Method for depositing high aspect ratio molecular structures

Abstract

A method for depositing high aspect ratio molecular structures (HARMS), which method comprises applying a force upon an aerosol comprising one or more HARM-structures, which force moves one or more HARM-structures based on one or more physical features and/or properties towards one or more predetermined locations for depositing one or more HARM-structures in a pattern by means of an applied force.

Claims

1. A method for depositing high aspect ratio molecular structures (HARM-structures), the method comprising: providing an aerosol comprising individual and bundled HARM-structures; applying a force upon the individual and bundled HARM-structures; moving at least part of the individual and bundled HARM-structures based on one or more physical features and/or properties towards one or more predetermined locations by means of the applied force; and depositing at least part of the individual and bundled HARM-structures in a pattern by means of the applied force, wherein the force is a thermophoretic force, and one or more individual and bundled HARM-structures are deposited in the pattern by patterned heating and cooling of a collection plate.

2. A method according to claim 1, wherein the individual and bundled HARM-structures comprise at least one selected from a carbon nanotube, a fullerene functionalized carbon nanotube, a boron-nitride nanotube, a nanorod including carbon, a phosphorous-, boron-, nitrogen- or silicon-containing nanorod, a filament, a tube, a rod and ribbon.

3. A method according to claim 2, wherein the method further comprises adding one or more reactants, agents, coating materials, functionalizing materials, surfactants and/or dopants to the aerosol comprising individual and bundled HARM-structures.

4. A method according to claim 1, wherein the method further comprises adding one or more reactants, agents, coating materials, functionalizing materials, surfactants and/or dopants to the aerosol comprising individual and bundled HARM-structures.

Description

LIST OF FIGURES

(1) In the following section, the invention will be described in detail by means of embodiment examples with reference to accompanying drawings, in which

(2) FIG. 1 illustrates one embodiment of the present method for moving HARM-structures and thereby separating bundled from individual HARM-structures.

(3) FIG. 2a illustrates one embodiment of the present method for continuously separating and selectively depositing bundled HARM-structures.

(4) FIG. 2b illustrates one embodiment of the present method for batch separation and selective deposition of bundled HARM-structures.

(5) FIGS. 3a, 3b and 3c illustrate other embodiments of the present method for separating and selectively depositing bundled from individual BARM-structures.

(6) FIG. 4 illustrates one embodiment of the present method for depositing individual EARM-structures.

(7) FIG. 5 illustrates one embodiment of the present invention for patterned deposition of mixed bundled and individual, only bundled and/or only individual HARM-structures.

(8) FIG. 6 illustrates one embodiment of the present method for deposition of mixed bundled and individual, only bundled and/or only individual HARM-structures by thermophoresis.

(9) FIGS. 7a and 7b illustrate other embodiments of the present invention for patterned deposition of mixed bundled and individual, only bundled and/or only individual HARM-structures by thermophoresis.

(10) FIG. 8 illustrates a continuous process for the synthesis and further processing of HARM-structures.

(11) FIG. 9a, illustrates mobility size distribution (measured using DMA without .sup.241Am bipolar charger) of the positively and negatively naturally charged fraction of bundled CNTs in the gas phase; FIG. 9b illustrates TEM image of the as-grown bundled CNTs.

(12) FIG. 10a illustrates comparison between the mobility size distributions of all and neutral CNTs and catalyst particles at different CO concentrations and FIG. 10b illustrates charged fraction of particles (N.sup.+/−) and particle number concentration plotted versus the CO concentration.

(13) FIG. 11a illustrates comparison between the mobility size distributions of all and neutral CNTs and catalyst particles at different hot wire heating powers and FIG. 11b illustrates charged fraction of aerosol particles (N.sup.+/−) and particle number concentration plotted versus the hot wire heating power.

(14) FIG. 12a illustrates TEM images of individual CNTs and FIG. 12b illustrates TEM images of both individual and bundles CNTs.

(15) FIG. 13a illustrates AFM images of individual CNTs deposited onto a temperature sensitive polymer-based substrate (SU-8; 10 jam thick layer) and

(16) FIG. 13b illustrates AFM images of individual CNTs deposited onto a Si.sub.3N.sub.4 substrate (119 μm thick layer).

(17) FIG. 14 illustrates AFM image of individual CNTs deposited on a Si.sub.3N.sub.4 membrane grid.

(18) FIG. 15 illustrates AFM image of individual CNTs deposited on a SiO.sub.2 substrate.

