the ratio of the mass of a heterogeneous set composition of discrete substances, optionally stored separately, to the total volume occupied by these substances. For carbon nanotubes according to the invention, the bulk density represents the ratio of the mass of a sample of carbon nanotubes by the total volume occupied by the sample.

By way of example, the multi-walled carbon nanotubes may for example comprise from 5 to 100 sheets and more preferably from 7 to 20 sheets. The present process is applicable to any source of commercially available NTCs, in particular by any method using one or more metal catalyst (s). An example of crude carbon nanotubes is in particular commercially available from Carbolex, Nanocyl (under the trade name NANOCYLTM NC 7000), Nanolab, CheapTubes, CNI, or ARKEMA (under the trade name Graphistrength® C100). The nanotubes can be in individualized form, or in agglomerated form in bundles or bundles. When the treatment of the nanotubes according to the present process is carried out at a temperature higher than the ambient temperature, that is to say> 25 ° C ± 5 ° C, any heating mode compatible with the process can be used. For example, it may be an oven, a filament resistor, an infrared lamp or a laser. Similarly, when the nanotube treatment according to the present process is carried out at a temperature below room temperature, ie <25 ° C ± 5 ° C, any method of refrigeration compatible with the process can to be used. For example, it may be an electric refrigeration mode (such as a refrigerator or freezer), or a chemical mode (for example, an acetone / dry ice bath or liquid nitrogen).

According to any one of the preceding embodiments, the treatment of nanotubes according to the present method can be carried out using a conventional heating device such as an oven, for example for 1 minute to 96 hours, for example from 30 minutes to 96 hours, preferably from 1 hour to 72 hours, preferably from 2 hours to 48 hours. The carbon nanotube treatment step can be carried out at a constant temperature throughout the duration of the treatment. For example, if it is an oven, it can be pre-heated to the desired temperature before the nanotube treatment step. Alternatively, the carbon nanotube treatment step can be carried out with a variable temperature profile during the treatment period, comprising at least one tray at the desired treatment temperature. For example, the temperature profile may include a temperature ramp, followed by a plateau at the processing temperature. The temperature profile may also include a temperature ramp down to the recovery temperature of the nanotubes after the processing step is complete (for example, the temperature may be reduced to 25 ° C ± 5 ° C). . According to any one of the preceding embodiments, the treatment of the nanotubes according to the present method can be carried out using a pulsed laser, continuously or quasi-continuously, for example by a series of pulses. Each pulse may be less than 1 second, or even less than 1 nanosecond, or less than 1 picosecond. In the series of pulses, the pulses can be of the same duration, or of different duration. It may be a gas laser (for example a sealed or pulsed CO2 laser TEA ("Transversely Excited Atmospheric Pressure CO2 Laser")), a chemical laser, a fiber laser, steam metallic, solid or semiconductor, of a diode-pumped laser (for example a DPSS (Diodes-Pumped Solid Laser)) or lamp-pumped laser, a UV laser, a red laser or a close laser infrared, or any type of laser that can operate in pulsed mode, continuous or quasi-continuous, capable of generating heat.

The person skilled in the art will be able to select the type of laser, the mode of operation (pulsed, continuous or quasi-continuous) and the intensity, the power and the wavelength of the irradiation, as a function of the heating conditions. desired.

The process according to the invention is not limited in terms of dihalogen and / or interhalogen gas pressure. It can be implemented both under-pressure, over-pressure and atmospheric pressure. According to any one of the preceding embodiments, the treatment of the nanotubes according to the present process can be carried out at a pressure of 1000 Pa at 1.106 Pa. Preferably, the treatment of the nanotubes according to the present process can be carried out at a pressure of 101 325 Pa (1 atm). It is known that the presence of too much water (H2O), carbon dioxide (CO2) and / or oxygen (O2) can lead to the oxidation of carbon nanotubes and damage them. Consequently, according to one embodiment, water, carbon dioxide and oxygen are preferably excluded from the gaseous atmosphere in the contact zone between the dihalogen and / or interhalogen gas and the nanotubes. Usually, the use of a gaseous atmosphere within the purification device comprising less than 5% by weight, more preferably less than 1% by weight of oxidizing compound (or mixture of oxidizing compounds) will be sufficient. For example, a gaseous atmosphere comprising less than 5% by weight, more preferably less than 1% by weight of water, carbon dioxide and / or dioxygen may be used (the% by weight refers to the sum of compounds oxidants (eg, H2O, O2 and / or CO2) present in the gaseous atmosphere). Preferably, the amount of water, carbon dioxide and oxygen will be less than 0.1% by weight. According to any one of the preceding embodiments, the dihalogen and / or interhalogen gas may be previously dried. For example, the water content of the dihalogen and / or interhalogen gas can be between 0 and 10% vol, for example between 0 and 8% vol, for example between 0 and 5% vol, for example between 0 and 3 % vol, for example between 0 and 1% vol. According to any one of the preceding embodiments, the dihalogenated and / or interhalogenated gas may be pretreated so as to reduce or eliminate traces of oxidizing compounds, such as O 2, CO 2 and / or H 2 O. For example, the content of O 2, CO 2 and / or H 2 O of the dihalogen and / or interhalogen gas may be between 0 and 10 vol%, for example between 0 and 8 vol%, for example between 0 and 5 vol%, for example, between 0 and 3% vol, for example between 0 and 1% vol. According to any one of the preceding embodiments, the dihalogen and / or interhalogen gas may be pretreated so as to reduce or even eliminate traces of hydrogen halide. This can especially be the case for H1, which breaks down into 12 and H2. For example, when the presence of H2 in the system is not desired (for example for safety reasons), the HI content of the dihalogen and / or interhalogen gas may be between 0 and 1%. For example, HI can be decomposed in a solution of acidified potassium permanganate. HI is then solubilized. Ions are oxidized to form 12 and H + reacts with MnO4 ions to form Mn2 + and water. For hydrogen halides other than H1, the presence of hydrogen halides in dihalogen and / or interhalogen is not a priori uncomfortable because they are stable. According to any of the previous embodiments, the dihalogen and / or interhalogen gas may be entrained by another gas. Any gas that does not damage carbon nanotubes and / or is compatible with the present process can act as a flushing gas. Preferably, the flushing gas is an inert gas such as nitrogen, helium, neon, argon, krypton, xenon, or mixtures of two or more of these gases. Preferably, the flushing gas is nitrogen or argon, or a mixture thereof. By "inert gas" is meant herein a gas or a gaseous mixture which does not react with carbon nanotubes, dihalogen and / or interhalogen. For example, the process is conducted under gaseous atmosphere free of oxygen. In particular, the process can be carried out under an argon or nitrogen atmosphere. According to any one of the preceding embodiments, the volume concentration of the dihalogen gas and / or the interhalogen may be between 0.1% vol and 100% vol, for example between 10 and 100% vol. for example between 30 and 100% vol, for example between 50 and 100 vol%, for example between 70 and 100 vol%, for example between 80 and 100 vol%. In the foregoing embodiments, the flow rate of the dihalogen gas and / or the interhalogen may range from 0 (static) to 100 I / min or more. In a dynamic configuration (gas flow), the maximum flow rate of the dihalogen gas and / or the interhalogen will be determined so as to avoid entrainment of the nanotubes in the gas stream. According to a variant of the invention, the process may be adapted to also eliminate, or reduce (i.e., decrease the amount), the carbonaceous impurities present in the carbon nanotube samples. Surprisingly, the inventors have discovered that this can be accomplished by introducing an oxidizing compound in gaseous form, such as dioxygen, together with the dihalogen and / or interhalogen gas, preferably in a concentration of between 10- 5% vol and 20% vol. Thus, according to any one of the preceding embodiments, it may be advantageous to introduce with the dihalogen gas and / or the interhalogen a certain amount of at least one gas capable of reducing (ie, decreasing the amount of ), or eliminate the carbon impurities contained in the sample of carbon nanotubes. For example, it may be an oxidizing compound (or mixture of oxidizing compounds) such as oxygen, carbon dioxide and / or water vapor. Thus, according to any one of the preceding embodiments, the treatment of the nanotubes may be carried out additionally in the presence of an oxidizing compound (or mixture of oxidizing compounds) such as dioxygen, carbon dioxide and / or steam. 'water. In one embodiment, the volume concentration of oxidizing compound (or mixture of oxidizing compounds) can be between 10-5% vol and 20% vol, for example 1 o between 10-3% vol and 15% vol, by example between 10-2% vol and 10% vol, for example between 10- '% vol and 5% vol, for example between 0.5% vol and 3% vol, for example between 0.5% vol and 1.5 %flight. The concentration of oxidizing compound used may be adjusted according to the treatment temperature of the carbon nanotubes in the presence of dihalogen and / or interhalogen gas. The higher the treatment temperature, the more the oxidation of the nanotubes by the oxidizing compound will be favored. This oxidation of nanotubes is of course to be avoided. The concentration of oxidizing compound will therefore be adjusted so as to limit the oxidation of the carbon nanotubes. The volume concentration of oxidizing compound (or mixture of oxidizing compounds) used will therefore be even lower than the treatment temperature will be high. Preferably, the treatment of the nanotubes may be carried out additionally in the presence of oxygen, for example in a volume concentration of between 10-5% vol and 20% vol, for example between 10-3% vol and 15% vol, for example between 10-2% vol and 25 10% vol, for example between 10- '% vol and 5% vol, for example between 0.5% vol and 3% vol, for example between 0.5% vol and 1.5 %flight. According to any of the previous embodiments, the dihalogen gas may be C12. As mentioned above, the oxidation treatments, whether they are carried out in the liquid phase (typically under reflux in HNO3, etc.) or in the gas phase in the presence of O2 or other oxidizing agent, are widely used in the field of CNTs. [ref 34, 55]. In particular, they play an important role in purification procedures because they aim to eliminate by combustion less stable carbon phases than CNTs, namely amorphous or poorly organized impurities. Thus, the methods of purification of conventional NTCs of the prior art, include an independent step of oxidation of NTCs samples most often in the gas phase using oxygen as an oxidizing species. A major drawback of these processes is the problem of controlling the selectivity of the combustion towards the carbon phases other than the CNTs. Indeed, the similarities of chemical nature and organization between the NTCs and the carbonaceous impurities present in the samples make it difficult to develop experimental conditions (eg, gaseous mixture, temperature, duration, etc.) which make it possible to obtain a combustion selectivity preserving the NTCs. Moreover, the temperature range in which CNT combustion occurs inevitably depends on the gaseous mixture used and the treatments undergone before oxidation [ref 56, 57]. From 300 ° C., CNTs can undergo structural damage due to the introduction of defects by the grafting of chemical groups on their surface in particular. When the processing conditions are more aggressive, or the higher temperatures, the removal of NTCs by combustion inevitably occurs. The combustion temperatures of the SWNTs commonly observed are generally around 400 ° C. and can in some cases reach 600 ° C. or 700 ° C. [ref 20, 58, 59]. It is therefore particularly difficult to develop experimental conditions for the oxidation treatments in order to direct the selectivity of the combustion towards the carbon phases other than the CNTs. Up to the present time, the efforts directed in this direction have remained inconclusive and elusive.

