WO2009020958A2

WO2009020958A2 - Production process for single-walled carbon nanotubes

Info

Publication number: WO2009020958A2
Application number: PCT/US2008/072193
Authority: WO
Inventors: Chad J. Unrau; Richard L. Axelbaum
Original assignee: Washington University
Priority date: 2007-08-06
Filing date: 2008-08-05
Publication date: 2009-02-12
Also published as: WO2009020958A3

Abstract

An improved process for the preparation of single wall carbon nanotubes by inverse diffusion flame combustion is provided. The present invention further provides an improved inverse diffusion flame combustion apparatus and improved catalysts for the preparation of single-walled carbon nanotubes.

Description

PRODUCTION PROCESS FOR SINGLE-WALLED CARBON NANOTUBES

[0001] This invention was made with government support under grant number NCC3-1063 awarded by the National Aeronautics and Space Administration. The government has certain rights in the invention. This application claims the benefit of U.S. Provisional Application Serial No. 60/945,211, filed August 6, 2007, the entire contents of which are incorporated herein by reference.

FIELD OF THE INVENTION

[0002] The present invention generally relates to an improved inverse diffusion flame combustion process for the preparation of single-walled carbon nanotubes, improved inverse diffusion flame catalysts, and an improved inverse diffusion flame apparatus.

BACKGROUND OF THE INVENTION

[0003] Carbon nanotubes ("CNT") and single-walled carbon nanotubes ("SWNT"), collectively termed "nanotubes", possess many unique properties making them useful for applications ranging from electronic devices to composite materials. Nanotubes are used in light weight, high strength composite materials since they have a high strength to weight ratio. Depending on chirality, they also have unique electrical properties making them useful for nano-electric devices. However, the commercialization of these applications has been inhibited by the difficulty in economically synthesizing nanotubes at rates high enough for industrial scale production. In an effort to achieve such rates, many synthesis techniques have been developed including chemical vapor deposition, laser ablation, arc discharge, spray pyrolysis, the HiPCO process (as described by P. Nicolev, Journal of Nanoscience and Nanotechnology 2004, 4, 307), and flame synthesis methods including gas-phase diffusion flame synthesis .

[0004] Gas-phase diffusion flame synthesis has been unsuitable for large-scale nanotube production due to problems concerning nanotube collection. First, in conventional diffusion flame processes, nanotubes are moved into a hot, oxidizing environment after formation where they can be oxidized. Thus, the un-oxidized nanotubes cannot be collected externally from the flame. Second, polycyclic aromatic hydrocarbons ("PAH") are formed in hydrocarbon diffusion flames and deactivate the catalyst particles, thereby inhibiting or arresting nanotube growth. Third, fuel combustion in conventional methods has been inefficient resulting in incomplete fuel combustion and the generation of fine carbonaceous particles ("soot") that are an impurity on synthesized nanotubes. Industrial scale production of nanotubes has been further limited by current analytical techniques that are performed off-line thereby preventing real-time process control. Due to the large number of process variables involved in nanotube preparation, process optimization is inhibited by the lack of on-line analysis and process control.

[0005] A need remains for processes and apparatuses that reduce oxidation of synthesized nanotubes, reduce the amount of PAH and soot formed, and that have online analysis for process control thereby enabling commercial scale preparation of nanotubes.

SUMMARY OF THE INVENTION

[0006] Among the various aspects of the present invention is the provision of an improved process for the preparation of carbon nanotubes by inverse diffusion flame combustion, improved catalysts for the preparation of carbon nanotubes and an improved inverse diffusion flame combustion apparatus. [0007] Briefly, therefore, one aspect of the present invention is a process for preparing carbon nanotubes by combusting a reaction mixture in an inverse diffusion flame. The reaction mixture comprises a catalyst precursor, oxygen, and a carbon-based combustible fuel mixture comprising, for example a hydrocarbon or an alcohol. In some embodiments, the hydrogen is added to the fuel mixture. The flame has a temperature of at least 1000⁰C.

[0008] The present invention is further directed to a catalyst nanoparticle for the preparation of carbon nanotubes, the nanoparticle comprising from about 0.5 to about 95 mol percent silicon, aluminum or a mixture thereof, from about 4 to about 99 mol percent iron, and from about 0.1 to about 70 mol percent oxygen.

[0009] Another aspect of the invention is directed to a catalyst nanoparticle for the preparation of carbon nanotubes, the nanoparticle comprising iron (II) oxide. The catalyst nanoparticle has an average peak diameter of from about 1 nm to about 15 nm.

[0010] Another aspect of the invention is directed to a process for preparing carbon nanotubes, the process comprising combusting a reaction mixture in an inverse diffusion flame wherein the reaction mixture comprises at least one catalyst precursor, oxygen, and a fuel mixture comprised of a carbon- based combustible fuel and hydrogen, wherein the flame has a temperature of at least 1000⁰C.

[0011] Yet another aspect of the invention is directed to a process for preparing carbon nanotubes, the process comprising combusting a reaction mixture in an inverse diffusion flame wherein the reaction mixture comprises a carbon-based combustible fuel, at least one catalyst precursor, and oxygen and the flame has a temperature of at least 1800⁰C.

[0012] The present invention is further directed to an inverse diffusion combustion apparatus comprising a first, innermost, tube for the introduction of an oxidizer stream to an inverse diffusion combustion flame. A second tube is annular and concentric to the first tube thereby forming a first opening between the exterior surface of first tube and the interior surface of the second tube for the introduction of a first fuel stream to the inverse diffusion combustion flame through the first opening, the first opening being continuous from a first fuel stream source to the flame. A third tube is annular and concentric to the second tube thereby forming a second opening between the exterior surface of second tube and the interior surface of the third tube for the introduction of a second fuel stream to the flame through the second opening, the second opening being continuous from a second fuel stream source to the flame. A fourth tube is annular and concentric to the third tube thereby forming a third opening between the exterior surface of third tube and the interior surface of the fourth tube for the introduction of an inert gas stream to the flame through the third opening, the third opening being continuous from an inert gas stream source to the flame. The first tube or the first opening is continuous from a catalyst precursor source to the flame, or both the first tube and the first opening are continuous from a catalyst precursor source to the flame.

[0013] The present invention is further directed to an inverse diffusion combustion apparatus comprising a first, innermost, tube for the introduction of an oxidizer stream to an inverse diffusion combustion flame. A second tube is annular and concentric to the first tube thereby forming a first opening between the exterior surface of first tube and the interior surface of the second tube for the introduction of a fuel stream to the inverse diffusion combustion flame through the first opening, the first opening being continuous from a fuel stream source to the flame. A third tube is annular and concentric to the second tube thereby forming a second opening between the exterior surface of second tube and the interior surface of the third tube for the introduction of an inert gas stream to the flame through the second opening, the second opening being continuous from an inert gas stream source to the flame. The first tube or the first opening is continuous from a catalyst precursor source to the flame, or both the first tube and the first opening are continuous from a catalyst precursor source to the flame.

[0014] Other objects and features will be in part apparent and in part pointed out hereinafter.

BRIEF DESCRIPTION OF THE DRAWINGS

[0015] FIG. 1 is an illustration of an inverse flame produced with a fuel/air mixture (IA), a diluted fuel/air mixture (IB) and a diluted fuel/pure oxygen mixture (1C) .

[0016] FIG. 2 is a graphical representation of the flame illustrated in FIG. IA.

[0017] FIG. 3 is a graphical representation of the flame illustrated in FIG. 1C.

[0018] FIG. 4 is a graphical representation of the particle size distribution for nanotubes prepared under the conditions of a Z_st of 0.85, 0.74, 0.69 and a maximum adiabatic flame temperature of 1647°C wherein the graphs of FIG. 4A, 4B and 4C correspond to a Z_st of 0.85, 0.74, and 0.69, respectively.

[0019] FIG. 5 is a graph of the major species and reaction rate plotted as a function of position (X) and temperature for flames with Z_st = 0.055 (5A) and Z_st = 0.65 (5B) .

[0020] FIG. 6 is the size distribution of flame A from FIG. 4A with addition of hydrogen (6A), reduction of ferrocene (6B), and addition of air (6C) .

[0021] FIG. 7 is a depiction of an inverse diffusion flame apparatus of the invention. [0022] FIG. 8 is a graph of the particle size distribution of nanotubes prepared under the conditions of a Z_st of 0.85 and a maximum adiabatic flame temperature of 1647°C.

[0023] FIG. 9A is a graph of the particle size distribution of nanotubes prepared under the conditions of a Z_st of 0.74 and a maximum adiabatic flame temperature of 1647°C. FIG. 9B is a high resolution electron microscopy image of a single-walled carbon nanotube prepared under the conditions of FIG. 9A.

[0024] FIG. 10 is a high resolution microscopy image of encapsulated catalyst particles observed from a sample collected at a DMA setting of 8 nm.

[0025] FIG. HA and HB are high resolution microscopy images of some typical single-walled carbon nanotubes observed from a sample collected at a DMA setting of 48 nm.

[0026] FIG. 12 is a graph of the particle size distribution of nanotubes prepared under the conditions of a Z_st of 0.69 and a maximum adiabatic flame temperature of 1647°C.

[0027] FIG. 13 is a graph of the shift of the single- walled carbon nanotube mode in the size distribution of nanotubes prepared under the conditions of a Z_st of 0.74 and a maximum adiabatic flame temperature of 1647°C.

[0028] FIG. 14 is a graph of single-walled carbon nanotube length and flame temperature corrected for radiation as a function of height above the burner.

[0029] FIG. 15 is a graph of single-walled carbon nanotube growth rate as a function of height above the burner.

[0030] FIG. 16 is a graph of number concentration versus diameter for single-walled carbon nanotubes synthesized with the flame apparatus of the invention and a silicon-iron catalyst . [0031] FIG. 17 is a high resolution microscopy image of single-walled carbon nanotubes synthesized with a silicon-iron catalyst and the flame apparatus of the invention.

[0032] FIG. 18 is a high resolution microscopy image of single-walled carbon nanotubes synthesized with an aluminum- iron catalyst and the flame apparatus of the invention.

