Popov V.N., Lambin P. (eds.) Carbon Nanotubes

Подождите немного. Документ загружается.

extend far within the 25 mm tube. Nucleation and SWNT growth could now be

carried out inside this tube.

Kingston et al.

were able to enhance the yield considerably by using two

Nd:YAG lasers, one (1.5J) operating in the pulsed-mode to ablate and the other

one in cw-mode to alter the rate of cooling of the condensing plume. The

synthesis rate of black soot surpassed 400mg/h.

Our set-up consists also of two different laser systems. For the first time (as

much as we know) a Nd:YAG and a small 100W cw-CO

-laser are used

simultanously. While the Nd:YAG (1.0J, 1064nm, 20Hz, 7ns) is used to ablate

material from the target, the CO

-laser systems is meant to influence the cooling

of the condensing plume as the cw-laser does in Kingston’s et al. system. We

use a 40mm-diameter quartz tube as reaction chamber and the already above

mentioned 25mm coaxial inner tube.

Figure 5. A novel LA set-up using a CO

and a Nd:YAG laser system simultaneously. Mass flow

and pressure control units as well as auxiliary devices are not depicted in this schematic.

2.4. LASER ABLATION PARAMETERS

The flow in the reaction tube should be kept constant in the range between 100

and 350 sccm. While ablating a pressure of 250 to 550 Torr is to be maintained.

Below 100 Torr the formation of amorphous carbon is favored. Munoz et al.

observed gas and pressure effects on the local temperature conditions. Using

pure helium no SWNT are formed. Helium and light gases are more efficient

for cooling so that the resulting temperature gradient is too large to favor

SWNT growth. In opposition, argon and nitrogen are heavy enough to keep the

temperature gradient at a point where suitable growth conditions prevail.

Increasing the pressure above 400 Torr will soar-up the collision propability of

the evaporated material with the gas molecules. This will diminish the

capabiltity of SWNT forming.

Adding no catalyst to the graphite target no SWNT will be formed. In case

of a small amout of Ni or other metal given to the target SWNT-forming will

occur. Possible transition metals may be Ni > Co >> Fe > Pt, where yields

decrease by a few percent in the same order. The synthesized tubes are few,

isolated and often covered by a-C. In case of a bi-material combination a web-

like structure is produced where thick and relatively clean bundels of SWNT

can be observed. Yields may be over a hundred times higher than achieved with

one-dopant targets. Guo et al. found the following sequence for bi-material

compositions and Nd:YAG systems: Ni/Co ~ Co/Pt > Ni/Pt >>Co/Cu. Munoz et

al.

examined bi-material mixtures for CO

-laser systems and concluded

Ni/Co~Ni/Y>>Ni/Fe~Ni/La~Co/Ni.

For our initial experiments we use mainly Ni/Y (4.2at.%/1at.%), Ni/Y

(2at.%/0.5at.%) and Ni/Co (2at.%/2at.%) plus carbon compositions. One target

is around 8g in weight. The very dryed mixture is mixed by a tooth-pick first

before we put it into a tumbling machine for at at least 48h to assume a

macroscopic homogenious intermixture. Afterwards a 15mm-diameter pellet is

pressed in a 150°C warm tool at 8kN. The optimum operation parameters of our

laser-furnace set-up, concerning preassure, carrier gas, flow and temperature as

well as applied laser power will be puplished elsewhere as soon as our

experiments are completed. Our momentary production record at a quite

acceptaple quality is at 200mg/h without S. With only Nd:YAG we attchieved

only 60mg/h. This is a triplication of the production rate. The composition of

the target was C:Ni:Co 96 at.%: 2 at.% : 2 at.%. The pure Ar gas flow was

controlled to be 350 sccm.

3. Synthesis product evaluation

Our experience is that a highly optimized process for arcD may yield up to 55

vol.% SWNT. The ArcD process delivers in contrast to LA rather short carbon

nanotubes (some microns) with a quite broad diameter distribution. Laser

ablated SWNTs are remarkably uniform in diameter and self-organized into

bundles via van-der-Waals forces. Consisting of 100 to 500 SWNTs, their

diameters range from 10 to 20 nm and their lengths can reach values between

10 and some 100 Pm. In the synthesis of arcD tubes quite low defect densities

can be observed. The high temperatures in the plasma usually ensure a complete

closure of the tube lattices. An even lower defect density can be expected in

single-walled nanotubes produced in the laser ablation set-up. Here, the few and

minor fluctuations in the process conditions are less important. The more stable

conditions are also the reason for the longer, more identical tubes and for the

higher purity. The energy in the arcD process is high enough to evaporate the

material carefully. But due to the alternating conditions a favorable

condensation of the material is hindered. By TEM one can observe structural

differences between both synthesis products concerning the catalysts: ArcD

tubes usually encase the catalyst particle with which help the tube grows. In LA

material it is more common to have the particle somewhere in the soot but not

encased or attached to the tube. For getting rid of the catalyst content in the

end-product more easily this cartelistic is opportune.