(19) FIGS. 16 (a and b) illustrates a SEM image of bundled CNTs deposited on a silica substrate at two different locations where an electrostatic charge was applied.

DETAILED DESCRIPTION OF THE INVENTION

(20) FIG. 1 illustrates one embodiment of the present method for moving and thereby separating bundled from individual HARM-structures. A mixture of bundled and individual HARM-structures (1) is collectively subjected to a force (2) which selectively acts on either the bundled or individual HARM-structures, based on at least one of their physical features, for example properties, distinguishing them, such that they are moved, for example accelerated, relative to one another so as to become separated in space into a plurality of individual (3) and bundled (4) HARM-structures.

(21) FIG. 2a illustrates one embodiment of the present invention, wherein a mixture of bundled and individual HARM-structures (1) is suspended in, for example, a carrier gas, a liquid or is suspended in a vacuum. Said dispersion is caused to pass through an electric field due to a voltage differential (6). Since bundled HARM-structuresare naturally substantially charged and the individual HARM-structures are substantially uncharged, the electric field causes the bundled HARM-structures (4) to migrate in the electric field (the direction dependant upon the sign of their net charge) and be separated from the individual HARM-structures (3), which pass through the electric field largely unaffected. In other words, the naturally charged HARM-structures are accelerated toward the channel walls (9) while the individual HARM-structures exit the channel.

(22) Similarly, FIG. 2b illustrates the method performed in batch mode, wherein the dispersion of bundled and individual HARM-structures (1) is put in a chamber (13), wherein an electrical potential or a voltage (6) is applied to cause the separation and deposition of bundled (4) and individual (3) HARM-structures. Naturally charged bundled HARM-structures are accelerated toward the walls (9) while the individual HARM-structures remain suspended.

(23) FIG. 3 illustrates another embodiment of the present method, wherein the mixture of bundled and unbundled HARM-structures (1) are suspended in a liquid or gas and subjected to a force, in this case an inertial or gravitational acceleration, thus causing the bundled (4) and individual (3) HARM-structures to separate. Here, bundled HARM-structures are selectively deposited from individual HARM-structures and a suspension of individual HARM-structures is generated via balancing inertia and drag. The dispersion is, in this embodiment, introduced into a curved channel (9) a) or directed towards a surface (4) b) and the bundled HARM-structures having a higher effective Stokes number are accelerated toward the surface while the individual HARM-structures remain suspended. Furthermore, the bundled HARM-structures can be made to deposit on a substrate. Alternatively c), the dispersion of individual and bundled HARM-structures (1) is fed into an expanding channel (7) in the opposite direction to the acceleration of gravity (8). The velocity of the suspending liquid or gas is decreased in the expanding volume. Thereby, the individual HARM-structures (3) having a lower Stokes number than the bundled HARM-structures (4) are carried further up the channel than the bundled HARM-structures, thereby causing them to separate.

(24) FIG. 4 illustrates that for example separated individual HARM-structures (3) can be deposited on a separate substrate (9) by electrostatic precipitation. Herein a voltage (11) is applied to a needle (12) to create an electron cloud, which charges the previously uncharged individual HARM-structures. In this device, HARM-structures are charged by field charging using a corona discharge that ionizes the gas and creates a small current between two plates. Thereafter, the use of a force, in this case an electrostatic migration velocity, will make them to deposit on the grounded collection plate (10) where the substrate (9) is located. Moreover, the deposition location can be determined by local variation of the electric field, thus allowing patterned deposition. Various means are available for the localization of the electric field.

(25) FIG. 5 illustrates one way of allowing patterned deposition by use of an electrostatic force. A pattern of charge can be localized on a semi or nonconductive substrate (9) by, for example, making a stamp or mask of conductive material (14) and applying the mask to the substrate so as to contact charge those areas of the substrate in contact with the stamp or mask (15). After the stamp or mask is removed, the contacted areas remain charged and a dispersion of a mixture of bundled and individual (1), bundled (4) or individual (3) HARM-structures are then brought into the vicinity of the charged substrate whereupon the local electric field causes those with an opposite charge to accelerate toward the predetermined pattern and to deposit according to the pattern of the stamp or mask. The resolution of the deposition pattern is thus approximately equal to that of the stamp or mask.

(26) FIG. 6 illustrates one embodiment of the present invention, wherein a thermoforetic force is used to deposit HARM-structures (1,3,4) on a substrate. A notable advantage of depositing particles using the thermophoretic precipitator is the possibility of employing any type of substrates. Here, an aerosol of carrier gas and HARM-structures is made to pass between a gap between a heated plate (16) and a cooled plate (17). Various means known in the art can be used to heat and cool the plates, but in the preferred embodiment, the hot plate is heated via an electric current and the cold plate is cooled conductively via a flow of cold water. The HARM-structures then migrate from the hot plate to the cold and are deposited on the attached substrate (9).