[0033] Corresponding reference characters indicate corresponding parts throughout the drawings.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0034] The present invention is directed to an improved process for the preparation of nanotubes by inverse diffusion flame combustion. The present invention is further directed to an improved inverse diffusion flame combustion apparatus and improved catalysts for the preparation of nanotubes. The present invention enables the preparation of nanotubes having lengths of over one micron at a growth rate of at least 10 μm/s. The nanotubes form on at least about 10, 20, 30, 40, 50, 60, 70, 80, 85, 90 or 95 percent by number of the catalyst nanoparticles formed during the combustion reaction. The nanotubes can be formed essentially free of soot and PAH contamination .

[0035] The process of the present invention generally involves the preparation of nanotubes by inverse diffusion flame combustion from vaporized (i.e., aerosol) feed streams comprising a carbon-based combustible fuel mixture (hereafter termed "combustible fuel"; an oxidizer; and a metal catalyst precursor. Examples of typical carbon-based combustible fuels include acetone, the hydrocarbons acetylene and ethylene, the alcohols methanol and ethanol, and mixtures thereof. In some embodiments, a mol ratio of hydrocarbon to alcohol of from 95:5 to 5:95, from 80:20 to 20:80, from 30:70 to 70:30 or even 50:50 is used. The combustible fuel mixture may be optionally diluted with a gas such as, for example, nitrogen, argon or hydrogen, and combinations thereof. Without being bound to any particular theory, and based on experimental evidence to date, it is believed that nanotube synthesis from the aerosol generally occurs by a gas aggregation mechanism wherein catalytic nanoparticles are formed in the gas phase by nucleation and growth from a supersaturated metal catalyst precursor vapor. The catalyst nanoparticles serve as nucleation sites for the growth of nanotubes. It is believed that carbon from the combusted fuel dissolves into the catalyst until a saturation concentration is reached. Thereafter, a nanotube begins to grow from the saturated catalyst with additional carbon being added onto the growing nanotube .

Nanotube Synthesis in Inverse Diffusion Flames

[0036] A challenge in diffusion flame synthesis is to minimize soot formation and PAH formation while maintaining an environment that is conducive to nanotube growth and minimizing nanotube oxidation. Soot and PAH formation, and nanotube oxidation can be minimized by using an inverse diffusion flame configuration. Oxidation that occurs in conventional flames is termed global oxidation or burnout, which occurs when the post-flame environment contains hot, oxidizing species. For nanotube synthesis, burnout should be avoided so that the products of the flame can be collected. This can be accomplished by altering the flame configuration, i.e. the location where the fuel and oxidizer are introduced.

Typical diffusion flames are established by adding combustion fuel through a central tube into a surrounding oxidizer stream such that catalyst particles and nanotubes are carried into a hot oxidizing environment. In contrast, an inverse diffusion flame is established by adding the oxidizer through a central tube into a surrounding combustion fuel, and the catalyst nanoparticles form near the flame. The flow field carries these catalyst nanoparticles into a fuel-rich region away from the hot, oxidizing environment, which may allow for nanotubes to form. The inverse flame configuration allows for catalyst nanoparticles of the optimum size and composition to be produced in an essentially soot and PAH free and fuel-rich environment, leading to the synthesis of relatively pure nanotubes at high growth rates.

[0037] It has been discovered that by a combination of oxidizer enrichment and combustion fuel dilution, the flame structure in an inverse flame configuration can be changed thereby minimizing the amount of formed soot and PAH while increasing flame temperature and maximizing nanotube growth rate and purity. "Flame structure" defines the relationship between the flame temperature and the species distributions of the fuel stream species and oxidizer stream species. Flame structure is characterized by the stoichiometric mixture fraction Z_st defined in equation (1) :

Z_st = ( 1 + Y_Y W₀ V₀ / Y₀ W_Έ V _Ε ) ^{~ λ} ( 1 ) where Y is the mass fraction, W is the molecular mass, and v is the stoichiometric coefficient (see J. Du, R. L. Axelbaum, Proc. Combust. Instit. 1996, 26, 1137) . Subscripts F and O refer to the fuel stream and oxidizer stream, respectively. Equation (1) indicates that fuel dilution and increased oxidizer concentration both act to increase Z_st. Increasing Z_st in this manner results in a significant change in the structure of a flame, which has been shown to play a primary role regarding the sooting tendency of conventional diffusion flames (see R. L. Axelbaum, W. L. Flower, CK. Law, Combust. Sci. Tech. 1988, 61, 51; R. L. Axelbaum, CK. Law, Proc. Combust. Instit. 1991, 23, 1517; B. H. Chao, S. Liu, R. L. Axelbaum, Combust. Sci. Tech. 1998, 138, 105; R. Chen, R. L. Axelbaum, Combust. Flame 2005, 142, 62; J. Du, R. L. Axelbaum, Combust. Flame 1995, 100, 367; and B. M. Kumfer, S. A. Skeen, R. Chen, R. L. Axelbaum, Combust. Flame 2006, 147, 233) .

[0038] FIG. IA illustrates experimental results for an undiluted inverse ethylene/air flame. The yellow to orange midsection and red tip indicates that a large amount of soot and PAH was formed. The FIG. IA flame has a low Z_st of 0.064 and a maximum adiabatic flame temperature ("T_ad") of 2097⁰C. The structure of the FIG. IA flame is shown graphically in FIG. 2 where a high temperature region is defined where T/ Tf is greater than 0.6, wherein T is the local flame temperature and Tf is the maximum flame temperature. The peak temperature corresponds to the location of the flame. In the high temperature region, the amount of oxidizing species present was small compared to the amount of fuel. In addition, the local atomic carbon to oxygen ratio or C/O ratio reached values much greater than one. The high C/O ratio in the high temperature region was a consequence of the flame structure and since a C/O ratio greater than unity implies that the carbon cannot be converted to gaseous CO, this type of flame readily formed soot particles. The soot gave rise to the yellow luminosity like that shown in Fig. IA.

[0039] FIG. IB illustrates experimental results for an ethylene/air flame where the ethylene stream was diluted to a blow off concentration limit of about 21% resulting in a Z_st value of 0.26 and a T_ad of 1901⁰C. The FIG. IB flame has a blue base and midsection and a red tip indicating that some soot and PAH was formed.

[0040] FIG. 1C illustrates experimental results for a flame produced by a combination of fuel stream dilution to a concentration of about 8.1% ethylene with nitrogen in combination with oxidizer enrichment wherein pure oxygen was used as the oxidizer resulting in a Z_st value of 0.78 and a T_ad of 2097⁰C. The FIG. 1C flame was completely blue indicating that it was essentially soot free although a small amount of PAH may have formed. The FIG. 1C flame had a higher adiabatic temperature than the FIG. IB flame indicating that it was more stable at a lower fuel concentration. The structure of the FIG. 1C flame is shown graphically in FIG. 3 where, again, a high temperature region is defined where T/ Tf is greater than 0.6. For this flame, the amount of oxidizing species in the high temperature region was large compared to the amount of fuel and unlike the low Z_st flame, the local C/O ratio did not reach values greater that 0.6 within the high temperature region. Thus, the effect of increasing Z_st on flame structure is to place more oxygen containing species in the high temperature region at the expense of carbon containing species. This promotes oxidation of the fuel rather than pyrolysis and soot chemistry, and leads to a flame that burns blue (due to the absence of blackbody radiation from soot) .

[0041] These experimental results indicate that by the selection of flame configuration, temperature and species concentration gradients (as indicated by elevated Z_st achieved through the combination of fuel dilution and oxidizer enrichment) essentially soot-free nanotubes can be synthesized and an environment is created that is favorable for the synthesis of high purity SWNTs at high rates over other nanotube species such as double or multi-wall CNTs.

[0042] The species concentration gradients are selected to achieve a Z_st value of from about 0.01 to about 0.95, more preferably about 0.02 to about 0.95, more preferably about 0.1 to about 0.95, more preferably from about 0.1 to about 0.85, more preferably about 0.65 to about 0.85, still more preferably from about 0.7 to about 0.8. The preferred maximum flame temperature range is from about 1000⁰C to about 3000⁰C, from about 1300⁰C to about 3000⁰C, preferably from about 1800⁰C to about 3000⁰C, for example, 1000⁰C, 1100⁰C, 1200⁰C, 1300⁰C, 1400⁰C, 1400⁰C, 1500⁰C, 1600⁰C, 1700⁰C, 1800⁰C, 1900⁰C, 2000⁰C, 2100⁰C, 2200⁰C, 2300⁰C, 2400⁰C or even 2500⁰C.

[0043] In embodiments wherein the oxidizer stream is diluted with an inert gas, a molar ratio of inert gas (as nitrogen) to oxidizer (as oxygen) of from about 0.01 to about 1.5 is preferred, more preferably from about 0.4 to about 1.2.

In embodiments wherein the combustible fuel mixture is diluted with an inert gas, a molar ratio of inert gas (as nitrogen) to combustible fuel (such as ethylene) of from about 13 to about 20 is preferred, more preferably from about 14.5 to about 18.

[0044] In some embodiments in which hydrogen is not added to the fuel stream, SWNTs having a purity of from about 1 to about 50 wt%, from about 10 to about 50 wt% or even from about 25 to about 50 wt% can be prepared. The ratio of SWNTs to non-SWNT forms is from about 0.02 to about 0.125, from about 0.05 to about 0.125 or even from about 0.1 to about 0.125.

[0045] SWNT growth rate is typically from about 10 μm per second ("μm/s") to about 100 μm/s, from about 10 μm/s to about 200 μm/s or even from about 10 μm/s to about 500 μm/s or more.

Average SWNT diameter is from about 1 nm to about 2.5 nm, from about 1 nm to about 3 nm, from about 0.7 nm to about 5 nm, from about 1 nm to about 5 nm or even from about 0.7 nm to about 10 nm. Average SWNT length is from about 200 nm to about 1 μm, from about 200 nm to about 5 μm, from about 200 nm to about 10 μm, from about 200 nm to about 50 μm or even from about 200 nm to about 100 μm.