The production rate in the arcD is by far higher. While the rate of an arcD

process is usually measured on the 100 mg/min scale the production rate in the

LA is expressed by the 10mg/h or 100mg/h scale. But one has to keep in mind

that the yield of SWNT in the as produced arcD soot does not surpass 55 vol.%

today. Of course, today it is not possible to produce SWNT of LA quality by

ArcD.

There are possibilities to clean the soot from all unwanted content, like

catalysts, a-C, fullerenes and graphite by purification. However, according to

our experiences so far, this might be applicable for small quantities or for a-C

and fullerenes but in case of an overall purification and a mass production these

cleaning steps for arcD material would turn out to be the most cost intensive.

Furthermore, by running through certain purification steps, not just the

unwanted material will be attacked but also the SWNTs will be harmed to some

extent.

Quality characterization in the SWNT field is a complicated issue. To

ensure the comparability of our measurements we have been working on a

protocol for quality control of single-walled carbon nanotube material. This

protocol is regularly updated.

The momentary version of the protocol includes

standards for homogenization, bucky paper preparation, composites, electrical,

x-ray diffraction SEM, TEM, optical spectroscopy and Raman measurements.

We stress the importance of working with large homogenized batches

(Presently, we use 100 g batches of ArcD material, whenever possible). A short

excerpt of our protocol is presented here.

3.1. ELECTRICAL MEASUREMENTS

Figure 6 shows a simple device to measure the conductivity of bucky papers

and of thin composite films using the 4 leads method. A strip punched off from

a bucky paper or a composite film is placed under the two outer (current)

electrodes and held with 5 mm wide clamps. Two inner (voltage) electrodes

with 10mm distance are put onto the sample orthogonally to the surface and

pressed with a constant force of 0.6 N. To this end a weight made of a stainless

steal block is placed on the body holding and separating the measuring

electrodes (insulating PTFE/PE, 10

:cm).

Figure 6. Device to measure the electric conductivity with the four-leads-method.

Measurement: Constant current (mA) is applied to determine the voltage

drop between the inner electrodes. The thickness of the strip (usually between

60 and 100 µm) is measured. At least three strips of one bucky paper should be

measured to estimate the mean electrical conductivity and the standard

deviation. The conductivity

is calculated as follows:

R = ǻU / I

ı = 1/R ǜ L/(T*W) § conductivity in Siemens per centimeter

[S/cm = (ȍcm)

-1

]

L = 10 mm § distance between the inner electrodes (constant)

W = 2 or 5 mm § width of the sample (constant)

T § lowest thickness of the bucky paper strip

______

Remark: A strong scattering of the conductivity values (larger than 15%) is evidence of

problems in the preparation of bucky paper and the preparation should be repeated.

3.2. ELECTRON MICROSCOPY

Note that electron microscopy (SEM and TEM) is not well suited to

characterise large batches. Sample preparation and image evaluation is very

"selective", i.e. the result depends on the skill and patience of the spectroscopist

and on what he hopes to see. Usually "best" data are shown rather than

"representative" data. But the methods are very useful to check whether in a

particular sample "there are nanotubes at all". When new synthetic routs for

nanotube production are developed, SEM very often gives the first positive hint

of success. SEM characterization can be carried out on powder deposited on the

sticky side of scotch tape or on a bucky paper.

Figure 7. Visual comparison of SWNT-material produced by ArcD (left) and LA (right). Both

TEM-pictures were taken at same magnification. In the ArcD material dirt and irregular large

particles are clearly recognizable.

3.3. X-RAY POWDER DIFFRACTION

We execute x-ray measurements in two ways. The first one is to use a

Lindeman tube. Here, the nanotube powder is filled into a Lindeman tube (thin

glass capillary, as commonly used for X-ray powder diffraction), diameter 0.3

or 0.5 mm. The second possibility is to use a flat bed sample holder. About 5

mg of powder is distributed on a scotch tape and put into the sample holder.