For example, the purification of SWNTs in the presence of C12 and water as an oxidizing agent has been unsuccessfully tested (see Zimmerman et al., [Ref 60]). In particular, the experimental conditions used, ie temperatures below 500 ° C., do not make it possible to obtain purified nanotubes. In addition, the treatments carried out by Zimmerman et al. under pure dichlor were ineffective under the conditions used for the purification of NTCs [ref 60]. The present invention provides a solution to this problem by providing a method for purifying NTCs involving a nanotube treatment carried out in the presence of a dihalogen or interhalogen gas, and an oxidizing compound (or mixture of oxidizing compounds) such as oxygen, carbon dioxide and / or water vapor. This variant of the process of the invention is illustrated in Example 7. Thus, quite unexpectedly, the treatment of nanotubes in an oxidizing atmosphere according to the method of the invention, especially at temperatures much higher than those reported to date in the literature, has led to an efficient purification of carbon nanotubes, with two major advantages: (i) the elimination in one step of catalytic impurities and carbonaceous impurities, (ii) by preserving the carbon nanotubes; integrity of the nanotubes (see Example 7 where temperatures of the order of 1000 ° C. and 1% vol of 02 in a C12 / N2 mixture (90% vol / 10% vol) were used). This result was unexpected in view of the work of the prior art. According to any one of the preceding embodiments, it may be advantageous to wash the purified nanotubes with an organic or inorganic solvent and / or with an acid treatment so as to eliminate the residual metal halides formed during the treatment. stage of treatment of nanotubes in the presence of dihalogen and / or interhalogen gas. This washing and / or acid treatment can be carried out with an aqueous solution of an inorganic or organic acid, for example an aqueous solution of HCl, nitric acid or sulfuric acid, hydrofluoric acid or other. They can be used pure or diluted in water at concentrations of preferably 6 to 2 mol / L. The washing and / or acid treatment conditions may be adapted to optimize the solubilization of the residual metal halides in the sample of carbon nanotubes, while avoiding the oxidation of the nanotubes. According to any one of the preceding embodiments, it may be advantageous to treat the purified nanotubes with an acidic solution in order to rid them of any residual mineral and / or carbonaceous impurities originating from their preparation process, for example carbon amorphous, graphite, fullerenes or ceramic support residues (catalytic support) that may be present in the samples of carbon nanotubes. Thus, according to any one of the preceding embodiments, the process according to the invention may further comprise a step of washing the nanotubes by acid treatment. This acid treatment may be carried out with an aqueous solution of an inorganic or organic acid, for example an aqueous solution of HCl, nitric acid or sulfuric acid, hydrofluoric acid or the like. These can be used pure or diluted in water at concentrations of preferably 6 to 2 mol / L. Preferably, the nanotubes are treated with an aqueous solution of an organic or inorganic acid at a concentration sufficiently high to attack the amorphous carbon in a practical period of time, but not high enough for the material of the carbon nanotubes to be attacked to a significant degree. In the washing operation and / or acid treatment, the weight ratio of the NTCs to the mineral acid may especially be between 1: 2 and 1: 100 or less, limits included. The washing and / or acidic treatment operation may furthermore be carried out at a temperature of from -10 to 3000 ° C, preferably from 20 to 25 ° C, for example for a period of 1 min to 48 hours. This operation may advantageously be followed by rinsing steps with water and / or with an aqueous basic solution to return to a neutral pH, and drying of the purified NTCs.

The present invention also relates to a device for purifying carbon nanotubes to implement the method according to the present invention. In one embodiment, the device comprises: a system for introducing a dihalogen and / or interhalogen gas, and a thermal compartment in which temperatures of -20 to 2000 ° C. can be used. All the previously described embodiments relating to the dihalogen and / or interhalogen gas (including its nature, its mode of introduction (eg, continuous or static flow), its origin (eg, commercial or in situ preparation), its flow rate, concentration, and purity thereof (eg, content of O2, CO2 and / or H2O "apply mutatis mutandis to the device, and are therefore not repeated here.) Thus, according to any of the embodiments relating to device, the dihalogen and / or inter-halogen gas introduction system can be adapted to allow the introduction of said gas into the device, preferably in a controlled manner, preferably in a continuous flow. dihalogen and / or inter-halogen gas may comprise a valve for controlling the flow of gas introduced into the device The introduction system may be in the form of a gas cylinder (for example F2 or C12 gas),that can be connected to the device by a suitable waterproof connection. The introduction system may be in the form of a container, connected to the device by a suitable sealed connection, possibly equipped with a valve for regulating the rate of introduction of the gas into the device, the container may contain bromine liquid or solid diiod. The container may optionally be provided with a heating system, for moderately heating the liquid dibrome or the solid diode to increase the flow of Br2 or 12 gas at the desired flow rate. Alternatively, the dihalogen and / or interhalogen gas introduction system can be adapted to allow the preparation of dihalogen and / or interhalogen gas "in situ" just before its introduction into the nanotube purification device. Thus, in one embodiment, the system for introducing dihalogen and / or interhalogen gas may comprise a device for producing said dihalogen and / or interhalogen gas. The reader will be able to refer to the discussion of gas preparation methods F2, C12, Br2 and 12 detailed above. The dihalogen and / or halogen gas feed system may comprise any suitable system for carrying out these methods of preparation. For example, when the dihalogen gas is chlorine, the dihalogen gas feed system may comprise a device as illustrated in FIG. 2, namely, a flask containing potassium permanganate crystals surmounted by a hydrogen dome. addition containing a concentrated aqueous solution of HCl, for example a 12N solution of HCl. Alternatively, when a chlorine industry technique is used to produce chlorine, the dihalogen gas feed system may include an electrolyzer. For example, it may be an asbestos membrane or diaphragm electrolyser. In a preferred embodiment, the introduction system is a chlorine introduction system. In a preferred embodiment, the dihalogen gas feed system comprises a chlorine producing device comprising a vessel containing KMnO 4 connected to an additive device for an aqueous solution of HCl, for example a concentrated aqueous solution of HCI like HCI 12N. In the same way, all the embodiments described above concerning the treatment temperature apply mutatis mutandis to the device, and are therefore not repeated here. Thus, according to one embodiment, the present invention covers a device for purifying carbon nanotubes comprising: a system for introducing a dihalogenated and / or interhalogenated gas, and a thermal compartment in which temperatures for example between 200 and 2000 ° C, for example between 300 and 2000 ° C, for example between 700 and 2000 ° C, for example between 800 and 2000 ° C, for example between 850 and 2000 ° C, for example between 900 and 2000 ° C, for example between 950 and 2000 ° C, for example between 1000 and 2000 ° C, for example between 1050 and 2000 ° C can be implemented. According to one embodiment, the present invention covers a device for purifying carbon nanotubes comprising: a system for introducing a dihalogenated and / or halogenated gas, and a thermal compartment in which temperatures <200 300 ° C, 300 ° C <2000 ° C, preferably 850 ° C <T <2000 ° C, preferably 900 ° C <2000 ° C, for example 900 ° C <T <1800 ° C, for example 900 ° C., for example 900 ° C., for example 900 ° C., 1500 ° C., for example 900 ° C., at 1400 ° C., for example 900 ° C. For example, 900 ° C. <1300 ° C., for example 900 ° C. <1200 ° C., for example 900 ° C. <1100 ° C. can be used.