Catalyst

[0046] Catalysts for the synthesis of SWNTs are formed in the gas phase from one or more catalyst precursors. The catalyst precursor can be introduced with the oxidizer stream or fuel stream, or can be split between the oxidizer and fuel streams. In embodiments where two or more catalyst precursor compounds are used, likewise, each catalyst precursor can be introduced with the oxidizer stream or fuel stream, or can be split between the oxidizer and fuel streams. It is believed that the composition, size, morphology, concentration and spatial distribution of the catalytic nanoparticles define the type and concentration density of the formed nanotubes.

[0047] Experimental evidence to date indicates that for a constant adiabatic flame temperature, a small change in fuel concentration can result in a change in nanotube formation and morphology. It is believed, without being bound to any particular theory, that the fuel and oxidizer concentration profiles can change the oxidation state of the catalyst particles which, in turn, can affect the characteristics of the synthesized nanotubes. In the case of an iron catalyst, experimental evidence to date indicates Z_st can affect the oxidation state of the catalyst particle. Experimental results for the size distributions for nanotubes formed in high Z_st combustion flames are depicted in FIG. 4A to 4C. In the case of iron catalysts, experimental evidence indicates that if the change in diameter of catalyst particles is attributed to oxidation of the catalyst particles and the peak at 6.5 nm (FIG. 4C) corresponds to pure iron particles, then a transition from metallic iron to iron (II) oxide to iron (III) oxide would cause the particle diameter to change from 6.5 nm to 8 nm to 13 nm. This closely corresponds to the diameters observed in Fig. 4A to 4C with Fig. 4A corresponding to metallic iron, Fig. 4B corresponding to iron (II) oxide and Fig. 4C corresponding to iron (III) oxide. Under one theory, and without being bound to any particular theory, it is believed that a change in the oxidation state of the catalyst particles produced by the three flames may be explained by examining the amount of oxygen and radicals present in the high temperature zone. It has been shown that as Z_st is increased, the amount of oxygen and radicals in the high temperature zone increases (see R. Chen, R. L. Axelbaum, Combust . Flame 2005, 142, 62) . This increase is shown experimentally in FIG. 5 for values of Z_st of 0.055 (FIG. 5A) and 0.65 (FIG. 5B) . As Z_st increases, both the oxygen concentration profile and the radical pool (OH and O) shift toward the fuel side (C2H4) where the particles are forming, and this shift would be expected to contribute to an increase in particle oxidation. Thus, the flames in FIG. 5 with values of Z_st of 0.055 and 0.65 would be associated with metallic iron and iron (III) oxide, respectively. [0048] Iron catalyst precursor compounds for the present invention generally comprise carbon and iron, preferably in a molar ratio of carbon to iron of about 5:1 to about 10:1. The compounds may optionally further comprise oxygen, nitrogen, hydrogen, and/or chlorine atoms. Examples of iron precursor compounds include, without restriction, ferrocene, iron pentacarbonyl, ferric nitrate, and iron chloride. Catalysts formed from such iron precursors generally do not contain appreciable quantities of carbon and preferably comprise from about 40 to about 100 mole percent ("mol%") iron, more preferably from about 40 to about 50 mol%. The formed catalysts can further comprise from about 0.01 to about 70 mol% oxygen, preferably from about 0.1 to about 70 mol%, more preferably from about 50 to about 60 mol%. The molar ratio of oxygen to iron is preferably from about 0 to about 1.5:1 and more preferably from about 1:1 to about 1.5:1. In some embodiments, the formed catalysts comprise from about 40 to about 60 mol% iron and from about 40 to about 60 mol% oxygen.

In some embodiments of the present invention, the catalyst precursor compound is selected from ferrocene, iron pentacarbonyl, ferric nitrate, and iron chloride and the formed catalyst comprises iron (II) oxide.

[0049] In some embodiments of the invention, the catalyst precursors include silicon, iron and carbon. The compounds may optionally further comprise oxygen and/or hydrogen atoms.

Silicon containing precursor compounds for the present invention generally have a molar ratio of carbon to silicon from about 1:1 to about 5:1, from about 1:2 to about 4:1, from about 1:2 to about 1:3, or even from about 1:2 to about 1:2.5.

Examples of silicon catalyst precursor compounds include, without restriction, alkylated siloxanes such as hexamethyldisiloxane and tetramethyldisiloxane ("TMDS") . Catalysts formed from silicon precursors and iron precursors as described herein generally do not contain appreciable quantities of carbon and preferably contain from about 0.5 to about 95 mol% silicon, more preferably from about 1 to about 95 mol% silicon, more preferably from about 1 to about 90%, even more preferably from about 15 to about 80 mol%; from about 4 to about 99 mol% iron, preferably from about 5 to about 99 mol%, more preferably from about 10 to about 99%, even more preferably from about 20 to about 85 mol%; and up to about 70 mol% oxygen, preferably from about 0.01 to about 70 mol% oxygen, more preferably from about 0.1 to about 70 mol%, more preferably from about 30 to about 50 mol%. The molar ratio of oxygen to iron in the catalyst nanoparticles is preferably from about 0 to about 1.5:1 and more preferably from about 1:1 to about 1.5:1. The molar ratio of oxygen to silicon is preferably from about 0 to about 4:1 and more preferably from about 1:1 to about 3:1. The molar ratio of silicon to iron is preferably from about 1:100 to about 10:1, more preferably from about 1:2 to about 1.5:1.

[0050] In some further embodiments of the present invention, the catalyst comprises aluminum or a mixture of aluminum and silicon. For example, 1%, 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or even all of the silicon can be replaced with aluminum. Aluminum containing precursor compounds generally comprise carbon and aluminum and have a molar ratio of carbon to aluminum from about 3:1 to about 15:1. The compounds may optionally further comprise oxygen, nitrogen, hydrogen, and/or chlorine atoms. Examples of aluminum catalyst precursor compounds include, without restriction, aluminum acetylacetonate, aluminum isopropoxide, and triethylaluminum. Catalysts formed from aluminum precursors, silicon precursors or mixtures thereof, and iron precursors as described herein contain from about 0.5 to about 95 mol% silicon, preferably from about 1 to about 95 mol%, more preferably from about 1 to about 90%, more preferably from about 10 to about 80 mol%, even more preferably from about 15 to about 80 mol%; from about 0.5 to about 95% aluminum, preferably from about 1 to about 95%, more preferably from about 0.5 to about 90%, from about 1 to about 90%, more preferably from about 0.1 to about 80%, more preferably from about 10 to about 80%, more preferably from about 15 to about 80%; from about 4 to about 99 mol% iron, preferably from about 5 to about 99 mol%, more preferably from about 10 to about 99%, even more preferably from about 20 to about 85 mol%; and up to about 70 mol% oxygen, such as from 0 to 70 mol%, preferably from about 0.01 to about 70 mol%, more preferably from about 0.1 to about 70 mol% oxygen, more preferably from about 30 to about 50 mol%. The molar ratio of oxygen to iron in the catalyst nanoparticles is preferably from about 0 to about 1.5:1 and more preferably from about 1:1 to about 1.5:1. The molar ratio of oxygen to silicon is preferably from about 0 to about 4 : 1 and more preferably from about 1:1 to about 3:1. The molar ratio of oxygen to aluminum is preferably from about 0 to about 4:1 and more preferably from about 1:1 to about 3:1. The molar ratio of silicon to iron is preferably from about 1:100 to about 10:1, more preferably from about 1:2 to about 1.5:1. Formed catalysts generally do not comprise carbon. The molar ratio of aluminum to iron is preferably from about 0.1:1 to about 10:1, more preferably from about 0.1:1 to about 1.5:1. The molar ratio of aluminum to silicon is preferably from about 0.1:1 to about 1000:1, more preferably about 0.1:1 to about 100:1. Formed catalysts generally are essentially carbon free. In some other embodiments, formed catalysts comprise from about 25 to about 35 mol% iron, from about 10 to about 15 mol% silicon, aluminum or a mixture thereof, and from about 50 to about 60 mol% oxygen. In yet other embodiments, formed catalysts comprise from about 27 to about 35 mol% iron, from about 11 to about 15 mol% silicon, from 0.1 to about 5 mol% aluminum, and from about 50 to about 60 mol% oxygen. In still other embodiments, formed catalysts comprise from about 4 to about 99 mol% iron, from about 10 to about 65 mol% silicon, from 0.1 to about 15 mol% aluminum, and from about 0.1 to about 70 mol% oxygen. Catalysts formed from aluminum and iron catalyst precursors in the absence of a silicon precursor contain from about 0.5 to about 95 mol% aluminum, more preferably from about 1 to about 90%, even more preferably from about 15 to about 80 mol%. The iron and oxygen content and the ratios of oxygen to iron, oxygen to aluminum and aluminum to iron are as described above.

[0051] Replacing some or all of the silicon in the catalyst with aluminum results in a catalyst that produces single-walled carbon nanotubes with growth rates, lengths, diameters, and purity similar to that of silicon-iron catalysts that do not contain aluminum. An example of a single-walled carbon nanotube produced from an aluminum-iron catalyst with no silicon present is shown in Fig. 18.

[0052] In some further embodiments of the present invention, the catalyst can further comprise from about 0.1 to about 5 mol%, for example, about 0.1 mol%, 0.05 mol%, 0.1 mol%, 0.5 mol%, 1 mol%, 2 mol%, 3 mol%, 4 mol% or even 5 mol% of a metal selected from Group IIIA, Group IB, Group VB, Group VIB, Group VIIB, iron, cobalt, nickel, lanthanide series metals, and combinations thereof. In one embodiment, the metals are selected from copper, silver, vanadium, chromium, molybdenum, tungsten, manganese, cobalt, nickel, cerium, and combinations thereof.

[0053] Preferably the catalyst particles are nanoparticles having an average diameter ranging of from about 1 nm to about 15 nm.

[0054] Thus, with the appropriate flame configuration, flame temperature and Z_st, catalyst particles with optimal size and composition can be formed in a soot-free, fuel-rich environment, which leads to the formation of nanotubes.