(27) FIG. 7a illustrates a way of depositing HARM-structures, wherein the HARM-structures are deposited in a pattern by patterned heating and cooling of the collection plate. Here a plurality of heating (18) and cooling (19) elements are positioned on one side of the substrate (9) and the aerosol of mixed, bundled and/or individual HARM-structures (1,3,4) are introduced on the other side of the substrate and between a heating plate (16). The HARM-structures are then deposited by the thermophoretic force, i.e. thermophoresis, onto the relatively cooler portions of the substrate. Other means can be used to create the heating and cooling patterns on the substrate. For example, for low heat conducting substrates, radiation (for example laser irradiation) can be used. For example, a pattern of laser beams can be directed toward the cooled substrate to heat particular regions. The method according to the present invention can be used to deposit HARM-structures having different properties to be deposited in different positions as is illustrated in FIG. 7b. Here, at time interval 1 (t1), the heating and cooling of the substrate is in a given pattern whereupon the HARM-structures of type 1 are deposited. Subsequently, the heating and cooling of the substrate is changed and HARM-structures of type 2 are deposited. The process can be repeated to create complex, multi-property deposition patterns.

(28) FIG. 8 illustrates the incorporation of the method according to the present invention into a HARM-production process. In FIG. 8 the method according to the present invention is incorporated into a floating catalyst HARM production process known in the art. Here, catalyst particles or catalyst particle precursors (20) are introduced into a HARM reactor (21) together with appropriate source(s) (22) and additional reagents (23) as required. An aerosol of HARM-structures exit the reactor and are separated in a separation apparatus (24) which operates according to any of the methods described to separate bundles (4) and individual (3) HARM-structures. Thereafter, the individual HARM-structures can be charged, coated, functionalized or otherwise modified in a conditioning reactor (25) and then deposited on a substrate (9) in a deposition reactor (27) according to any of the methods described to move bundled and/or individual HARM-structures. The deposition layer can be homogeneous layered or patterned according to the methods described in the invention. Furthermore, the deposition layer can be further processed in any suitable way. Further, unused precursors and/or reagents can be recovered in a recovery reactor (30) by means known in the art and fed back into the production cycle. The process can be repeated.

EXAMPLES

(29) In the following examples, bundled and individual HARM-structures, in this example carbon nanotubes (CNT), were moved and thereby separated from each other and separately deposited according to the described invention.

(30) In all examples, the CNTs were continuously synthesized upstream of the separation and deposition steps to produce an aerosol containing a mixture of bundled and individual CNTs. A hot wire generator (HWG) method was used for the synthesis of CNTs as is known in the art. In the method Fe catalyst particles were produced by vaporization from a resistively heated catalyst wire in a H.sub.2/Ar (with a 7/93 mol ratio) flow (400 cm.sup.3/min). Particles were formed and grown by vapor nucleation, condensation and particle coagulation processes. Subsequently, the produced particles were introduced into a ceramic tubular reactor at about 400° C., mixed with a carbon monoxide (CO) flow of 400 cm.sup.3/min and heated to induce CNT formation (from 700° C. to 900° C.). A porous tube dilutor (6 l/min) was installed downstream of the reactor to prevent the product deposition on the walls. 12 cm.sup.3/min of CO.sub.2 was introduced in the reactor as an etching agent. Unless otherwise stated, all the experiments were carried out using a heating power to the wire of 19 W, a CO concentration of 53% in a CO/(Ar—H.sub.2) (93-7 mol ratio) mixture, and a peak reactor temperature of 700° C. Mobility size distributions of aerosol particles (i.e. catalyst particles, individual CNTs and CNT bundles dispersed in the gas phase) were measured by a differential mobility analyzer system consisting of a classifier, a condensation particle counter, and an optional .sup.241Am bipolar charger. Adequate power supplies for applying both positive and negative polarity to the internal electrode were used, while the external electrode was kept grounded. An electrostatic filter (ESF) was located downstream of the reactor and used to filter out the charged aerosol particles (when required). The ESP is comprised of two metallic plates with dimensions of 15 cm, length, and 2 cm, height, separated each other by a distance of 1 cm. This device enabled the filtering out of charged aerosol particles by connecting one of the plates to high voltage (around 4000 V) while the other one was kept grounded. Aerosol particles including catalyst particles and CNT HARM-structures were collected on carbon coated copper grids for their structural characterization by TEM.