According to any of the previous embodiments, the device may further comprise a compartment for purifying the dihalogen gas and / or interhalogen. For example, the purification compartment may comprise a system adapted to reduce or even eliminate traces of oxidizing compounds, such as O 2, CO 2 and / or H 2 O. For example, the purification compartment may comprise a system for trapping water, such as a column containing a desiccant such as CaCl 2, and / or a container containing a concentrated inorganic acid, such as 36N sulfuric acid, in which passes the dihalogenic gas and / or interhalogen before its introduction into the thermal compartment containing the nanotubes. The purification compartment may further comprise a system adapted to reduce or even eliminate traces of hydrogen halide (HF, HCI, HBr, HI). For example, the purification compartment may comprise a container containing a concentrated solution of NaCl, NaBr, or other alkaline halide solution into which the dihalogenated and / or interhalogenated gas is passed before it is introduced into the thermal compartment containing the nanotubes. According to any one of the preceding embodiments, the device may furthermore comprise a compartment for treating the gaseous effluent at the outlet of the thermal compartment. For example, the treatment compartment may comprise a system adapted to reduce or even avoid any backscattering of moisture in the nanotube purification device, in particular in the thermal compartment. For example, the treatment compartment may comprise a system for trapping water, such as a column containing a desiccant such as CaCl 2. The treatment compartment may further comprise a system adapted to decompose the dihalogen or interhalogen gas. For example, when the gas is chlorine, the treatment compartment may comprise a container containing a concentrated solution of sodium hydroxide (NaOH), potassium hydroxide (KOH) or calcium hydroxide (Ca (OH) 2) by for example, in which the gaseous effluent is passed to the outlet of the thermal compartment, to decompose C12 chloride and sodium hypochlorite. For the decomposition of other alkali or alkaline earth hydroxides such as Br 2, a concentrated solution of NaOH, KOH or Ca (OH) 2 for example may also be used. According to any of the previous embodiments, the device may further comprise a gas introduction compartment and / or carrier for driving the dihalogen or interhalogen gas in the purification device. Any gas that does not damage the carbon nanotubes and / or is compatible with the present process can act as a flushing gas. Preferably, the flushing gas is an inert gas such as nitrogen, helium, neon, argon, krypton, xenon, or mixtures of two or more of these gases. Preferably, the flushing gas is nitrogen or argon, or a mixture thereof. Thus, in one embodiment, the device may further include a compartment for introducing an inert carrier gas, such as nitrogen or argon. In one embodiment, the carrier gas introduction compartment ("carrier gas" in English) can take the form of a valve, in particular to control the flow of the carrier gas. According to any one of the preceding embodiments, the device may further comprise guard systems placed in the mounting of the device in order to avoid any contact between the different reagents used in the event of depression or overpressure. According to any of the previous embodiments, the device 25 may further comprise a compartment for introducing at least one oxidizing gas capable of reducing or eliminating the carbonaceous impurities contained in the sample of carbon nanotubes. For example, it may be dioxygen, carbon dioxide and / or water vapor. Of course, this compartment can be adapted to allow the introduction of the oxidizing gas in a controlled manner, for example via a valve for controlling the rate of introduction of the oxidizing gas into the device. Preferably, the device may further comprise a dioxygen introduction compartment.

The present invention also relates to carbon nanotubes that can be obtained according to a method as described above. In the present, we will denote these nanotubes "purified carbon nanotubes". The purified carbon nanotubes according to the invention can in particular be used for the reinforcement of polymeric matrices; for the manufacture of packaging materials for electronic components (for example for electromagnetic shielding and / or antistatic dissipation), such as housings for mobile telephones, computers, electronic devices on motor vehicles, rail or air ; for the manufacture of inks for the electrical connection between two electronic components; or for the manufacture of medical instruments, fuel lines (gasoline or diesel), adhesive materials, antistatic coatings, thermistors, or electroluminescent diode electrodes, photovoltaic cells or supercapacitors. The purified carbon nanotubes according to the invention can also be used as biomaterials (i.e., materials for medical use), as mechanical reinforcements, drug vectors or for the manufacture of electronic or microelectronic components. In one embodiment, the electronic or microelectronic component is a capacitor or a transistor. The present invention therefore also relates to the use of purified carbon nanotubes as defined above for the aforementioned purposes.

Thus, in one embodiment, the present invention relates to the use of so-called "purified" carbon nanotubes obtainable by a process according to the invention for the preparation of composite materials.

According to one embodiment, the composite material is obtained by a process comprising a step of mixing the purified carbon nanotubes and a solution of polymer or a mixture of polymers. According to one embodiment, the composite material is obtained by a process comprising a step of polymerization in situ of a monomer or mixture of monomers in a suspension of purified carbon nanotubes. Such polymers and their method of manufacture are described for example in Matyjaszewski, K.; Eds. ; Advances in Controlled / Living Radical Polymerization, (American Chemical Society 2003) [ref 24]; Hsieh, H. L.; Quirk, R.P .; Eds. ; Anionic Polymerization Principles and Practical Applications, (Marcel Dekker 1996) [ref 25]; Matyjaszewski, K .; Davies, T. P .; Eds .; Handbook of Radical Polymerization, (Wiley-Interscience 2002) [ref 26] or Fontaine, L.; Initiation to Macromolecular Chemistry and Physico-Chemistry (French Group for Studies and Applications of Polymers Volume 12 (Chapter 3)) [ref 27]. According to the invention, the polymer may be any polymer for carrying out the present invention. It can be chosen for example from the group comprising polystyrene; polyolefins, for example polyethylene, polypropylene, poly (alpha-olefin) s, polyisobutene and polypropylene; polyethers; polyesters; polyamides; polyacrylamides; polyacrylates; polysilanes; polysiloxanes. According to the invention, the polymer may be a linear block copolymer or a random copolymer. Those skilled in the art will be able to identify suitable operating conditions and the polymer (s) to be used to produce a composite material having the required properties. In particular, those skilled in the art can draw inspiration from the methods described in FR 2 870 251 [ref 28] and / or WO 2006/136715 [ref 29] which describe the preparation of composite materials from carbon nanotubes and polymers or polymer blends. Those skilled in the art will be able to adapt the methods described in these documents to carry out the preparation of composite materials from purified carbon nanotubes obtainable by the process of the present application. The polymer (s) may be selected in order to optimize the carbon nanotube / polymer surface interactions, and to allow better dispersion of the purified carbon nanotubes in the polymer matrix. Such materials may be used, for example, in paint or solvent formulations, in coatings, or as additives or antistatic materials. By "block copolymer" herein is meant a block polymer comprising more than one species of monomer. In a block copolymer, identical monomers are grouped together. Such polymers and their method of manufacture are described for example in Matyjaszewski, K.; Eds. ; Advances in Controlled / Living Radical Polymerization, (American Chemical Society 2003) [ref 24] or Hsieh, H. L .; Quirk, R.P .; Eds. ; Anionic Polymerization Principles and Practical Applications, (Marcel Dekker 1996) [ref 25]. By "random copolymer" is meant herein a polymer in which the different monomers are mixed according to the reactivity and concentration thereof. Such polymers and their methods of manufacture are described, for example, in Matyjaszewski, K .; Davies, T. P .; Eds .; Handbook of Radical Polymerization, (Wiley-Interscience 2002) [ref 26] or Fontaine, L.; Initiation to Macromolecular Chemistry and Physico-Chemistry (French Group for Studies and Applications of Polymers Volume 12 (Chapter 3)) [ref 27].