Hydrogen Addition

[0055] In some embodiments of the invention, a small about of hydrogen or a combination of a low concentration of hydrogen and an inert gas such as nitrogen, helium or argon is used to dilute the combustible fuel mixture to affect the oxidation state of catalysts comprising iron. A small amount of hydrogen refers to less than 1 mol%. Hydrogen is believed to reduce catalyst surface oxidation and thus addition of the appropriate amount of hydrogen to the flame of FIG. 4A may increase the number of catalyst nanoparticles resulting in nanotubes in a manner comparable to those illustrated in FIG. 4B. Iron (II) oxide is preferred. A molar ratio of inert (as nitrogen) to hydrogen of about 90:1 to about 110:1 is preferred, more preferably from about 95:1 to about 105:1. In some other embodiments of the invention, at least 1%, 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95% or even all of the inert gas is replaced by hydrogen. In one embodiment, no inert gas is used thereby resulting in feed streams comprising: an admixture of a combustible fuel, preferably ethylene and/or acetylene and hydrogen; and an oxidizer, preferably oxygen. It has been discovered that, in addition to optimization of the catalyst oxidation state (i.e., predominantly iron (II) oxide), replacement of inert gasses with hydrogen for combustible fuel dilution increases the maximum flame temperature to about 1800⁰C to about 3000⁰C, and minimizes soot and PAH formation. The combination of optimized oxidation state and high temperature has been found to experimentally result in SWNT growth at a rate of at least 100 μm per second. Preferably, the maximum flame temperature is from about 1000⁰C to about 3000⁰C, from about 1300⁰C to about 3000⁰C, from about 1700⁰C to about 3000⁰C, from about 1800⁰C to about 3000⁰C, from about 1900⁰C to about 3000⁰C, or even from about 2000⁰C to about 3000⁰C, for example 1000⁰C, 1100⁰C, 1200⁰C, 1300⁰C, 1400⁰C, 1400⁰C, 1500⁰C, 1600⁰C, 1700⁰C, 1800⁰C, 1900⁰C, 2000⁰C, 2100⁰C, 2200⁰C, 2300⁰C, 2400⁰C or even 2500⁰C. In some embodiments, where hydrogen is added to the combustible fuel mixture, Z_st is preferably from about 0.01 to about 0.95, more preferably about 0.02 to about 0.95, more preferably from about 0.1 to about 0.95, more preferably from about 0.1 to about 0.85, more preferably about 0.1 to about 0.6, still more preferably from about 0.1 to about 0.4. Hydrogen is preferably added to the combustible fuel mixture in a molar ratio of hydrogen to combustible fuel of from about 0.1:1 to about 200:1, preferably from about 3:1 to about 200:1, more preferably about 20:1 to about 180:1. Examples of ratios are 0.1:1, 3:1 and 200:1. In various embodiments, the concentration of hydrogen may be greater than 1 mol% in order achieve the desired molar ratio of hydrogen to combustible fuel .

[0056] Using hydrogen addition to the inverse diffusion combustion reaction, SWNTs having a purity of from 1 to 95 wt%, from 50 to 95 wt% or even from 70 to 95% or more can be prepared. The ratio of SWNTs to non-SWNT forms is from 1:50 to 10:1, from 1:1 to 10:1 or even from 5:1 to 10:1. SWNT growth rate is from 10 μm/s to 500 μm/s, from 10 μm/s to 1,000 μm/s or even from 10 μm/s to 10,000 μm/s. Average SWNT diameter is from 0.7 nm to 3 nm, from 0.7 nm to 4 nm or even from 0.7 nm to 5 nm. Average SWNT length is from 200 nm to 5 μm, from 200 nm to 10 μm or even from 200 nm to 100 μm.

Flame Apparatus

[0057] As depicted in FIG. 7, an improved inverse flame diffusion burner apparatus is provided that enables the production of nanotubes at high purity and velocity. The burner comprises a first, innermost, tube 1 for the introduction of an oxidizer stream 17 and optionally a catalyst precursor stream 18 into the combustion process flame 50. The first tube 1 is continuous from an oxidizer source and optionally a catalyst precursor stream source to flame 50.

The first tube 1 is located interior to a second tube 20, tube 20 being annular and concentric to tube 1, thereby forming an opening 25 between the exterior surface of tube 1 and the interior surface of tube 20. A first fuel stream 27 and optionally a catalyst precursor stream 28 are introduced into the combustion process through opening 25, opening 25 being continuous from a first fuel stream source and optionally a catalyst precursor stream source to flame 50. The second tube 20 is located interior to a third tube 30, tube 30 being annular and concentric to tube 20, thereby forming an opening 35 between the exterior surface of tube 20 and the interior surface of tube 30. A second fuel stream 37 is introduced into the combustion process through opening 35, opening 35 being continuous from a second fuel stream source to flame 50. The third tube 30 is located interior to a fourth tube 40, tube 40 being annular and concentric to tube 30, thereby forming an opening 45 between the exterior surface of tube 30 and the interior surface of tube 40. An inert gas stream 47 is introduced into the combustion process through opening 45, opening 45 being continuous from an inert gas stream source to flame 50.

[0058] Preferably, oxidizer stream 17 comprises oxygen or an admixture of oxygen and an inert gas such as argon or nitrogen. A molar ratio of inert gas (as nitrogen) to oxidizer (as oxygen) of from about 0 to about 1.5:1 is preferred, more preferably from about 0 to about 1:1. The flame temperature is affected by the amount of inert gas in oxidizer stream 17. The first fuel stream 27 comprises a gas comprising combustible fuel or a mixture of hydrogen and a combustible fuel. In one embodiment, first fuel stream 27 further comprises an inert gas such as nitrogen or argon wherein the molar ratio of inert gas to combustible fuel is from about 1:1,000 to about 33:1, more preferably about 1:1,000 to about 20:1. Hydrogen is preferably at least 90% pure, at least 95% pure or at least 99% pure. Where an admixture of hydrogen and a combustible fuel gas is used, a molar ratio of hydrogen to combustible fuel gas of from about 0.1:1 to about 200:1 is preferred, more preferably from about 3:1 to about 200:1, more preferably about 20:1 to about 180:1.

At least one of catalyst precursor streams 18 and 28 are present. Those streams comprise catalyst precursors, as described above, such as silicon and iron. The catalyst precursor stream 18 and/or 28 and the first fuel stream 27 are introduced in a small region near the flame. In that region, the precursors are quickly decomposed near the base of the flame thereby confining the formed catalyst nanoparticles to that region. It is believed that the microenvironment of that region exposes the catalysts to a uniform environment thereby favoring the formation of catalysts that are more likely to form nanotubes, the tubes being substantially homogenous in terms of length and diameter. If the precursors were instead introduced through openings 35 and 45, more catalyst particles would form in a broader region and away from the base of the flame resulting in heterogeneity. It is believed that nanotubes formed from such catalysts would have non- homogeneous particle size characteristics as compared to nanotubes formed near the base of the flame and a lower proportion of catalysts would form nanotubes. The catalyst particles not forming nanotubes would be present as an impurity on nanotubes that do form. A ratio of the diameter of tube 1 to tube 20 of from about 0.4:1 to about 0.95:1 is preferred. The second fuel stream 37 preferably comprises an admixture of hydrogen and a combustible fuel as described above, such as ethylene or acetylene. Inert gas stream 47 is introduced in the outermost opening 45 so that the flame is surrounded by inert gas thereby preventing oxidation of the nanotubes. As described above, for embodiments where hydrogen is added to the combustible fuel, the species concentration gradients are selected to achieve a Z_st value of from about 0.01 to about 0.95, more preferably about 0.02 to about 0.95, more preferably from about 0.1 to about 0.95, more preferably from about 0.1 to about 0.85, more preferably about 0.1 to about 0.6, still more preferably from about 0.1 to about 0.4.

The preferred maximum flame temperature range is as described above, e.g., most broadly from about 1000⁰C to about 3000⁰C. The flame can be laminar flow to fully turbulent flow.

[0059] In one alternative embodiment to the apparatus described immediately above, catalyst precursor stream 28 is present and the hydrogen and/or combustible fuel of first fuel stream 27 is replaced with an inert carrier gas.

[0060] In one embodiment, in reference to FIG. 7, tube 30, opening 45, and second fuel stream 37 are eliminated and second tube 20 is instead located interior to tube 40, tube 40 being annular and concentric to tube 20, thereby forming an opening 35 between the exterior surface of tube 20 and the interior surface of tube 40. The resulting burner comprises a first, innermost, tube 1 for the introduction of an oxidizer stream 17 and optionally a catalyst precursor stream 18 into the combustion process flame 50. The first tube 1 is located interior to a second tube 20, tube 20 being annular and concentric to tube 1, thereby forming an opening 25 between the exterior surface of tube 1 and the interior surface of tube 20. A fuel stream 27 and optionally a catalyst precursor stream 28 are introduced into the combustion process through opening 25, opening 25 being continuous from a fuel stream source and optionally a catalyst precursor stream source to flame 50. Fuel stream 27 comprises a gas comprising a mixture of hydrogen and the entire portion of the combustible fuel. Fuel stream 27 optionally further comprises an inert gas. Catalyst precursor streams 18 and 28 comprise catalyst precursors, such as silicon and iron, and at least one of streams 18 and 28 are present. An inert gas stream 47 is introduced into the combustion process through opening 35, opening 35 being continuous from an inert gas stream source to flame 50. Preferred Z_st values, gas ratios and flame temperatures are as described above. The flame can be laminar flow to fully turbulent flow.

Online Analysis

[0061] In another embodiment of the present invention, the above described embodiments, alone or in combination, can be coupled with online nanotube analysis to provide a basis for process control for nanotube synthesis. In particular, online analysis can be used to collect real time analytical measurements of nanotube formation, size and purity. The analytical measurements can then be used as inputs to a process control algorithm that compares those measurements versus predetermined set points and generates a set of outputs that continually adjust process parameters such as, for example, combustible fuel flow rate, oxidizer flow rate, catalyst precursor flow rate, inert gas flow rate, flow rate ratios of one more of those streams, and flame temperature in order to prepare nanotubes having the desired characteristics.