Example 1

Moving and Separation of Bundled and Individual CNT HARM-structures by Electrostatic Precipitation by Taking Advantage of the Naturally Charging of Bundled HARM-structures

(31) The mobility size distribution of the naturally charged aerosol particles (i.e. obtained without external bipolar charger prior the DMA) is illustrated in FIG. 9. The figure shows the dependence of the measured frequency on both the equivalent mobility diameter, D, calculated assuming a spherical shape and a single charge, and the inverse electrical mobility, 1/Z. As can be seen, a broad mobility distribution with a mean mobility diameter of around 45 nm was obtained regardless the polarity of the bias and was attributed to the presence of CNTs. TEM observation of the sample produced at 700° C. and directly collected onto the TEM grid from the gas phase showed that the nanotubes were single-walled and clearly aggregated in bundles (FIG. 9b). Since the DMA can classify only charged aerosol particles, these results indicate that the nanotubes coming from the reactor were naturally electrically charged. Furthermore, this phenomenon was observed independently of the polarity applied to the DMA. According to concentration measurements, the CNTs were approximately equally positively and negatively charged with fractions of N.sup.+=47% and N.sup.−=53%, respectively (table 1).

(32) Previous investigations on metal nanoparticle formation by a HWG indicated that the particles posses electrical charges after their formation. In order to study the possibility that Fe catalyst particles could also become charged in our system, and, consequently be the origin of the charging of the nanotubes, CO was replaced by N.sub.2 (thereby preventing the formation of CNTs). Our investigations carried out at temperatures from 25° C. to 900° C. showed that almost all the Fe particles (up to 99%) were electrically neutral (table 2), which suggests that catalyst particles are not directly involved in the observed charging of the nanotubes.

(33) To measure the mobility size distributions of the neutral aerosol particles, the charged aerosol was filtered out by applying a potential difference between electrodes in the ESF. The extracted neutral aerosol particles were artificially charged using the external bipolar charger (.sup.241Am) prior the mobility distribution measurement by DMA. A peak with a mean equivalent diameter of 5 nm was observed and assigned to Fe catalyst particles that remain inactive for the growth of CNTs. Thus, these results indicate that all the nanotubes were deposited in the ESF and, hence, were electrically charged. Similar results were obtained at 800° C. and 900° C. (table 1).

(34) TABLE-US-00001 TABLE 1 Charged fraction (N.sup.+/−) of CNTs, synthesized using 53% CO and a heating power of 19 W, at different reactor temperatures. (N.sup.+) and (N.sup.−) indicate the polarity distribution of charged CNTs. Temperature (° C.) N.sup.+/− (%) N.sup.+ (%) N.sup.− (%) 700 99 47 53 800 99 48 52 900 97 41 59

(35) TABLE-US-00002 TABLE 2 Charged fraction (N.sub.p.sup.+/−) of Fe catalyst particles produced via HWG method in N.sub.2 atmosphere, at different reactor temperatures. (N.sub.p.sup.+) and (N.sub.p.sup.−) indicate the polarity distribution of charged catalyst particles. Temperature (° C.) N.sub.p.sup.+/− (%) N.sub.p.sup.+ (%) N.sub.p.sup.− (%) 25 1 99 1 700 1 4 96 800 4 27 73 900 2 28 72

(36) It is known that gas surface reactions may induce electronic excitations at metal surfaces. When highly exothermic reactions take place, these excitations may lead to the ejection of ions and electrons from the surface. As a consequence, it can be speculated that the exothermic CO disproportionation reaction needed for the growth of CNTs might play a role in their electrical charging. In an attempt to study it, experiments were carried out varying the CO concentration. In order to quantitatively estimate the fraction of charged CNTs (N.sup.+/−), mobility size distributions were measured with the .sup.241Am bipolar charger prior the classifier. FIG. 10a shows the comparison between the mobility size distributions of all CNTs (ESF off) and the neutral CNTs (ESF on) at CO concentrations of 27%, 34% and 53%. As is expected, the CNT concentration increased with the concentration of carbon source (CO) introduced into the reactor. At the concentration of 27%, the mobility size distribution of all CNTs and the neutral fraction appeared to be identical indicating that almost all CNTs are electrically neutral. However, the fraction of neutral CNTs gradually decreased as the CO concentration was increased. Thus, at 53% of CO, almost all the CNTs were charged. FIG. 10b illustrates concisely the effect of the CO concentration on the total fraction of charged nanotubes (N.sup.+/−) and the concentration of the product.