According to the invention, when it is a block copolymer, it may be for example a diblock copolymer synthesized for example by controlled radical polymerization or by living anionic polymerization or by living cationic polymerization or a copolymer statistic synthesized by controlled radical polymerization or uncontrolled radical polymerization. Controlled radical polymerization (CRP) is a method of choice for preparing well-defined polymers and copolymers with flexible molecular weights and low polymolecularity indices. Techniques useful in the present invention are described for example in Matyjaszewski, K .; Davies, T. P .; Eds .; Handbook of Radical Polymerization, (Wiley-Interscience 2002) [ref 26]. By "living polymerization" is meant a polymerization in which there are no termination reactions, no transfer reactions, and where the polymer chains continue to grow as long as there are monomer molecules to be added to the chains. According to the invention, the living polymerization can be cationic or anionic. Such methods are described for example in Matyjaszewski, K.; Eds. ; Cationic Polymerizations Mechanisms, Synthesis, and Applications, (Marcel Dekker 1996) [ref 30] or Hsieh, H. L.; Quirk, R.P .; Eds. ; Anionic Polymerization Principles and Practical Applications, (Marcel Dekker 1996) [ref 25]. The monomers can be introduced in full during the polymerization step. They can also be introduced separately or in a mixture, continuously or discontinuously. A monomer supplement may also be introduced at the end of the polymerization to obtain the desired polymer composition. The adjuvants that may be incorporated during conventional polymerization processes can be used according to the method of the invention. Thus, initiators, chain transfer agents, catalysts, antioxidants and lubricants known to those skilled in the art can be used. As one skilled in the art can see from the present description, one of the main advantages of the present invention is the simplicity of implementation of the process, as well as its ability to provide purified carbon nanotubes more efficiently than the processes of the prior art. In particular, the process according to the invention makes it possible to dispense with the methods of the prior art which require a prior oxidation step with the aid of an oxygen-containing compound (eg, O.sub.2, CO.sub.2, H2O) to oxidize the metal impurities into metal oxides or metal hydroxides in order to eliminate them. The major disadvantage of this type of method is that the oxidation step is not selective: the carbon nanotubes are also oxidized, and thus damaged and even destroyed, at the end of these processes. Conversely, the process according to the invention does not require this preliminary oxidation step, which is problematic. In the same way, the method of the invention makes it possible to avoid a prior step of exfoliation of empty carbon shells or filled with metallic impurities in a sample of carbon nanotubes, especially carbonaceous multilayer shells. This prior exfoliation step was recommended in the state of the art in order to expose the metal particles before eliminating them. Thus, the process according to the invention is very selective, whatever the variant used (treatment of nanotubes with a dihalogen or interhalogen gas, in the presence or absence of an oxidizing compound (or mixture of oxidizing compounds) as oxygen, carbon dioxide and / or water vapor): it makes it possible to eliminate catalytic impurities, and possibly carbonaceous impurities (in its variant using an oxidizing atmosphere), in a single step, avoiding the chemical attack of nanotubes. The method according to the invention is therefore little consumer of sample. By way of comparison, conventional methods (i.e., involving an oxidation step) typically consume more than 900/0 of the sample [ref 48]. As demonstrated in Example 8, the process according to the invention makes it possible to reduce to about 200% by weight (including the mass of catalysts removed) of sample consumption under the experimental conditions of Example 8. This is a remarkable improvement. Although the method of the invention may be enhanced by a washing step and / or acid treatment, preferably in a liquid medium, this step is not crucial to obtain nanotubes of adequate purity for typical industrial applications of nanotubes. The single step of treatment with a dihalogen and / or an interhalogen, optionally in the presence of an oxidizing compound (or mixture of oxidizing compounds), is sufficient to effectively purify the nanotubes with a remarkably improved yield compared to the prior art, and with an acceptable degree of purity. In addition, the possibility offered by the process according to the invention to overcome this step of washing / treatment in a liquid medium (a step to which the conventional methods of the prior art have recourse) makes it possible to preserve the aerated state of the sample. After treatment in a liquid medium, the liquid in which the samples have been treated must be removed so that the solid can be recovered. Most often, this is accomplished by filtration or simple heating. It is also possible to remove the solvent from the sample by lyophilization (see Maugey et al [ref 53]), but this type of method is relatively heavy to implement and is expensive. A simple filtration or drying step may seem innocuous to the sample; however, it is responsible for densifying the sample in which the nanotubes strongly aggregate irreversibly. Manipulation and the realization of subsequent chemical treatments for their dispersion or functionalization proves for example very difficult to achieve on such samples because the accessibility of the nanotube walls to the reagents is greatly reduced. Thus, maintaining the aerated state of the nanotube samples is important because it fully conditions their "processability" for future uses. A major advantage of the process according to the invention is that it preserves the structural integrity of the nanotubes: the nanotubes are not etched. The advantageous properties of the nanotubes are therefore fully preserved. In summary, the process according to the invention proposes a new route for effectively removing catalytic impurities from carbon nanotube samples. The process according to the invention also makes it possible to remove amorphous or slightly crystalline carbonaceous impurities, in particular in embodiments in which the dihalogen or interhalogen gas is used in the presence of an oxidizing compound (or mixture of oxidizing compounds). ) such as oxygen, carbon dioxide and / or water vapor. It differs from commonly used procedures and has many advantages. In particular, it avoids the chemical etching of nanotubes, is therefore of low sample consumption, and thus allows purification with better yields than conventional methods.

Other advantages may still appear to those skilled in the art on reading the examples below, illustrated by the appended figures, given for illustrative purposes.

BRIEF DESCRIPTION OF THE FIGURES FIG. 1 schematically illustrates the principle of one embodiment of the invention, namely the spontaneous elimination of metal impurities from the sample by treatment of carbon nanotubes with a dihalogen gas (X2) to form metal halides within the multi-layer carbon shells. Under the action of a high temperature, the metal halides sublimate, thus eliminating metal impurities from carbon nanotubes. - Figure 2 schematically illustrates an embodiment of the invention. This is an assembly used to carry out the heat treatment under a continuous stream of dichlor for a sample of carbon nanotubes, for example SWNTs. - Figure 3 shows a typical image obtained by transmission electron microscopy (TEM) of a sample of SWNTs after electric arc synthesis. The SWNTs are clustered in bundles of ten to several hundred nanotubes. Perpendicular to the axis of the nanotubes, these beams form a triangular network. The latter nevertheless has arrangement defects which can be eliminated by a high temperature heat treatment. In addition, at the output of the synthesis reactor, in the raw state, the walls of the nanotubes have surface functions due to the low synthesis conditions. The samples also contain impurities in a significant amount. These impurities are in the form of amorphous carbon and more or less graphitized particles, carbon shells multi-layer protecting the catalytic impurities. Because of these multi-layer protections, metal impurities are very difficult to remove from the samples. Their presence is however not desirable because they can induce significant modifications or even mask the chemical and physical properties of the nanotubes. It is for example reported that these particles can catalyze certain undesired reactions during chemical treatments. Moreover, the electrical conductivity properties or the magnetic properties of the sample can be strongly disturbed by these non-carbonaceous impurities. FIG. 4 represents an image obtained by TEM of a sample of SWNTs after synthesis by electric arc, in which the impurities are identified. FIG. 5 shows images obtained by TEM of samples of SWNTs after dichlore heat treatment according to an embodiment of the present invention. (treatment at 1100 ° C. under a flow of chlorine at 90% vol. in N 2 as a carrier gas (flow rate: 100 ml / min) for 2 hours). FIG. 6 represents images obtained by TEM of samples of SWNTs after thermal treatment under chlorine in the presence of approximately 1% vol of oxygen according to one embodiment of the present invention. (Treatment at 1100 ° C. under flow of chlorine / O 2 at 90% vol in N 2 as a carrier gas (flow rate: 100 ml / min) for 2 hours). FIG. 7 shows TEM images of a sample of carbon nanotubes after treatment at 850 ° C. under a stream of 90% vol chlorine in N 2 as a carrier gas (flow rate: 100 ml / min) for 2 hours.