Both sample holders rotate while measuring. Fig. 8 (left) shows a typical

powder diffractogram of carbon arc material after various purification steps

(Copper KD radiation, 20 minutes exposure time, Lindeman tube,

multidetector). If there are nanotube bundles in the sample, a peak will show up

at 6°, which originates from the triangular lattice of the tubes in the bundle

(lattice constant about 10 Å). The central curve shows a slight indication of

such a peak. The peak at ~26° corresponds to graphite. At higher scattering

angles there are the peaks of metal (raw material) or metal oxides (after heating

in air). For quantitative measurements the SWNT material is mixed in a ratio of

1:1 with fullerene C

. X-ray powder diffraction patterns are recorded using the

flat bed method (By filling into a Lindeman tube the sample would de-mix).

The area of the graphite peak at 26° is compared with that of the second

fullerene peak at 16° (which is used as a marker).

Figure 8. A typical powder diffractogram of ArcD material after various purification steps.

Figure 9. The pattern of a sample with unknown graphite concentration is shown. Now the peak

area between graphite and fullerene marker compare like 1 to 5, indicating that the sample

contains about 20% of graphite contamination (In addition to crystalline graphite there probably

also is an important contamination of turbostratic graphite leading to the broad peak between 15°

and 25°).

3.4. OPTICAL SPECTROSCOPY

For a first orientation, absorption spectra of films sprayed onto a glass substrate

are useful. Therefore, SWNTs are suspended in 1% SDS aqueous solution and

sprayed onto microscope slides using the airbrush technique. Fig. 9 shows the

spectrum of SWCNT after subtraction of an appropriate baseline. The peaks are

assigned to transitions between the van Hove spikes in the electronic density of

states of nanotubes. S11 and S22 are related to semi-conducting tubes and M11

originates from metallic tubes. S11 depends very much on the chemical

environment of the sample (e.g. doping) and is not well suited for a quantitative

analysis. The background is due to carbonaceous by-products, like amorphous

carbon and graphitic particles, and also to excitations in the nanotubes

themselves (but not directly related to the van Hove spikes). For a quantitative

analysis we follow Itkis et al. method.

30,31

This analysis only yields relative

nanotube concentrations, relative to a standard of known concentration. For

ArcD material we use the standard (A(S

)/A(T) = 0.141) published in.

For LA

material we use A(S

)/A(T) = 0.211, a value which we measured on some

material which was delivered by FZ Karlsruhe (Germany) and purified by our-

selves. NIR spectroscopy and electrical measurements in bucky paper correlate

well as can be seen in Figure 10.

Figure 10. Spectrum of SWCNT after the subtraction of a proper baseline.

Figure 11. NIR spectroscopy in solution and electrical measurements in bucky paper correlate

well. Here we purified ArcD material in a hot furnace at a set temperature for differently long

periods.

3.5. RAMAN SPECTROSCOPY

Figure 12 shows a Raman spectrum measured by a Raman microscope,

excitation line O = 632 nm. The double line at |1600 cm

-1

is the so called G-line

coming from the stretching of conjugated double bonds. The lines around

200 cm

-1

are from the radial breathing mode (RBM) and depend on the

nanotube diameter, the line at 1300 cm

-1

(D-line) is from the multiple phonon

scattering and from the nanotubes defects. Spectra are normalized to the D*-

mode (2600 cm

-1

). Note that Raman spectra are resonance enhanced and depend

very much on the excitation line. For our believe, Raman spectra are not useful

for quantitative analysis, but the intensity of the D-line is often taken as

evidence of many defects on the tube or of the presence of amorphous carbon in

the sample. In addition, like SEM and TEM images, Raman spectra often give

the first hint of the presence of single-walled carbon nanotubes in a sample and

then justify further investigations.

Figure 12. Raman lines from ArcD and LA material at 632.817nm. LA nanotubes have a

narrower diameter distribution.

4. Summary

In comparison to the laser ablation method, the arc discharge SWCNT

synthesis is still the most efficient method to produce large quantities of

nanotubes. Additionally, it is also the most widly used one which also traces

back to the relative simple basic set-up. In combination of optimized process

parameters and additional ingredients like sulfur yield and quality can be

influenced positively. Although the relative amount of SWCNT in the soot can

be increased by varying the process parameters and by gas/ingredient variations

the synthesis process regime is difficult to control. Hence, in case of need for

very pure material with a narrow diameter distribution the laser ablation method

is the process of choice. Our approach to use a Nd:YAG system at 1064nm and

a CO

system at 10.6 microns simultaneously is already capable to triple the

production rate to 200mg/h. A main question is whether laser ablation may ever

be a synthesis method to produce bulk quantities or if it stays a method for

nanotubes for specialized applications. For the evaluation of SWNT material we

developed a protocol

which is regularly updated. It is one of the most

important means for our work on synthesis and purification.