(37) In a similar manner, mobility distributions were also measured varying the heating power applied to the wire from 16 W to 19 W when the CO concentration was kept constant at 53%. Increasing the power increases the concentration of CNTs, due to a higher concentration of Fe catalyst particles produced. Consequently, nanotube bundling increases. As can be seen in FIG. 11, the fraction of the charged CNTs increased with the power applied to the heated wire.

(38) The results show that a higher concentration of CNTs leads to more effective charging. This fact relates to the bundling of the CNTs, since the likelihood for bundling increases with their concentration in the gas phase. Accordingly, the natural charging of the CNTs may happen in the process of formation of bundles. This hypothesis was supported by TEM observation of the sample containing charged CNTs, where only bundled CNTs were found (FIG. 9b).

(39) In order to collect the neutral fraction of CNTs, the ESF was used to filter out the charged CNTs. CNTs were synthesized using a heating power of 16.5 W to maintain a low concentration of CNTs and, thereby to minimize their bundling. At these experimental conditions the fraction of charged CNTs was estimated to be around 12%. CNTs were collected directly from gas phase onto a TEM holey carbon film substrate using a point-to-plate electrostatic precipitator. TEM observations of the neutral CNTs revealed the presence of only individual CNTs (FIG. 12a). The collection of the whole product (i.e. without filtering charged CNTs out) revealed the presence of both bundles and individual CNTs (FIG. 12b). This indicates that individual CNTs were neutral whereas bundles were charged.

(40) The charging effect can be explained by the van der Waals energy released during the CNT bundling. In order to minimize the total free energy, CNTs form bundles consisting of individual tubes located parallel to each other. This results in a relatively high energy release: for example, the bundling of two armchair (10,10) CNTs leads to the total energy decrease by as much as 95 eV/100 nm. The bundle may be charged due to the emission of electrons and ions via dissipation of the released van der Waals energy. The high contact area to surface area ratio and high surface area to volume ratio of CNTs likely allows a significant charging that would not be detectable in large and/or low aspect ratio structures.

(41) As the charging process due to bundling is directly related to the high ratio of contact area to volume of these approximately one-dimensional structures, the findings are applicable to any High Aspect Ration Molecular Structures (HARM-structures) as mentioned above.

Example 2

Separation of Bundled and Individual CNTs in the Gas Phase and Separate Deposition Via Electrostatic Precipitation on Polymer-based Substrate and Si3N4 Substrates

(42) Bundled and individual CNTs were moved and thereby separated with the method according to the present invention. The separated CNTs were then separately deposited on a polymer-based substrate (SU-8, 10 μm thick layer), with a degradation temperature of ˜300° C., and a Si.sub.3N.sub.4 substrate (119 μm thick layer). The deposition was carried out using an electrostatic precipitator (FIG. 4). Atomic force microscopy (AFM) images illustrated in FIG. 13a-b show the presence of individual CNTs, which had been charged before the deposition, with diameters ranging from 0.7 to 1.1 nm determined from the height measurements, which is consistent with what is determined by TEM. In addition, AFM images of individual CNTs collected onto and Si.sub.3N.sub.4 substrates (100 nm thick) are shown in FIG. 14.

Example 3

Separation of Bundled and Individual CNTs in the Gas Phase and Separate Deposition Via Thermophoresis on a SiO2 Substrate

(43) Bundled and individual CNTs were moved and thereby separated with the method according to the present invention. The separated CNTs were then separately deposited on a polymer-based substrate (SU-8, 10 μm thick layer), with a degradation temperature of ˜300° C., and a SiO.sub.2 substrate. The deposition was carried out using a thermophoretic precipitator (FIG. 6). An atomic force microscopy (AFM) image illustrated in FIG. 15 show the presence of individual CNTs.

Example 4

Deposition of Bundles of CNTs from the Gas Phase on a Silica Substrate by Electrostatic Charging

(44) Bundles of CNTs were produced in a CNT reactor using ferrocene and carbon monoxide. The bundles were deposited from the gas phase onto a silica substrate. The substrate was prepared beforehand so as to have a local electrostatic charge by pressing the tip of an iron pin to the substrate surface. Bundles of CNTs were deposited only on the points where the pin pressure had been previously applied. Resulting deposits are shown in FIG. 16.

(45) The invention is not limited merely to the embodiment examples referred to above, instead many modifications are possible within the scope of the inventive idea defined in the claims.