Zone 1: hulls emptied after treatment; zone 2: bundles of SWNTs. FIG. 8 represents a curve Loss of mass = f (Temperature) obtained after characterization by ATG of a purified sample of SWNTs by treatment in a flow of C12 according to the invention. (See Example 3) (treatment at 1050 ° C. under a stream of chlorine at 90% vol. In N 2 as a carrier gas (flow rate: 100 ml / min) for 2 hours). FIG. 9 represents representative TEM images of samples of SWNTs purified according to the process of the invention after treatment at 800 ° C. under a stream of 90% vol chlorine in N 2 as a carrier gas (flow rate: 100 ml / min). Figure 9A: for 2 hours. Figure 9B: for 4 hours. FIG. 10 represents representative TEM images of samples of SWNTs purified according to the process of the invention after treatment at 1100 ° C. under a flow of chlorine at 90% vol in N 2 as a carrier gas (flow rate: 100 ml / min) for 2 minutes. hours. Figure 10A: Low magnification. Figure 10B: High magnification. FIG. 11 represents representative TEM images of samples of SWNTs purified according to the process of the invention after treatment at 1050 ° C. or 1100 ° C. under a stream of 90% vol chlorine in N 2 as a carrier gas (flow rate: 100 ml / min) for 2 hours. Figure 11A: 1050 ° C low magnification. Figure 11B: 1050 ° C high magnification. Figure 11C: 1100 ° C low magnification. Figure 11D: 1100 ° C high magnification. FIG. 12 represents a table listing the quantification by thermogravimetric analysis of the metal impurity concentration in a sample of SWNTs after purification by treatment in a C12 stream before and after washing the purified SWNTs in a solution of 4N HCl at room temperature for 15 hours. FIG. 13 is a graphical representation of the results of FIG. 12. Concentration of impurities of metal oxides (MXOy) obtained after ATG of a sample of SWNTs containing, before treatment under flow of dichlor, 11.6% by mass of MXOy. . After purification treatment under the conditions mentioned in the figure and before the washing step, the mass concentration of MXOy is between 50/0 and 7.50 / 0. After the washing step in a solution of 4N HCl at room temperature for 15 hours, the concentration of MXOy is around 3.50 / 0. 14 represents 3 typical RAMAN spectra of samples of SWNTs before and after purification treatment under C12 at 950 ° C for 2 hours and after washing in a 4N HCl solution at room temperature for 15 hours. In the 100-200 cm-1 wavenumber domain, the so-called RBM band for `Radial Breathing Mode 'is little modified by the different treatments and indicates the presence of SWNTs for the 3 samples. The band G linked to the vibrations of the C = C bonds of the SWNTs is typically visible around 1600 cm -1. The Raman intensity of all the spectra is normalized to that of the G band. The intensity of the D or ID band relative to the intensity of the G or IG band decreases after the purification treatment under C12 and is little modified after the washing step. This shows that the washing as described does not damage the structure of the SWNTs. FIG. 15 is a table comparing the ratios of the intensities of the D-band and the G-band noted Ip / IG for a sample of SWNTs before purification, after the heat treatment under chlorine flow and after washing. Band D is sensitive to the introduction of defects in the structure of the CNTs, the increase of the ratio Ip / IG would mean that the treatment undergone induces damage to the walls of the SWNTs. Its decrease observed in the table of Figure 15 indicates that the purification treatment improves the quality of the purified SWNTs structure and that the washing does not introduce defects on the walls of the SWNTs. EXAMPLES Unless otherwise indicated, all the experiments were carried out with the experimental set-up described in FIG. 2. EXAMPLE 1 Purification of SWNTs under a continuous flow of chlorine As illustrated in FIG. 2, the process according to the invention can be carried out in a device comprising a chlorine production zone, a chlorine purification zone, a chlorination zone and a neutralization zone for the gaseous effluent, the whole enabling purification of carbon nanotubes by treatment with dichlor continued. In this installation, the carbon nanotubes are arranged in a silica nacelle, itself placed in a silica tube in the center of a horizontal tubular furnace (Figure 2). They are gradually heated to a given temperature under a stream of chlorine gas, and then maintained at this temperature for a selected period. They are then cooled, still under current of chlorine. Dichloro gas is generated by the reaction between crystals of potassium permanganate (KMnO4) and concentrated hydrochloric acid (HCI) (Figure 2). The flow of chlorine is regulated by the acid flow. The use of a carrier gas (N2) makes it possible to establish a controlled gas flow. The gaseous mixture passes into a saturated solution of sodium chloride (NaCl), which makes it possible to remove all traces of hydrochloric acid. It is then carefully dehydrated by passing successively on a column containing calcium chloride (CaCl2) and then in a flask containing concentrated sulfuric acid (36N H 2 SO 4). The chlorine thus dehydrated then passes into the oven (700 - 1200 ° C.) in order to react with the sample. At the outlet of the oven, a column containing calcium chloride makes it possible to avoid any backscattering of moisture. The excess chlorine passes at the end of the assembly into a soda solution in order to be decomposed into chloride and sodium hypochlorite. Guard flasks are systematically placed in the assembly in order to avoid any contact between the different reagents used in this protocol in the event of depression or overpressure. The advantage of such an assembly lies in benefiting from a continuous flow of chlorine since it is directly generated during the treatment of the sample. This dynamic geometry also allows faster diffusion of the metal chlorides formed during the reaction.

The preparation of the assembly consists of the connection of the different elements and then the filling of the vials. 200 ml of saturated solution of sodium chloride (NaCl) are prepared by adding until saturation of solid NaCl to 200 ml of demineralized water and introduced into the dedicated flask; 200 ml of concentrated sulfuric acid (36N H 2 SO 4), used as received, are introduced into the dedicated flask. 30 g of sodium hydroxide (NaOH) are added to 170 g of demineralised water and solubilized with magnetic stirring and then this solution of 100% NaOH is introduced into the dedicated flask. Columns filled with CaCl2 are connected to the rest of the assembly. For a typical treatment, 250 g of potassium permanganate (KMnO4) are placed in the flask and 400 ml of hydrochloric acid (HCI, 12N) are introduced into the ampoule. The flow of chlorine is regulated by the flow rate of HCI itself controlled by a tap. The flow rate is kept constant throughout the treatment and set at 100 mL / min. The concentration of chlorine in N2 is 90% vol. A carrier gas (N2, alphagaz 2) is used which makes it possible to establish a controlled gaseous flow and to modify the concentration of chlorine.

For a typical experiment, about 200 mg of a sample of SWNTs is placed in a silica nacelle. This nacelle is placed in the center of a tubular oven and heated with a rise in temperature of 10 ° C / min at the chosen temperature (800-1100 ° C) for several hours (2h-4h). Then let the temperature drop naturally. Up to a temperature in the region of 300 ° C., the flow conditions of the chlorine / N 2 mixture remain unchanged. From 300 ° C to room temperature, the HCI valve is closed to no longer produce chlorine. The sample is left under a flow of N2 to remove the chlorine from the assembly before it is opened for the recovery of the sample.

Samples of SWNTs were purified at 800 ° C, 950 ° C, 1050 ° C or 1100 ° C, for 2 or 4 hours (see Example 5). EXAMPLE 2 Characterization of Metallic Impurities by MET Transmission electron microscopy (TEM) is a technique of choice for evaluating the efficiency of the purification method used because it allows a direct visualization of the presence or absence of metal particles. Characterization by MET has several advantages, namely: - Its excellent resolution in real space can reach the atomic scale (can go up to 0.8 Angstrom), with magnifications ranging from one thousand to one million. The possibility of combining the information obtained in real space in image mode with the information obtained in the Fourier space (or reciprocal) in diffraction mode. For crystalline samples, another way of using is to visualize the diffraction pattern of the sample. However, it should be noted that the MET allows only qualitative and in no case quantitative studies: in fact, the observed area measures a few tens of micrometers maximum and can not be generalized to the entire sample. To overcome this drawback, the number of shots has been multiplied and only a few have been selected and are considered the most representative.

The metallic particles, which come from the catalysts necessary to obtain SWNTs, observed by MET, are in the form of dark points (because they have a richer electronic structure than that of carbon) and are often covered with several carbon layers. They are therefore easily recognizable. Amorphous carbon is often arranged in blocks and forms greyish masses that can extend over several hundred nanometers. The bundles of nanotubes resemble more or less thick wires depending on the number of SWNTs they contain. The quality of their wall guarantees the low number of defects of synthesized SWNTs. Carbonated multihulls are often empty spheres but they can also be filled by the catalytic metals which are thus protected, which greatly complicates their elimination.

To prepare the samples before placing them in the microscope, the carbon nanotubes were added to an ethanol solution which was placed in an ultrasonic bath to be dispersed. Then, a drop of this liquid is deposited on a grid which itself is placed on a sample holder introduced into the MET. Figures 5, 6 and 7 are representative TEM images of the samples after the different treatments described herein. The most striking difference with the starting sample is the virtual disappearance of the particles that appear dark on these brightfield images: the catalysts. A detailed observation of the samples for treatment conditions deviating from the optimal conditions makes it possible to demonstrate the presence of empty multi-layered carbon shells (FIG. 7). Before the treatment, the latter were filled with catalytic impurities, the chlorine vapors were able to diffuse through the carbon layers and react with the metals present to form metal chlorides. Under the effect of heat, the highly volatile chlorides could sublimate and be removed from the sample leaving the shells empty (part 1 of the image, Figure 7). Moreover, we can notice that the quality of the walls of the nanotubes is preserved; chlorine does not attack the structure of nanotubes. In addition, the comparison of the images 5 and 6 shows that when oxygen is added to the C12 / N2 mixture, the carbonaceous particles have practically disappeared (FIG. 6) whereas they remain present in a significant quantity when a mixture C12 / N2 (Figure 5) is used, that is to say without addition of 02. Example 3: Characterization of the metal impurities by thermogravimetric analysis: ATG Thermogravimetric analysis is a quantitative method which makes it possible to analyze a quantity of macroscopic sample (approximately 10 mg). It consists (after a preliminary vacuum carried out on the measuring chamber) in heating in dry air from the ambient temperature to the temperature of 1000 ° C according to an identical ramp for all the samples studied (3 ° C / min). The mass loss of the sample is measured as a function of the temperature in the enclosure. The heating results in the combustion elimination of all the carbonaceous particles and also leads to the oxidation of the catalytic particles. Thus, at the end of the ATG measurement, a certain mass is recovered which corresponds only to the oxidized catalytic residues, that is to say to the nickel NiO and yttrium Y 2 O 3 oxides. Since no trace of carbon remains, smaller mass quantities measured by ATG are indicative of a higher purity sample. A crucible that will contain the sample is tared, then a few milligrams of the sample to be studied are placed inside and the crucible is introduced into the oven. After purging the chamber to remove ambient air, dry commercial air is introduced and the temperature control program is started, raising the temperature from 20 ° C to 1000 ° C with a ramp of 3 ° C / min. The mass loss curve = f (T) as shown in Figure 8 is obtained. The loss of mass has several waves depending on the stability of the carbon species present in the sample. We can see a significant mass loss around 500 ° C and another less marked around 750 ° C. Beyond this, the sample contains only the metal oxides. Subsequently, we will focus only on the final mass loss (at 1000 ° C for example). For the example given in Figure 8 the remaining mass is 4.70%.