[0062] A differential mobility analyzer ("DMA") is an example of an online analytical instrument suitable for the practice of the present invention. The DMA uses particle electrical mobility to measure size distribution. The electrical mobility of a non-spherical particle is given by equation (2)

Z = neC(d_a) /3πμχ₀d_v (2) where n is the number of charges, e is the elementary charge, C(d_a) is the slip correction factor based on an adjusted sphere diameter, μ is the viscosity, χ₀ is the dynamic shape factor, and d_v is the equivalent volume diameter equal to (1.5D²L) ^H where D and L are the diameter and length of the nanotube, respectively (see Cheng, Y. S., Chem Eng. Commun . 108 (1991) 201-223) . The shape factor and the adjusted sphere diameter are both dependent on orientation relative to the direction of flow. Thus, electrical mobility of a nanotube is a function of its shape, orientation, diameter, length, and the number of charges carried by the nanotube.

[0063] In aerosol processes, some of the catalyst particles will not produce nanotubes and therefore catalyst particles should be classified separately from nanotubes. The spherical catalyst particles and the nanotubes often have similar diameters and, if they also carry a similar amount of charge, nanotubes will have a lower mobility than the catalyst particles because their aspect ratios are much greater than unity. The difference in mobility translates to different DMA classification voltages for the two types of particles. Thus, the formation of catalysts and nanotubes can be identified with the DMA. This difference in classification voltage has been theoretically demonstrated (see CJ. Unrau, R. L. Axelbaum, P. Biswas, P. Fraundorf in: T. George, G. Zhang, L. Assoufid, G. AIi Mansoori (Eds.), Molecular Building Blocks for Nanotechnology: From Diamondoids to Nanoscale Materials and Applications, Springer-Verlag, New York (2007)) and has been observed experimentally in flow reactor synthesis of nanotubes (see A. G. Nasibulin, A. Moisala, D. P. Brown, H. Jiang, E.I. Kauppinen, Chem. Phys . Lett. 402 (1-3) (2205) 227- 232; and S. H. Kim, M. R. Zacharial, Nanotechnology 16 (10) (2005) 2149-2152) .

[0064] The amount of impurities present on the synthesized nanotubes can also be determined from the size distribution measured by the DMA. If both inactive catalyst particles and nanotubes are present, the size distribution will likely be bimodal . The ratio of the bimodal peaks can indicate the amount of catalyst impurities on the synthesized nanotubes. More catalyst impurities will exist on the nanotubes if a large fraction of the catalyst particles do not produce nanotubes. Finally, the DMA can provide online information about nanotube size. In particular, if the diameter in equation (2) is known or can be estimated from, for example, the catalyst particle size distribution, the number of charges can be calculated (see CJ. Unrau, R. L. Axelbaum, P. Biswas, P. Fraundorf in: T. George, G. Zhang, L. Assoufid, G. AIi Mansoori (Eds.), Molecular Building Blocks for Nanotechnology: From Diamondoids to Nanoscale Materials and Applications, Springer-Verlag, New York (in press)) .

[0065] Having described the invention in detail, it will be apparent that modifications and variations are possible without departing from the scope of the invention defined in the appended claims.

Examples

[0066] The following non-limiting examples are provided to further illustrate the present invention. Example 1

[0067] In this example, a high Z_st ethylene inverse diffusion flame was used to synthesize carbon nanotubes. Catalyst particles were produced in situ by introducing ferrocene with the fuel stream. A DMA coupled to a sampling probe was used to monitor the size distribution of the flame while varying the fuel concentration. Samples were size- selected using the DMA and subsequently collected for validation of the online size distribution results using transmission electron microscopy.

[0068] An inverse diffusion flame of an enriched-oxidizer and diluted fuel was established on a coflow burner with concentric tubes of 8 and 51 mm diameter. Fuel, nitrogen, and ferrocene were introduced through the outer tube with flow rates given in Table 1. Ferrocene was chosen as the precursor primarily because of its favorable decomposition temperature.

Ferrocene was introduced at a mass flow rate of 0.5 mg/min by flowing nitrogen carrier gas through a bubbler filled with ferrocene powder. Oxygen and nitrogen were introduced through the inner tube. Hydrogen and air, when present, were introduced with the fuel stream at flow rates of 0.1 mg/s and 3.0 mg/s, respectively. For all experiments the adiabatic flame temperature was maintained at 1647°C and the flame height was maintained at 28 mm ± 1 mm. Z_st was varied from 0.69 to 0.85. Nitrogen was added to the oxidizer stream as the fuel concentration was increased to maintain the same adiabatic flame temperature. All flows were controlled using calibrated rotameters. Based on these calibrations, the concentrations presented are accurate to within ±3%.

Table 1: Run Conditions for T_ad = 1647°C and Flame Height of 2\ mm

Zst [ C₂H₄ ] [ O₂ ] m inner ( mg / s ) ήl outer (mg/ s )

+ 3 % ±3%

Flame A 0 . 85 0 . 051 1 4 . 8 145 . 3

Flame B 0 . 74 0 . 059 0 . 75 6 . 2 14 6 . 5

Flame C 0 . 69 0 . 0 63 0 . 45 7 . 3 147 . 2

[0069] The flame was surrounded by a large plexiglass chimney to minimize disturbance. The aerosol was sampled with a quartz probe mounted vertically above the flame, and the probe tip was 3 cm from the flame to minimize its influence. The probe was connected to a Scanning Mobility Particle Sizer

(SMPS) consisting of an impactor, a Kr ⁵ bipolar ion source, a TSI 3085 nano-DMA, a TSI 3025A condensation particle counter

(CPC), and a PC to run the software. SMPS is useful for the present invention because it provides information in near real-time. Thus, the effect of process variables can be determined rapidly. SMPS can also be used to measure the size distribution at specific locations in the axial direction above the flame. The change in the size distribution with axial location allows for SWNT lengths to be estimated as they form and thus, a profile of the growth rate can be determined in near real-time. The SMPS system was used to monitor the size distribution of the aerosol as Z_st was varied. The peak diameters and number concentrations in the size distributions presented were reproducible to within ±5%. Samples were collected directly on TEM grids by selecting a DMA voltage and collecting the resultant particle stream at the exit of the nano-DMA using a high-efficiency electrostatic sampler. The samples were analyzed using a high-resolution transmission electron microscope (HR-TEM) .

[0070] As stated above, Z_st was varied between 0.69 and 0.85 while maintaining a maximum adiabatic flame temperature of 1647°C. Under these high Z_st conditions, the flames were free of soot (pure blue with no visible yellow emission) . The flames were far from their sooting limits. For example, for Z_st = 0.78 yellow emission was not observed until the maximum adiabatic flame temperature was 2027⁰C, or 400⁰C greater than the adiabatic flame temperatures used in this study. A flame near the sooting limit will have a tendency to form more PAH, which would be detrimental to nanotube growth.

[0071] The first operating condition was obtained by burning in pure oxygen and decreasing the fuel concentration until the flame was near the blow-off limit. This flame, flame A in Table 1 and FIG. IA, had a Z_st of 0.85 and an adiabatic flame temperature (T_ad) of 1647°C. The size distribution measured by the DMA at these settings is shown in Fig. 8. This size distribution contains only one peak located at an equivalent mobility diameter of about 14 nm. The lack of a second peak in this distribution indicates that no significant carbon nanotubes are being produced. The size distribution was also measured at these settings without the addition of ferrocene and no particles were detected, suggesting that the particles of FIG. 8 are either iron or iron oxide.

[0072] The fuel concentration was increased to 5.9% and the oxygen concentration was reduced to 75% producing flame B in Table 1 and FIG. IB, with a Z_st of 0.74. The corresponding size distribution is shown in Fig. 9A. This distribution is bimodal with one peak at an equivalent mobility diameter of 8 nm and the other at 48 nm. The presence of the second peak and the large separation between the peaks provides a good indication that nanotubes or nanofibers were produced. Each peak was size-selected using the DMA and a sample was collected for TEM analysis. The equivalent mobility diameter on the x-axis refers to the diameter of a spherical particle with the same electrical mobility as a nanotube or, for spherical particles, simply the diameter. Observation of TEM samples collected from the two different modes revealed that the mode on the left in FIG. 9Aa corresponds to catalyst particles encapsulated in a carbonaceous shell, while the mode on the right corresponds to single-walled carbon nanotubes. The SWNTs observed from the second mode in FIG. 9A were long and reasonably straight with diameters of 1-3 nm and lengths ranging from 200 nm to over one micron. The particular SWNT shown in the high resolution electron microscopy ("HREM") image of FIG 9B had a diameter of 1 nm and a length of 800 nm, yielding an aspect ratio of nearly 1000. The impurity present on the surface of this nanotube is typical of that observed on the SWNTs. It is believed that this impurity is primarily due to excess catalyst particles colliding with the nanotube. In addition, when compared to SWNTs synthesized in diluted fuel/air flames, high Z_st flame synthesized SWNTs contain significantly less soot impurities. This is a strong indication that high Z_st combustion is a viable approach for avoiding soot as an impurity on diffusion flame synthesized carbon nanotubes.

[0073] For the sample collected at a DMA setting of 8 nm, only catalyst particles encapsulated by disordered carbon were observed in the HREM image depicted in FIG. 10. A distribution of catalyst particle sizes is present, ranging from 2 nm to greater than 10 nm.

[0074] FIG. HA is a HREM image of some typical single- walled carbon nanotubes observed from a sample collected at a DMA setting of 48 nm. Only SWNTs were observed and these nanotubes were generally straight with an average diameter of 2 nm and an average length of about 1 μm based on a sample size of 20 SWNTs. A coating of amorphous carbon and catalyst particles is observed on some parts of the tubes while other sections are clean. A higher magnification shown in FIG. HB is a HREM image illustrating the lighter core of the tube relative to the walls. The walls of this tube were initially straight but were deformed by the electron beam of the TEM at high magnification.

[0075] Flame C in Table 1 and FIG. 1C, was produced with a Z_st of 0.69 by increasing the fuel concentration to 6.3% and decreasing the oxygen concentration to 45%. The size distribution was measured with and without the addition of ferrocene to ensure that the particles were not soot and no particles were detected without the addition of ferrocene. The size distribution resulting from the addition of ferrocene is shown in FIG. 12. This distribution is still bimodal indicating some nanotube formation. The second peak is at the same location as that in FIG. 9, although the number density has been reduced by nearly two orders of magnitude. The first peak had shifted to about 6.5 nm, which is less than its location in either FIG. 8 or FIG. 9. In general, as fuel concentration increased, the second peak corresponding to nanotube production appeared and then disappeared while the first peak shifted steadily to the left, corresponding to a smaller equivalent mobility diameter.