References

1. S. Iijima, Helical microtubules of graphitic carbon, Nature 354, 56–58 (1991).

2. S. Iijima and T. Ichihashi, Single-shell carbon nanotubes of 1-nm diameter, Nature 363,

603–605 (1993).

3. D. S. Bethune, C. H. Kiang, M. S. de Vries, G. Gorman, R. Savoy, J. Vazquez, and R.

Beyers, Cobalt-catalysed growth of carbon nanotubes with single-atomic-layer walls, Nature

363, 605–607 (1993).

4. W. Krätschmer, L. O. Lamb, K. Fostiropoulos, and D. R. Huffman, Solid C60: a new form

of carbon, Nature 347, 354–358 (1990).

5. Y. L. Hsin, K. C. Hwang, F.-R. Chen, and J.-J. Kai,

Production and in-situ metal filling of

carbon nanotubes in water, Advanced Materials 13(11), 830-833 (2001).

6. H. W. Zhu, X. S. Li, B. Jiang, C. L. Xu, Y. F. Zhu, D. H. Wu and X. H. Chen Formation of

carbon nanotubes in water by the electric-arc technique. Chem. Phys. Lett. 366, 664-669

(2002).

7. H. Lange, M. Sioda, A. Huczko, Y. Q. Zhu, H. W. Kroto and D. R. M. Walton, Nanaocarbon

production by arc discharge in water, Carbon 41, 1617-1623 (2003).

8. L. P. Biró, Z. E. Horváth, L. Szalmás, K. Kertész, F. Wéber, G. Juhász, G. Radnóczi and J.

Gyulai, Continuous carbon nanotube production in underwater AC electric arc, Chemical

Physics Letters 372, 399-402 (2003).

9. Ch. Bower, O. Zhou,

W. Zhu, D. J. Werder, and S. Jin, Nucleation and growth of carbon

nanotubes by microwave plasma chemical vapor deposition, Applied Physics Letters 77,

2767-2769 (2000).

10. M. Chhowalla, K. B. K. Teo, C. Ducati, N. L. Rupesinghe, G. A. J. Amaratunga, A. C.

Ferrari, D. Roy, J. Robertson, and W. I. Milne, Growth Process Conditions of Vertically

Aligned Carbon Nanotubes Using Plasma Enhanced Chemical Vapor Deposition, J. Appl.

Phys. 90(10), 5308-5317 (2001).

11. L. Delzeit, C. V. Nguyen, R. M. Stevens, J. Han, and M. Meyyappan, Growth of carbon

nanotubes by thermal and plasma chemical vapour deposition processes and applications in

microscopy, Nanotechnology 13, 280-284 (2002).

12. J. Han, J.-B. Yoo, C. Y. Park, H.-J. Kim, G. S. Park, M. Yang, I. T. Han, N. Lee, W. Yi, S.

G. Yu, and J. M. Kim, Tip Growth Model of Carbon Tubules Grown on the Glass Substrate

by Plasma Enhanced Chemical Vapor Deposition, J. Appl. Phys. 91, 483-486 (2002).

13. T. Guo, P. Nikolaev, A. Thess, D. T. Colbert, and R. E. Smalley, Catalytic growth of single-

walled nanotubes by laser vaporization, Chem. Phys. Lett. 243, 49–54 (1995).

14. M. E. Itkis, D. E. Perea, S. Niyogi, J. Love, J. Tang, A. Yu, C. Kang, R. Jung, and R. C.

Haddon, Optimization of the Ni-Y Catalyst Composition in Bulk Electric Arc Synthesis of

Single-Walled Carbon Nanotubes by Use of Near-Infrared Spectroscopy, J. Phys. Chem. B

108, 12770-12775 (2004).

15. Y. S. Park, K. S. Kim, H. J. Jeong, W. S. Kim, J. M. Moon, K. H. An, D. J. Bae, Y. S. Lee,

G.-S. Park, and Y. H. Lee, Low pressure synthesis of single-walled carbon nanotubes by arc

discharge, Synthetic Metals 126, 245-251 (2002).

16. C. Liu, H. M. Cheng, H. T. Cong, F. Li, G. Su, B. L. Zhou, M. S. Dresselhaus. Synthesis of

macroscopically long ropes of well-aligned single-walled carbon nanotubes, Advanced

Materials 12, 1190-1192 (2000).

17. H. W. Zhu, B. Jiang, C. L. Xu, and D. H. Wu, Synthesis of High Quality Single-walled

Carbon Nanotube Silks by the Arc-discharge Technique, J. Phys. Chem. B 107, 6514-6518

(2003).