Example 4: Characterization of SWNTs Samples by Raman Spectroscopy Raman spectroscopy is an efficient analytical technique for studying the vibration properties of CNTs that are directly related to their structural and electronic properties. Thanks to their quasi-one-dimensional structure, the response of the NTCs obtained by Raman spectroscopy is resonant, which has the advantage of amplifying their signal strongly compared to that of the impurities present in the medium and thus of directly probing the behavior of the NTCs. in samples that are far from pure.

In the range 70-1800 cm-1, several characteristic bands are to be reported: - The band RBM ("Radial Breathing Mode") in the range 100-200 cm- 'is the so-called breathing band of SWNTs. It corresponds, in fact, to the radial deformation of their diameter. Its position is also inversely proportional to the diameter of the SWNTs that resonate. - Band D (around 1340 cm-1) represents the band of defects. Its intensity will increase if the number of defects in the structure of SWNTs increases. The band G corresponds to the vibrations of the carbon-carbon bonds of the nanotube. It is characterized by two components. The band G- (low frequency component around 1560 cm -1) corresponds to vibrations along the circumferential axis of the tube. The G + band (high frequency component around 1595 cm -1) corresponds to the vibrations along the longitudinal axis of the tube.

The ratio between the intensity of the D and G bands provides an indication of the number of defects in the NTCs after a chemical treatment for example. For sample preparation for Raman analyzes, the samples are first dispersed in ethanol using a low power ultrasonic bath for 5 min.

A drop of this solution is then deposited on the surface of a glass slide and dried at room temperature. Several (average 3) Raman spectra are recorded for each sample on different areas to ensure some statistics. The sample is illuminated, via an optical microscope (objective x50), with a light beam of wavelength λ = 514.5 nm (green) from an Argon laser, whose output power is maintained at the value of 15 mW. The power of the incident laser beam is then focused on a small area of the sample, the beam diameter being about 5 pm for the objective x50. The samples being very absorbent, the incident power is decreased by optical filters chosen according to the samples and objectives used in order to overcome any changes in the spectra (or nanotubes themselves) that may be caused by a heating up too much. The Raman signal from the sample is collected by the objective of the microscope which was used to focus the incident light (backscattering geometry) and directed to the input of the spectrometer; the scattered light is finally directed, at the output of the spectrometer, to a CCD camera cooled to -70 ° C. by a Peltier effect device. For the analysis of the data, in order to bring the baseline back to zero in order to be able to compare the spectra between them, the gross spectrum is subtracted from a constant value. The recordings being made at room temperature in a non-air-conditioned room, reference spectra are systematically and regularly recorded, to check for any drifts in frequency. The recorded spectra were thus corrected by taking the line at 520.70 cm -1 of silicon as a reference. The ratio of bands D and G noted Ip / IG is determined from the heights of peaks D and G +. FIG. 15 represents a table comparing the intensity ratios of the D-band on the intensity of the G-band noted Ip / IG for a sample of SWNTs before purification, after thermal treatment under chlorine flow and after washing. The observed decrease in Ip / IG indicates that the purification treatment improves the quality of the SWNTs structure after the purification treatment and that the washing does not introduce defects in the structure of the purified SWNTs.

EXAMPLE 5 Results Obtained at Different Treatment Times and Temperatures Samples of SWNTs were purified according to Example 1 at 800 ° C, 1050 ° C or 1100 ° C, for 2 or 4 hours.

The samples thus purified were analyzed by MET (see FIGS. 9 to 11). By analyzing the images of Figures 9A and 9B, it is noted that there are no significant differences both in terms of the amount of impurities such as catalytic metals, than from the point of view of that of the nanotubes. The samples were analyzed by ATG. The results are given in the table below: Remaining mass (%) Treatment at 800 ° C for 2h 7.32 Treatment at 800 ° C for 4h 6.23 The results obtained by ATG show that when the duration of treatment is increased from 2 hours to 4 hours, the remaining catalyst mass is reduced by about 10%. Increasing the duration of treatment at this temperature improves the removal of catalytic impurities. We can therefore think that the formed chlorides take time to diffuse through the protective shells. It can be assumed, however, that increasing the duration of the treatment promotes their elimination because they can leave the sample. As illustrated in FIG. 10, the sample obtained after a treatment carried out according to Example 1 at 1100 ° C. for 2 hours has a remarkable purity. To verify the reproducibility of these results, we performed two experiments for two hours at 1050 ° C and 1100 ° C (see Figure 11). As confirmed by the TEM clichés in Figure 11, the quality of the sample is generally improved. In fact, the concentration of nanotubes is significantly increased for these high temperature experiments (1050 ° C., 1100 ° C.) compared to the experiments carried out at 800 ° C. and even at 950 ° C. In fact, the carbonaceous impurities present in a rather large quantity at the end of the treatment at 800 ° C. (see FIG. 9) were strongly removed from the samples treated at 1050.degree. C. and 1100.degree. In addition, the amount of catalytic particles also appears to decrease as the temperature is increased. These results were confirmed by ATG analysis. The results of the experiments carried out at 1050 ° C and 1100 ° C compared with those obtained at lower temperatures (800 ° C and 950 ° C) are reported in the table of Figure 12. As shown in the first column ("mass remaining before washing "), the increase in temperature improves the removal of metal impurities due to the catalysts. Indeed, their concentration (mass) i o is around 7% for the treatments at 800 and 950 ° C and it decreases to values close to 50/0 after the treatments at 1050 and 1100 ° C. EXAMPLE 6 Purification of SWNTs under a continuous stream of dichlor, followed by washing with an aqueous acid solution The carbon nanotubes are purified according to Example 1. After cooling, the nanotubes were washed with a solution of hydrochloric acid 4N concentration at room temperature, for about 15 hours to solubilize the metal chlorides that would remain trapped in the sample. The pH is then brought back to neutrality with successive washes. Typically 3 washes are required for pure water. After filtration, the sample is dried at 50 ° C for 24 to 48 hours before being analyzed. The results obtained are given in FIG. 12 (second column "mass remaining after washing") and show that this washing step, for all the conditions used, leads to a decrease in the amount of catalysts. A real influence of the washing step in hydrochloric acid is observed to improve the elimination of the catalysts. Indeed, for all the experiments carried out, the remaining mass obtained after ATG is significantly lower after washing. It is even approximately 2 times lower for the experiments carried out at 800 ° C. and 950 ° C. For treatments performed at higher temperatures, however, its effect is less pronounced (see also Figure 13). Example 7 Purification of SWNTs under a continuous flow of chlorine in the presence of oxygen The device illustrated in FIG. 2 is used.