[0076] Size distributions were also determined for all flames using a long DMA to extend the measurement range out to an equivalent mobility diameter of 250 nm. No peaks were observed beyond 48 nm for any Z_st and it is not likely that any would appear beyond 250 nm since this would correspond to SWNTs significantly longer than a few microns.

[0077] A comparison of FIGS. 8, 9 and 12 demonstrates that SWNT formation generally occurred in a window of Z_st. Fuel concentration was varied only by 1.2% from flame A to flame C yet the change in the size distribution was significant. For a given temperature, an increase in fuel concentration might suggest that more carbon would be deposited on the catalyst particles, which would lead to a shift to a larger size. On the contrary, these figures show that the first peak of the distributions, which corresponds to catalyst particles encapsulated in disordered carbon, shifted to a smaller size for an increase in fuel concentration. In addition, the change in fuel concentration was small and thus, it is not anticipated that this fuel concentration is the cause of the peak shift or the window of SWNT formation.

[0078] One possible reason for the shift in the size distribution of the first peak, without being bound to any particular theory, is that when Z_st is varied, there is a change in oxygen concentration in the high temperature zone and a shift in the location of the radical pool. As Z_st is increased, the oxygen concentration in the high temperature zone increases (see D. X. Du, R. L. Axelbaum, C. K. Law, Combust Flame 102 (1-2) (1995) 11-20), and the radical pool shifts towards the fuel side (see R. Chen, R. L. Axelbaum, Combustion and Flame 142 (1-2) (2005) 62-71) . This combination of increasing oxygen concentration and a shift in the radical pool towards the fuel side with increasing Z_st could influence the extent of oxidation of the iron catalyst particles, with greater oxidation occurring for higher Z_st. A transition of the catalyst particle from Fe to FeO to Fe2θ3 would cause the size of the catalyst particle to increase. This increase would result in a shift in the first peak to the right as Z_st is increased, which corresponds to what is observed. Assuming that the first peak for flame C (FIG. 12) corresponds to iron with a size of 6.5 nm, a transition to FeO would cause a size increase to 7.8 nm, which is close to 8 nm diameter observed for flame B (FIG. 9) . If the particles were further oxidized to Fe2θ3, these particles would grow to 13 nm in diameter. This is reasonably close to the diameter of the first peak for flame A. Thus, oxidation of the particles may contribute to the trend of increasing particle size with increasing Z_st.

[0079] The change in oxygen concentration and the shift in the radical pool may also provide some explanation for the window of SWNT formation with Z_st because if the oxidation state and the size of the catalyst particle change this could change the activity of the catalyst particle towards SWNT growth. The prior art indicates that metal-oxides are the active catalyst for nanotube growth under flame conditions

(see M. J. Height, J. B. Howard, J. W. Tester, J. B. V. Sande, Carbon 42 (11) (2004) 295-2307; L. M. Yuan, T. X. Li, K. Saito, Carbon 41 (10) (2003) 1889-1896; and Y. Li, W. Kim, Y. Zhang, M. Rolandi, D. Wang, H. Dai, J Phys Chem B 105 (46)

(2001) 11424-11431) . It has also been shown that formation of SWNTs requires catalysts of ~2 nm in size (see H. Dai, Rinzlev, A. G., Nikolaev, P., Thess, A., Colbert, D., Smalley, R. E., Chem Phys Lett 260 (1996) 471-475) . FIGS. 9 and 12 show that the size distributions of flames B and C both contain numerous particles of the appropriate size for SWNT formation

(about 2 nm) . Nonetheless, flame B yields an order of magnitude more nanotubes that flame C. If, as suggested above, the catalyst particles of flame B consist of FeO, then a possible reason flame C produces less nanotubes is because Fe is less active than FeO.

[0080] The diameter of the first peak of flame A (FIG. 8) may correspond to further oxidation of the catalyst particles to Fe2θ3. Although this is a metal oxide, no nanotubes were observed for this flame. This can be understood by comparing FIG. 8 with FIGS. 9 and 12. The particle concentration for sizes less that 4 nm in FIG. 8 is observed to be two orders of magnitudes less than those of flames B and C in FIGS. 9 and 12. Thus, in this case it may be particle size, not particle composition, that is limiting nanotube growth. [0081] The SWNTs produced in flame B had an average length of about 1 μm, which is the longest size known for gas phase diffusion flame synthesis of SWNTs. These nanotubes, shown in FIG. 11, clearly demonstrate the ability of a high Z_st inverse diffusion flame to produce SWNTs at significant lengths. In addition, these nanotubes are relatively free of amorphous carbon impurities due to the use of high Z_st combustion. Although amorphous carbon is present at some locations on the tubes, there is far less than that found in prior art diluted-fuel/air diffusion flames (see R. L. Vander WaI, T. M. Ticich, V. E. Curtis, Chem Phys Lett 323 (3-4) (2000) 217-223) . Using a different precursor, for example iron pentacarbonyl, may further reduce these impurities (see R. L. Vander WaI, T. M. Ticich, V. E. Curtis, Chem Phys Lett 323 (3-4) (2000) 217-223) . A number of catalyst particles are also present as impurities on the synthesized SWNTs. These impurities can be decreased by introducing the ferrocene through a narrow region close to the flame instead of over the entire region of the outer flow. Some of the catalyst particles formed in the current configuration never pass through a region conducive to SWNT growth and thus, they only contribute to impurities.

[0082] The discovery of the window for SWNT formation clearly demonstrates the capability of the DMA to provide useful online information about the nanotube synthesis process. By measuring the size distribution with the DMA at each value of Z_st, the window of conditions suitable for SWNT formation can be quickly identified as indicated by the appearance of the bimodal distribution in FIG. 9. In addition, the optimal condition for SWNT growth at this temperature, residence time, and ferrocene concentration was identified to be at a Z_st of 0.79 and a T_ad of 1647°C.

[0083] In addition to information about the formation of nanotubes, FIG. 9 can also provide information about nanotube size and purity. A comparison of this concentration to that of the first peak gives an estimate of the amount of catalyst impurities. A larger difference between the two peaks corresponds to a larger amount of catalyst impurities. Finally, an estimate of the actual size of the nanotubes can be determined. The nanotubes were confirmed to be SWNTs by TEM analysis and the diameter was estimated to be about 2 nm.

Using this diameter and the methodology outlined by Unrau above, and assuming a charge of unity, an average length was estimated. This length was 0.8 //m, which is close to the average of 1 //m estimated from TEM analysis.

Example 2

[0084] To experimentally investigate whether oxidation of the catalyst particles is the cause of the change in nanotube production, hydrogen was added to flame A described in Example 1. Hydrogen is known to reduce surface oxide and thus, addition of the appropriate amount of hydrogen to flame A may reduce the catalyst particles to the composition present in flame B of Example 1. If oxidation of the catalyst particles is the determinant for nanotube production between Example 1 flames A, B and C, then flame A should produce nanotubes if enough hydrogen is added to reduce the catalyst particles to that of flame B. With the addition of 0.1 mg/s of hydrogen to flame A, the size distribution, which is shown in FIG. 6A, is nearly identical to that of flame B (FIG. 4B) . This provides an indication that a change in catalyst particle composition is the reason for the difference in nanotube production for flames A, B, and C.

[0085] Nonetheless, the size of the first peak for flame A was reduced to 8 nm by the addition of hydrogen and it might be that the large size of the catalyst peak is the reason for the lack of SWNT growth. To determine if catalyst particles in flame A were simply too large to produce SWNTs, the diameter of the first peak in flame A was also reduced to 8 nm by decreasing the ferrocene concentration. As seen in FIG. 6B, a second mode was not observed, further indicating that the oxidation of the catalyst particles plays an important role in nanotube formation in these flames. Finally, air (3.0 mg/s) was added to the fuel stream of flame C to shift the catalyst particle composition to that of flame B. Once again, the size distribution, shown in FIG. 6C, is nearly identical to that of flame B (FIG. 4B) . These results indicate that the optimal catalyst particle composition is iron (II) oxide and that high Z_st combustion may not only provide an essentially soot-free environment, but by adjusting Z_st within this high range, catalyst particle composition can be optimized for single-walled nanotube production.

Example 3

[0086] To characterize the growth rate of SWNTs in high Z_st flames, an axial size distribution profile of the post- flame region of flame B was obtained using a small quartz probe with an inlet diameter of 0.5 mm. The axial profile was obtained at a radial distance of 2 mm from the flame centerline. This radial distance corresponds to the maximum of SWNT formation at each axial height considered, thus ensuring the axial profile was along a streamline. The flame tip was located at 30.0 mm above the burner and size distributions were measured from 31.6 mm above the burner up to 43.6 mm in 2 mm increments. FIG. 13 shows the change in the SWNT mode of the size distribution with height above burner (HAB) . The first mode has been subtracted out for clarity. As seen in FIG. 13, the SWNT peak continues to shift to a larger equivalent mobility diameter as the axial location of the probe is increased. The location of the peak continues to shift until the probe reaches a height of 41.6 mm HAB. A further increase in probe height did not cause any further shift in the SWNT mode. [0087] The shift in the SWNT mode with axial location indicates growth of the single-walled nanotubes. Thus, these results demonstrate that the SMPS can be utilized to observe SWNT growth in near real-time. The peak of each mode shown in FIG. 13 can be used to estimate length and the growth rate can be obtained from velocity data. The equivalent mobility diameter of each peak can be converted to an average SWNT length using the following equation (3)

where χ₀ is the dynamic shape factor, d_v is the equivalent volumetric diameter equal to (1.5D²L)¹ where D and L are the diameter and length of the nanotube, n is the number of charges and C(d_a) is the slip correction factor based on the adjusted sphere diameter (see CJ. Unrau, R. L. Axelbaum, P. Biswas, P. Fraundorf, Online size characterization of nanofibers and nanotubes. In Molecular Building Blocks for Nanotechnology: From Diamondoids to Nanoscale Materials and Applications , G. A. Mansoori, T. F. George, L. Assoufid, G. Zhang, Eds.; Springer: New York, 2007; Vol. 109; pp 212) . The left-hand side of equation (3) is known from the SMPS settings. The right-hand side is a function of the diameter and length of the nanotube and the number of charges it carries. Due to the small cross-section of a single-walled carbon nanotube, it is reasonable to assume that the overwhelming majority of SWNTs passing through the SMPS system carry only one charge. For multiply charged nanotubes, the method developed by Unrau (above) may be used. In addition, the diameter distribution of SWNTs was found to be narrow from TEM analysis, ranging from 1-3 nm, and so a diameter of 2 nm was used to determine the average length with reasonable accuracy (see CJ. Unrau, R. L. Axelbaum, P. Biswas, P. Fraundorf Combust Inst 2007 , 31, 1865) . [0088] The above assumptions were used in equation (3) to determine the average length of the SWNTs corresponding to the peak of each mode in FIG. 13. Nanotube length versus axial location is shown in FIG. 14. To validate the assumptions made using the SMPS system, the average length was also measured by TEM imaging. For the TEM measurement, samples were collected via thermophoretic sampling at each height and the average length was measured based on a sample size of 50 nanotubes. As shown in FIG. 14, there is good agreement between the SMPS length measurement and the TEM measurement even though two different sampling techniques were used.