About 200 mg of a sample of SWNTs is placed in a silica nacelle. This nacelle is placed in the center of a tubular furnace and heated with a rise in temperature of 10 ° C / min at the chosen temperature (800-1100 ° C) under a stream of dichlor gaseous containing about 1% vol of oxygen for several hours (2h-4h). The flow of oxygen from a bottle is introduced into the gaseous mixture (C12 / N2) via a flow meter at a rate of 1 mL / min for the duration of the treatment under chlorine. Then let the temperature drop naturally. The chlorine is generated as in Example 1. The oxygen is introduced by a valve provided for this purpose, from a commercial gas cylinder connected to the device. Up to a temperature in the region of 300 ° C., the flow conditions of the chlorine / O 2 / N 2 mixture remain unchanged. From 300 ° C to room temperature, the HCI valve and the oxygen valve are closed so as not to produce chlorine and no longer introduce oxygen into the system. The sample is allowed to flow under N2 to remove the chlorine and oxygen from the assembly before it is opened for sample recovery. A TEM analysis shows that this sample purified in the presence of a small amount of oxygen is of better quality in terms of content of carbonaceous impurities than the samples prepared according to Example 1, that is to say without adding oxygen. In particular, the content of carbonaceous impurities in the purified sample according to Example 7 is much lower than that obtained with the procedure of Example 1. The content of carbonaceous impurities in an electrically synthesized sample is typically 40. % vol (see Figure 3). An analysis from the images obtained by TEM (for example FIG. 6) makes it possible to estimate the volume concentration remaining after the procedure of Example 7 at less than 2% vol (FIG. 6D). EXAMPLE 8 Demonstration of the Low Consumption of SWNT Samples with the Purification Process According to the Invention A simple weighing of the sample before and after each treatment makes it possible to monitor the loss of mass due to the treatment undergone. The consumption of SWNT samples purified according to the method of the invention, as mentioned in Examples 1 to 7 above, and in Figures 5 to 7 was measured. In all cases, it is less than 200/0 by mass. Comparative Example 8A: Consumption of SWNTs with a Traditional Purification Method of the Prior Art Cho et al. [ref 48] discuss the problem of the yield of multi-step purification methods including oxidation steps, exfoliation, micro-filtration, acid treatment ... Most procedures developed do not achieve yields greater than 50% and these yields are often around 10% and can be less than 10% for the highest purities. A conventional three-step purification procedure (ie, (i) thermal oxidation under dry air, (ii) acidic treatment with 6N hydrochloric acid, and (iii) high temperature vacuum treatment) was carried out on samples of SWNTs (see Vigolo et al [ref.54]). Using optimized experimental conditions, the mass losses of the sample were 60, 90 and 15% mass. after steps (i) (oxidation), (ii) (acid treatment) and (iii) (heat treatment), respectively. Which gives a total consumption of the sample greater than 95% mass. EXAMPLE 9 Demonstration of the Aerated Aspect of SWNTs Samples with the Purification Process According to the Invention The Loss of Accessibility of the Solvent or Reagent to the Walls of Nanotubes Due to Excessive Intensification Due to the filtration of the sample after a treatment carried out in a liquid medium can be demonstrated by comparing the behavior of two purified SWNT samples: a sample which has undergone a purification procedure comprising a treatment in a liquid medium and another sample which has undergone the purification method described herein. It is possible to effectively and spontaneously disperse SWNTs directly from the synthesis in a polar solvent (for example dimethyl sulphoxide or DMSO) after they have been reduced by an alkali metal in the liquid phase (see Penicaud et al. [ref 51]) or gaseous (see Vigolo 20 et al [ref 52]). The accessibility of the metal atoms to the walls of the SWNTs completely conditions the success of the dispersion since if the alkali metal can not diffuse within the material, the reduction reaction will not be able to occur and the dispersion will not take place . For samples of SWNTs purified by a conventional method involving a liquid treatment step (e.g., reflux in an acidic solution for several hours), the nanotubes do not disperse after the alkali metal reduction treatment; they remain aggregated and precipitate in the solvent.

On the other hand, by using the purification method according to the present invention, the behavior of the purified SWNTs is similar to that of SWNTs directly derived from the synthesis. They disperse easily. This validates the advantage of the present purification method, a good dispersion of the nanotubes being difficult to obtain with other purification treatments. This advantage conferred by the process according to the present invention offers the possibility of manipulating and applying chemical treatments (reduction, functionalization, etc.) to SWNTs having been purified.

Comparative Example 9A: appearance of the sample of SWNTs having undergone a conventional purification method of the prior art The evolution of the bulk density is determined by measuring the volume occupied by three different samples of known mass. (i) A sample at the outlet of the reactor having undergone no treatment named ap_SWNTs (for `as-produced SWNTs'). (ii) a sample of SWNTs having undergone a three-step purification procedure according to reference [54], namely step 1: oxidation under dry air at 350 ° C. for 90 min; step 2: reflux in HCI (6N) for 24 hours; step 3: vacuum annealing at 1100 ° C. This sample is noted cp_SWNTs (for'commonly purified SWNTs'). (iii) a sample having undergone the procedure according to Example 1, named hp_SWNTs (for `halogen-purified SWNTs'). The calculation of the apparent density is carried out as follows. A known volume is filled with a known known mass of the sample by allowing it to occupy naturally this volume, that is to say without applying any compacting action of the powder. The comparison of the apparent densities of the carbon phase of the samples ap_SWNTs, cp_SWNTs and hp_SWNTs requires taking into account the quantity of catalysts present; this is all the more important as the catalyst concentration is different for each sample. The apparent densities obtained are presented in the table below: apparent density sample (g / cm 3) ap_SWNTs 0.058 cp_SWNTs 0.182 hp_SWNTs 0.086 Thus, for the sample having undergone the purification procedure according to Example 1, the apparent density is only slightly increased compared to that of a sample which has not undergone any treatment, whereas it is more than 3 times higher for a sample of SWNTs having undergone a standard purification procedure according to reference [54]. On reading the present application and illustrative examples above, those skilled in the art will recognize that the present process is general and is applicable to the purification of any carbon material whose synthesis involves the use of metal catalysts. .

List of References [1] Ijima, S. "Helical Microtubules of Graphitic Carbon." Nature, 1991; 354: 56-58 [2] Ijima et al., Nature, 1993; 363: 603-605 [3] Bethune et al., Nature, 1993; 363: 605-607 [4] Schaffer, M.S.P., Windle, A.H., "Manufacturing and Characterization of Carbon Nanotube / Poly (vinyl alcohol) Composites", Adv. Mater., 1999; 11: 937-941 [5] Ye, Y., Ahn, CC, Witham, C., Fultz, B., Liu, J., Rinzler, AG, Colbert, D., Smith, KA, Smalley, RE, " Hydrogen Absorption And Cohesive Energy Of Single-Walled Carbon Nanotubes ", App. Phys. Lett., 1999; 74: 307-2309 [6] Liu, C., Fan, YY, Liu, M., Gong, HT, Cheng, HM, Dresselhaus, MS, "Hydrogen Storage in Single-walled Carbon Nanotubes at Room Temperature", Science, 1999; 286: 1127-1129 [7] Kong, J., Chapline, M.G., Dai, H., "Functionalized Carbon Nanotubes For Molecular Hydrogen Sensors", Adv. Mater., 2001; 13: 1384-1386 [8] Aldissi, M .; Schmitz, B .; Lazaro, E .; Bhamidipati, M .; Dixon, B., "Conducting Polymers In Ultracapacitor Applications," 56th Annu. Tech. Conf .-- Soc. Piast. Eng., 1998; 2: 1197-1201 [9] An, K.H .; Kim, W. S .; Park, Y. S .; Moon, J.-M .; Bae, D. J .; Lim, S. C .; Lee, Y. S .; Lee, Y. H. "Electrochemical Properties Of High-Power Supercapacitors Using Single-walled Carbon Nanotube Electrodes", Adv. Funct. Mater., 2001; 11: 387-392 [10] Yu, R., Chen, L., Liu, Q., Lin, J., Tan, K.-L., Ng, SC, Chan, HSO, Xu, G.-Q ., Hor, TSA "Platinum Deposition On Carbon Nanotubes Via Chemical Modification", Chem. Mater., 1998; 10: 718-722 [11] Planeix, J. M .; Coustel, N .; Rooster, B .; Brotons, V .; Kumbhar, P. S .; Dutartre, R .; Geneste, P .; Bernier, P .; Ajayan, P.M., "Application Of Carbon Nanotubes As Supports In Heterogeneous Catalysis", J. Am. Chem. Soc., 1994; 116: 7935-7936 [12] Tans, S.J., Verschueren, A.R.M., Dekker, C., "Boom-Temperature Transistor Based On A Single Carbon Nanotube", Nature, 1998; 393: 49-52 [13] Bachtold, A .; Hadley, P .; Nakanishi, T .; Dekker, C., "Logic Circuits With Carbon Nanotube Transistors". Science, 2001; 294: 1317-1320 [14] Muradov et al., / Nt. J. Hydrogen Energy, 1993; 18: 211-215 [15] Ishihara et al., Chem. Lett., 1995; 2: 93-94 [16] Ebbeson, T.W., Ajayan, P.M. "Broad Scale Synthesis Of Carbon Nanotubes", Nature, 1992; 358: 220-222 [17] Saito et al., Chem. Phys. Lett., 1995; 236: 419-426 [18] Thess, A .; Lee, R .; Nikolaev, P .; Dai, H .; Petit, P .; Robert, J .; Xu, C .; Lee, Y. H .; Kim, S. G .; Rinzler, A. G .; Colbert, D.T .; Scuseria, G. E .; Tomanek, D .; Fischer, J. E .; Smalley, R. E., "Crystalline Ropes Of Metallic Carbon Nanotubes". Science, 1996; 273: 483-487 [19] Nikolaev et al., Chemical Physics Letters, 1999; 313: 91-97 [20] Chiang et al., J. Phys. Chem. 8, 2001; 105: 1157-1161 [21] Chiang et al. J. Phys. Chem. 8, 2001; 105: 8297-8301 [22] Liu et al. Science, 1998; 280: 1253-1256 [23] Bandow et al. J. Phys. Chem. 8, 1997; 101: 8839-8842 [24] Matyjaszewski, K.; Eds. ; Advances in Controlled / Living Radical Polymerization, (American Chemical Society 2003) [25] Hsieh, H.L .; Quirk, R.P .; Eds. ; Anionic Polymerization Principles and Practical Applications, (Marcel Dekker 1996) [26] Matyjaszewski, K .; Davies, T. P .; Eds .; Handbook of Radical Polymerization, (Wiley-Interscience 2002) [27] Fontaine, L.; Initiation to Macromolecular Chemistry and Physico-Chemistry (French Group for Studies and Applications of Polymers Volume 12 (Chapter 3)) [28] FR 2 870 251 [29] WO 2006/136715 [30] US 7,090, 819 [31] US 6,752,977 [32] US 7,135,158 [33] US 7,074,980 [34] Hou PX, Liu C., Cheng HM Multi-step purification of carbon nanotubes. Carbon, 2008; 46: 2003-25. [35] Moon J.M., An K.H., Lee Y.H., Park Y.S., Bae D.J., Park G.S. High-Yield Purification Process of Single Walled Carbon Nanotubes. J. Phys. Chem. 8, 2001; 105: 5677-81 [36] Chattopadhyay D., Galeska I., Papadimitrakopoulos F. Complete elimination of metal catalysts from single wall carbon nanotubes. Carbon, 2002; 40: 985-8 [37] Strong K.L., Anderson D.P., Lafdi K., Kuhn J.N. Purification process for single-wall carbon nanotubes. Carbon, 2003; 41: 1477-88. [38] Sen R., Richard S. M., Itkis M. F., Haddon R. Controlled Purification of Single-walled Carbon Nanotube Films by Selective Oxidation and Near-IR Spectroscopy. Chem. Materials, 2003; 15: 4273-79. [39] Martinez M.T., Callejas M.A., Benito A.M., M. Cochet, Seeger T. et al. Nanotubes to oxidative processing: structural modification, intercalation and functionalization. Carbon, 2003; 41: 2247-56. [40] Yudasaka M., Ichihashi T., Kasuya D., Kataura H., lijima S. Structure changes of single-walled carbon nanotubes and single-walled carbon nanohorns caused by heat treatment. Carbon, 2003; 41: 1273-80. [41] Furtado C.A., Kim U.J., Gutierrez H.R., Pan L., Dickley E.C., and Eklund P.C. Debundling and Dissolution of Single-Walled Carbon Nanotubes in Amide Solvents. J. Am. Chem. Soc., 2004; 126: 6095-105 [42] Harutyunyan A. R., Pradhan B. K., Chen G., Chang J., Eklund P. C. Purification of Single-Wall Carbon Nanotubes by Selective Microwave Heating of Catalyst Particles. J. Phys. Cham. 8, 2002; 106: 8671-75. [43] US 7,537,682 [44] US 7,604,790 [45] US 7,670,583 [46] WO 2008/54869 [47] Greenwood, Norman N .; Earnshaw, A. (1997), Chemistry of the Semants (2nd ed.), Oxford: Butterworth-Heinemann [48] Hyeon Gu Cho, Sang Won Kim, Hyeong Jun Lim, Chang Hoon Yun, Lee Heon Sang, Chong Rae Park. A simple and highly effective process for the purification of single-walled carbon nanotubes synthesized with arc-discharge. Carbon, 2009; 47: 3544 - 3549 [49] FR 2 864 454 [50] WO 2005/073127 [51] Alain Pénicaud, Philippe Poulin, Alain Derré, Eric Anglaret, and Pierre Petit. Spontaneous Dissolution of a Single-Wall Carbon Nanotube Salt, J. Am Cham, 2005, 127: 8-9 B. Vigolo, C. Hérold, J.-F. Maréché, P. Bourson , S. Margueron, J.

Ghanbaja, E. McRae. "Direct Revealing of the Occupation Sites of Heavy Alkali Metal Atoms in Single-Walled Carbon Nanotube Intercalation Compounds." Journal of Physical Chemistry C, 2009; 113: 17648-17654 [53] M. Maugey, W. Neri, C. Zakri, A. Derré, A. Pénicaud, L. Noé, M.

Chorro, P. Launois, M. Monthioux, and P. Poulin. "Substantial Improvement of Nanotube Processability by Freeze-Drying". Journal of Nanoscience and Nanotechnology, 2007; 7: 2633-2639 [54] B. Vigolo, C. Hérold, J.-F.Marché, J.Ghanbaja, M. Gulas, F. The Norman, R. Almairac, L. Alvarez, J.-L. Bantignies . A comprehensive scenario for commonly used purification procedures of arc-discharge as-produced single-walled carbon nanotubes. Carbon, 2010; 48: 949-963 [55] Pillai S.K. Ray S.S., Moodley M. Purification of Single-Walled Carbon Nanotubes. J. Nanosci. Nanotechnol. 2007; 7: 3011-47. [56] Landi B.J, Cress C.D., Evans C.M., Raffaelle R.P. Thermal Oxidation Profiling of Single-Walled Carbon Nanotubes. Chem. Mater., 2005; 17: 6819-34. [57] Smith Jr. M.R., Hedges S.W., LaCount R., Kern D., Shah N., Huffman G.P., Bockrath B. Selective oxidation of single-walled carbon nanotubes usingcarbon dioxide. Carbon 2003; 41: 1221-30. [58] Dillon A.C., Jones K.M., Bekkedahl T.A., Kiang C.H., Bethune D.S., Heben M.J. Storage of hydrogen in single-walled carbon nanotubes. Nature 1997; 386: 377-9. [59] Yan D.-W., Zhong, J., Wang C.-R., Wu Z.-Y. Near-edge X-ray fine absorption structure spectroscopy-assisted purification of singlewalled carbon nanotubes. Spectrochimica Acta Part B. 2007; 62: 711-6. [60] Zimmerman J.L., Bradley R.K., Huffman, C.B., Hauge R.H., Margrave J.L. Gas-Phase Purification opf Single-Wall Carbon Nanotubes. Chem. Mater., 2000; 12: 1361-6.

Claims (15)

CLAIMS1 Process for purifying carbon nanotubes comprising a step of treating the nanotubes in the presence of dihalogen gas and / or an interhalogen gas at 200-2000 ° C., without prior oxidation step and without prior exfoliation of the carbonaceous multilayer shells in the sample. 2. Method according to claim 1, wherein the temperature is between 900 and 1100 ° C 3. Method according to claim 1 or 2, wherein the treatment in the presence of dihalogen gas and / or inter-halogen is carried out in continuous flow or static system of dihalogen gas and / or interhalogen. 4. Method according to any one of claims 1 to 3, wherein the treatment in the presence of dihalogen gas and / or interhalogen is carried out for 2 to 4 hours. 5. Method according to any one of claims 1 to 4, wherein the volume concentration of dihalogen gas and / or interhalogen is between 0.1% vol and 100% vol. 2o 6. Method according to any one of claims 1 to 5, wherein the treatment of the nanotubes is carried out further in the presence of oxygen in volume concentration between 10-5% vol and 20% vol. 7. Process according to any one of claims 1 to 6, wherein the dihalogen gas is C12. The process according to any of claims 1 to 7, further comprising a step of washing the carbon nanotubes with an organic or inorganic solvent and / or using an acid treatment. 9. Method according to any one of claims 1 to 8, wherein the carbon nanotubes are single-walled nanotubes. 10. Device for purifying carbon nanotubes comprising: - a dihalogen and / or inter-halogen gas introduction system, and - a thermal compartment in which temperatures of 200-2000 ° C can be implemented. 11. Device according to claim 10, further comprising a compartment for purifying the dihalogen gas and / or interhalogen. 12. Device according to claim 10 or 11, further comprising a treatment compartment of the gaseous effluent at the outlet of the thermal compartment. 13. Device according to any one of claims 10 to 12, further comprising a compartment for introducing an inert carrier gas. The device of any one of claims 10 to 13, further comprising a dioxygen introduction compartment. 2o 15. Device according to any one of claims 10 to 14, wherein the dihalogen gas introduction system comprises a device for producing said dihalogen gas comprising a container containing KMnO4 connected to an aqueous solution addition device. of HCI, and the dihalogen gas is C12. 25