[0089] FIG. 14 shows that in the first 7 mm of the post- flame region, the SWNTs nucleate and grow rather slowly. Then, over the next 4 mm, the growth rate is rapid. Finally, over the last 2 mm, SWNT growth declines rapidly until it ceases all together. A temperature profile of the post-flame region was also measured and is shown in FIG. 14. One interpretation of these three growth regions can be made by considering the gradients in fuel concentration and temperature in the post-flame region. Near the flame, the fuel concentration is relatively low and thus, there is not enough carbon present to sustain high growth rates even though the temperature is sufficiently high. Since this flame configuration is inverted such that fuel is supplied from the outer flow, the fuel concentration increases downstream of the flame tip where it is nearly zero. Thus, as the SWNTs move away from the flame, they encounter an increase in fuel concentration. In the region where SWNT growth rate is a maximum, the temperature is decreasing but is still relatively high and the fuel concentration is high. Finally, as the SWNTs continue to move further from the flame, the temperature decreases and the catalyst particles on which the nanotubes have nucleated become poisoned with carbon deposition until growth ceases. This deactivation was observed on samples collected throughout the post-flame region, which contained catalyst particles encapsulated in amorphous carbon.

[0090] The length measurements shown in FIG. 14 can be combined with flame velocity measurements to estimate the growth rate. The velocity profile of the post-flame region was obtained using laser Doppler velocimetry. In the SWNT growth region, the velocity reached a maximum value of nearly 1.5 m/s. Combining this profile with the SWNT length profile, the growth rate as a function of axial location was determined and is shown in FIG. 15.

[0091] Based on these measurements, a maximum growth rate for SWNTs produced in this flame of over 100 //m/s was estimated. This growth rate is consistent with an average growth rate of 125 //m/s calculated from an experimentally determined growth time of 8 ms and a nanotube length of 1 //m.

The prior art has reported a similar growth rate in low- pressure premixed flames (see M.J. Height, J. B. Howard, J. W. Tester, J. B. V. Sande, Carbon 2004, 42, 2295), but, in contrast, the present invention the growth rate was attained at atmospheric pressure.

Example 4

[0092] The process involved an apparatus of the invention comprising a center tube for oxygen feed, an intermediate tube concentric and annular to the center tube for silicon and iron catalyst precursor, hydrocarbon fuel and hydrogen feed, and an outer tube concentric and annular to the intermediate tube for inert gas feed. The oxygen feed rate was about 25 mg per second ("mg/s"), but can be acceptably varied over a wide range. The mol ratio of hydrogen to ethylene fuel was about 140, the flow rate of hydrogen was about 12.3 mg/s, and the flow rate of ethylene was about 1.23 mg/s. A silicon/iron catalyst was generated from TMDS and ferrocene. A TMDS catalyst precursor source was connected by a tube to the hydrogen/ethylene gas stream and TDMS combined with the gas stream by diffusion so that a low concentration was achieved.

The TMDS feed rate was estimated to be about 0.1 mg/min. TMDS feed rate can be controlled by altering the diffusion surface area or by adding the gas stream subsurface to TMDS. Ferrocene catalyst precursor was contained in a vertical column through which the hydrogen/ethylene gas stream was passed resulting in a ferrocene feed rate estimated to be about 0.5 mg/min. Ferrocene feed rate can be controlled by heating the column. The inert gas (N₂) was fed through the outer tube so that it surrounded the flame. The N₂ feed rate can be suitably varied and was estimated to be about 250 mg/s.

Maximum flame temperature was estimated to be about 2700⁰C and Z_st was estimated to be about 0.02.

[0093] FIG. 16 is a size distribution for SWNTs synthesized from the flame apparatus of the present invention using the Fe/Si/O catalyst and FIG. 17 is a HREM image of those SWNTs. When compared to the size distribution of FIG. 4B (a process wherein oxygen was fed to the flame in an inner tube and hydrocarbon fuel and a ferrocene catalyst precursor to the flame in an outer tube), it is different. The flame design used for FIG. 4B and catalyst produced many catalyst particles without nanotubes (large peak at 8 nm) and about an order of magnitude less nanotubes (peak at 48 nm) as depicted in FIG. 4B. With the flame design and catalyst of the present invention, the peak corresponding to nanotubes (at 70 nm in the graph above) is an order of magnitude larger than the catalyst peak (at about 12 nm) . Thus, the flame apparatus and catalyst of the present invention cause nearly all catalyst particles to produce SWNTs in contrast to the FIG. 4B process.

Using acetylene or ethylene as the hydrocarbon fuel produces similar results.

Example 5 [0094] SWNTs were prepared using an aluminum based catalyst containing no silicon according to the method of Example 4_except aluminum isopropoxide was introduced as a catalyst precursor compound at a flow rate of 0.5 mg/minute and TMDS was not introduced. A Fe/Al/O catalyst was thereby formed in the flame. SWNTs were produced that were similar to those produced by the Fe/Si/O catalyst.

Example 6

[0095] SWNTs were prepared according to the method detailed in Examples 1 and 4, except the hydrocarbon fuel was replaced with ethanol. SWNTs produced from the flame are depicted in FIG. 18 and were similar in length, diameter, purity and growth to those produced in Examples 1 and 4 and depicted in FIGS. 9b and 17.

[0096] When introducing elements of the present invention or the preferred embodiments (s) thereof, the articles "a", "an", "the" and "said" are intended to mean that there are one or more of the elements. The terms "comprising", "including" and "having" are intended to be inclusive and mean that there may be additional elements other than the listed elements.

[0097] In view of the above, it will be seen that the several objects of the invention are achieved and other advantageous results attained.

[0098] As various changes could be made in the above compositions, processes and apparatuses without departing from the scope of the invention, it is intended that all matter contained in the above description and shown in the accompanying figures shall be interpreted as illustrative and not in a limiting sense.

Claims

WHAT IS CLAIMED IS:

1. A catalyst nanoparticle for the preparation of carbon nanotubes, the nanoparticle comprising from about 0.5 to about 95 mol percent silicon, aluminum or a mixture thereof, from about 4 to about 99 mol percent iron, and from about 0.1 to about 70 mol percent oxygen.

2. The catalyst nanoparticle of claim 1 comprising from about 1 to about 95 mol percent silicon, from about 5 to about 99 mol percent iron, and from about 0.1 to about 70 mol percent oxygen.

3. The catalyst nanoparticle of claim 1 consisting essentially of from about 0.5 to about 95 mol percent silicon, aluminum or a mixture thereof, from about 4 to about 99 mol percent iron, and from about 0.01 to about 70 mol percent oxygen.

4. The catalyst nanoparticle of claim 1 consisting essentially of from about 0.5 to about 95 mol percent silicon, from about 4 to about 99 mol percent iron, and from about 0.01 to about 70 mol percent oxygen.

5. The catalyst nanoparticle of claim 1 comprising from about 10 to about 80 mol percent silicon, aluminum or a mixture thereof, from about 4 to about 99 mol percent iron, and from about 0.1 to about 70 mol percent oxygen.

6. The catalyst nanoparticle of claim 5 comprising from about 10 to about 65 mol percent silicon and about 0.1 to about 15 mol percent aluminum.

7. The catalyst nanoparticle of claim 1 comprising from about 10 to about 15 mol percent silicon, aluminum or a mixture thereof, from about 25 to about 35 mol percent iron, and from about 50 to about 60 mol percent oxygen.

8. The catalyst nanoparticle of claim 1 having an average peak diameter of from about 1 nm to about 15 nm.

9. A catalyst nanoparticle for the preparation of carbon nanotubes, the catalyst nanoparticle comprising iron (II) oxide, the catalyst nanoparticle having an average peak diameter of from about 1 nm to about 15 nm.

10. A process for preparing carbon nanotubes, the process comprising combusting a reaction mixture in an inverse diffusion flame wherein the reaction mixture comprises at least one catalyst precursor, oxygen, and a fuel mixture comprised of carbon-based combustible fuel and hydrogen, wherein the flame has a temperature of at least 1000⁰C.

11. The process of claim 10 wherein the hydrogen concentration is greater than 1 mol percent.

12. The process of claim 10 wherein the molar ratio of hydrogen to carbon-based combustible fuel is from about 0.1 to about 200.

13. The process of claim 10 wherein the molar ratio of hydrogen to carbon-based combustible fuel is from about 3 to about 200.

14. The process of claim 10 wherein the catalyst precursor comprises iron.

15. The process of claim 14 wherein the reaction mixture further comprises a second catalyst precursor comprising silicon or aluminum.

16. The process of claim 14 wherein the reaction mixture further comprises a second catalyst precursor comprising silicon and a third catalyst precursor comprising aluminum.

17. The process of claim 10 wherein the flame temperature is from 1300⁰C to about 3000⁰C.

18. The process of claim 17 wherein the flame temperature is from 1800⁰C to about 3000⁰C.

19. The process of claim 10 wherein the flame has a stoichiometric mixture fraction (Z_st) of from about 0.1 to about 0.95.

20. The process of claim 10 wherein the carbon nanotube growth rate is at least 10 μm per second.

21. The process of claim 10 wherein the average diameter of the carbon nanotubes is from about 0.7 nm to about 5 nm.

22. The process of claim 10 wherein the average length of the carbon nanotubes is from about 200 nm to about 100 microns.

23. The process of claim 10 wherein the carbon nanotubes comprise single-walled carbon nanotubes.

24. The process of claim 10 wherein catalyst nanoparticles are formed and the carbon nanotubes form on at least about 10% by number of the catalyst nanoparticles.

25. The process of claim 10 wherein the inverse diffusion flame further comprises an inert gas.

26. The process of claim 10 wherein the inverse diffusion flame is in the presence of an inert atmosphere, and the combustion occurs in an apparatus comprising: a first, innermost, tube for the introduction of the oxygen to the inverse diffusion combustion flame; a second tube annular and concentric to the first tube thereby forming a first opening between the exterior surface of first tube and the interior surface of the second tube wherein the hydrogen and a first portion of the carbon-based combustible fuel are introduced to the inverse diffusion combustion flame through the first opening, the first opening being continuous from a hydrogen source and a carbon-based combustible fuel source to the flame; a third tube annular and concentric to the second tube thereby forming a second opening between the exterior surface of second tube and the interior surface of the third tube wherein the remaining portion of the carbon-based combustible fuel stream is introduced to the inverse diffusion combustion flame through the second opening, the second opening being continuous from a carbon-based combustible fuel stream source to the flame; and a fourth tube annular and concentric to the third tube thereby forming a third opening between the exterior surface of third tube and the interior surface of the fourth tube wherein an inert gas is introduced to the inverse diffusion combustion flame through the third opening, the third opening being continuous from an inert gas source to the flame, wherein the catalyst precursor is introduced to the inverse diffusion combustion flame through (i) the first tube, the first tube being continuous from a catalyst precursor source to the flame, (ii) the first opening, the first opening being continuous from a catalyst precursor source to the flame, (iii) the second opening, the second opening being continuous from a catalyst precursor source to the flame or (iv) a combination thereof.

27. The process of claim 10 wherein the inverse diffusion flame is in the presence of an inert atmosphere, and the combustion occurs in an apparatus comprising: a first, innermost, tube for the introduction of the oxygen stream to the inverse diffusion combustion flame; a second tube annular and concentric to the first tube thereby forming a first opening between the exterior surface of first tube and the interior surface of the second tube wherein the hydrogen stream and the carbon-based combustible fuel stream are introduced to the inverse diffusion combustion flame through the first opening, the first opening being continuous from a hydrogen source and a carbon-based combustible fuel stream source to the flame; and a third tube annular and concentric to the second tube thereby forming a second opening between the exterior surface of second tube and the interior surface of the third tube wherein an inert gas stream is introduced to the flame through the second opening, the second opening being continuous from an inert gas stream source to the flame, wherein the catalyst precursor is introduced to the inverse diffusion combustion flame through (i) the first tube, the first tube being continuous from a catalyst precursor source to the flame, (ii) the first opening, the first opening being continuous from a catalyst precursor source to the flame, (iii) the second opening, the second opening being continuous from a catalyst precursor source to the flame or (iv) a combination thereof.

28. The process of claim 26 or 27 wherein the apparatus is operatively coupled to nanotube analysis device.

29. The process of claim 28 wherein the nanotube analysis device is a differential mobility analyzer.

30. A process for preparing carbon nanotubes, the process comprising combusting a reaction mixture in an inverse diffusion flame wherein the reaction mixture comprises a carbon-based combustible fuel, at least one catalyst precursor, and oxygen and the flame has a temperature of at least 1800⁰C.

31. The process of claim 30 wherein the reaction mixture further comprises hydrogen.

32. The process of claim 31 wherein the molar ratio of hydrogen to carbon-based combustible fuel is from about 0.1 to about 200.

33. The process of claim 31 wherein the molar ratio of hydrogen to carbon-based combustible fuel is from about 3 to about 200.

34. The process of claim 30 wherein the catalyst precursor comprises iron.

35. The process of claim 34 wherein the reaction mixture further comprises a second catalyst precursor comprising silicon or aluminum.

36. The process of claim 34 wherein the reaction mixture further comprises a second catalyst precursor comprising silicon and a third catalyst precursor comprising aluminum.

37. The process of claim 30 wherein the flame temperature is from 1800⁰C to about 3000⁰C.

38. The process of claim 30 wherein the flame has a stoichiometric mixture fraction (Z_st) of from about 0.1 to about 0.95.

39. The process of claim 30 wherein the carbon nanotube growth rate is at least 10 μm per second.

40. The process of claim 30 wherein the average diameter of the carbon nanotubes is from about 0.7 nm to about 5 nm.

41. The process of claim 30 wherein the average length of the carbon nanotubes is from about 200 nm to about 100 microns.

42. The process of claim 30 wherein the carbon nanotubes comprise single-walled carbon nanotubes.

43. The process of claim 30 wherein catalyst nanoparticles are formed the carbon nanotubes form on at least about 10% by number of the catalyst nanoparticles.

44. The process of claim 30 wherein the inverse diffusion flame further comprises an inert gas.

45. The process of claim 30 wherein the inverse diffusion flame is in the presence of an inert atmosphere, and the combustion occurs in an apparatus comprising: a first, innermost, tube for the introduction of the oxygen into the inverse diffusion combustion flame; a second tube annular and concentric to the first tube thereby forming a first opening between the exterior surface of first tube and the interior surface of the second tube wherein the hydrogen and a first portion of the carbon-based combustible fuel are introduced to the inverse diffusion combustion flame through the first opening, the first opening being continuous from a hydrogen source and a carbon-based combustible fuel source to the flame; a third tube annular and concentric to the second tube thereby forming a second opening between the exterior surface of second tube and the interior surface of the third tube wherein the remaining portion of the carbon-based combustible fuel stream is introduced to the inverse diffusion combustion flame through the second opening, the second opening being continuous from a carbon-based combustible fuel stream source to the flame; and a fourth tube annular and concentric to the third tube thereby forming a third opening between the exterior surface of third tube and the interior surface of the fourth tube wherein the inert gas is introduced to the inverse diffusion combustion flame through the third opening, the third opening being continuous from an inert gas source to the flame, wherein the catalyst precursor is introduced to the inverse diffusion combustion flame through (i) the first tube, the first tube being continuous from a catalyst precursor source to the flame, (ii) the first opening, the first opening being continuous from a catalyst precursor source to the flame, (iii) the second opening, the second opening being continuous from a catalyst precursor source to the flame or (iv) a combination thereof.

46. The process of claim 30 wherein the inverse diffusion flame is in the presence of an inert atmosphere, and the combustion occurs in an apparatus comprising: a first, innermost, tube for the introduction of the oxygen stream into the inverse diffusion combustion flame; a second tube annular and concentric to the first tube thereby forming a first opening between the exterior surface of first tube and the interior surface of the second tube wherein the hydrogen stream and the carbon-based combustible fuel stream are introduced to the inverse diffusion combustion flame through the first opening, the first opening being continuous from a hydrogen source and a carbon-based combustible fuel stream source to the flame; and a third tube annular and concentric to the second tube thereby forming a second opening between the exterior surface of second tube and the interior surface of the third tube wherein an inert gas stream is introduced to the flame through the second opening, the second opening being continuous from an inert gas stream source to the flame, wherein the catalyst precursor is introduced to the inverse diffusion combustion flame through (i) the first tube, the first tube being continuous from a catalyst precursor source to the flame, (ii) the first opening, the first opening being continuous from a catalyst precursor source to the flame, (iii) the second opening, the second opening being continuous from a catalyst precursor source to the flame or (iv) a combination thereof.

47. The process of claim 45 or claim 46 wherein the apparatus is operatively coupled to nanotube analysis device.

48. The process of claim 47 wherein the nanotube analysis device is a differential mobility analyzer.

49. An inverse diffusion combustion apparatus comprising: a first, innermost, tube for the introduction of an oxidizer stream to an inverse diffusion combustion flame; a second tube annular and concentric to the first tube thereby forming a first opening between the exterior surface of first tube and the interior surface of the second tube for the introduction of a first fuel stream to the inverse diffusion combustion flame through the first opening, the first opening being continuous from a first fuel stream source to the flame; a third tube annular and concentric to the second tube thereby forming a second opening between the exterior surface of second tube and the interior surface of the third tube for the introduction of a second fuel stream to the flame through the second opening, the second opening being continuous from a second fuel stream source to the flame; and a fourth tube annular and concentric to the third tube thereby forming a third opening between the exterior surface of third tube and the interior surface of the fourth tube for the introduction of an inert gas stream to the flame through the third opening, the third opening being continuous from an inert gas stream source to the flame, wherein the first tube and/or the first opening is continuous from a catalyst precursor source to the flame.

50. The apparatus of claim 49 wherein the ratio of the first tube diameter to the second tube diameter is between about 0.4 and about 0.95.

51. The apparatus of claim 49 wherein the first opening is continuous from a hydrogen source to the flame.

52. The apparatus of claim 49 wherein the apparatus is operatively coupled to a nanotube analysis device.

53. The apparatus of claim 52 wherein the nanotube analysis device is a differential mobility analyzer.

54. An inverse diffusion combustion apparatus comprising: a first, innermost, tube for the introduction of an oxidizer stream to an inverse diffusion combustion flame; a second tube annular and concentric to the first tube thereby forming a first opening between the exterior surface of first tube and the interior surface of the second tube for the introduction of a fuel stream to the inverse diffusion combustion flame through the first opening, the first opening being continuous from a fuel stream source to the flame; and a third tube annular and concentric to the second tube thereby forming a second opening between the exterior surface of second tube and the interior surface of the third tube for the introduction of an inert gas stream or fuel stream to the flame through the second opening, the second opening being continuous from an inert gas stream source to the flame, wherein the first tube and/or the first opening is continuous from a catalyst precursor source to the flame.

55. The apparatus of claim 54 wherein the first opening is continuous from a hydrogen source to the flame.

56. The apparatus of claim 54 wherein the apparatus is operatively coupled to a nanotube analysis device.

57. The apparatus of claim 56 wherein the nanotube analysis device is a differential mobility analyzer.