Review

Recent advances in carbon nanotube-based electronics

Prithu Sharma *, Prerit AhujaIndian Institute of Technology Madras, Chennai 600036, India

Received 10 December 2006; received in revised form 24 June 2007; accepted 18 October 2007

Available online 23 October 2007

Abstract

CNT-electronics is a field involving synthesis of carbon nanotubes-based novel electronic circuits, comparable to the size ofmolecules, the practically fundamental size possible. It has brought a new paradigm in science as it has enabled scientists to increasethe device integration density tremendously, hence achieving better efficiency and speed. Here we review the state-of-art currentresearch on the applications of CNTs in electronics and present recent results outlining their potential along with illustrating somecurrent concerns in the research field. Unconventional projects such as CNT-based biological sensors, transistors, field emitters,integrated circuits, etc. are taking CNT-based electronics to its extremes. The field holds a promise for mass production of highspeed and efficient electronic devices. However, the chemical complexity, reproducibility and other factors make the field achallenging one, which need to be addressed before the field realizes its true potential.# 2007 Elsevier Ltd. All rights reserved.

Keywords: A. Nanostructures; B. Vapor deposition; C. Atomic force microscopy; D. Electrical properties; D. Thermal conductivity

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2518

2. Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2518

2.1. Engineering carbon nanotubes and nanotube circuits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2518

2.2. Manipulating CNT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2518

2.3. Large-scale integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2519

2.4. Collating nanotubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2519

3. Current initiatives by the community . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2519

3.1. Integrated carbon nanotube sensors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2519

3.2. CNT rings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2519

3.3. Carbon nanotube—CMOS chemical sensor integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2520

3.4. Light-induced electron transfer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2520

3.5. Nano-oscillator. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2520

3.6. Ring oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2520

3.7. Carbon nanotube field-effect transistors (CN-FET) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2520

3.8. Diodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2521

3.9. Millipede technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2521

www.elsevier.com/locate/matresbu

Materials Research Bulletin 43 (2008) 2517–2526

* Corresponding author. Tel.: +91 9840286464; fax: +91 44 22574752.E-mail addresses: prithu129@gmail.com, prithu@iitm.ac.in (P. Sharma).

0025-5408/$ – see front matter # 2007 Elsevier Ltd. All rights reserved.doi:10.1016/j.materresbull.2007.10.012

3.10. Quantum capacitance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2522

4. Applications of CNT in electronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2522

4.1. CNT-enhanced ultracapacitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2522

4.2. Bio-chips. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2522

4.3. Nonvolatile electro-fluidic memory devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2523

4.4. Field emission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2523

4.5. Constructive destructive technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2524

4.6. Transistors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2524

4.7. Logic circuits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2524

4.8. Organic light emitting diodes (OLEDs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2524

5. Obstacles to success . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2524

5.1. Thermal conductivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2524

5.2. Reproducibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2524

5.3. Chemical complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2525

5.4. Expensive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2525

5.5. Limitations of CNT field emission electron sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2525

5.6. Friction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2525

6. Future of CNT-electronics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2525

7. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2526

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2526

1. Introduction

In 1991 Iijima synthesized new type of fullerenes, quasi-one-dimensional crystalline structures of carbon atoms,generally referred to as carbon nanotubes [1]. A nice thing about nanotechnology is that as we look into the options, theyexpand and become truly incredible. Nanotubes are promising building blocks for nanoscale electronic circuits, and theypossess novel properties due to their small sizes, which offer exciting possibility for smaller and faster devices with betterperformance. In the rapidly developing field of nanotechnology – doing things at a scale 100,000 times narrower than ahuman hair – nanodevices are becoming an increasingly key component in everything from drug delivery to improving oreven replacing the microprocessors in computers or optical switches in telecommunications networks. Nanotubes, tinyhollow carbon filaments about one ten-thousandth the diameter of a human hair, are already famed as one of the mostversatile materials ever discovered. A hundred times as strong as steel and one-sixth as dense, able to conduct electricitybetter than copper or to substitute for silicon in semiconductor chips, carbon nanotubes have been proposed as the basisfor everything from elevator cables that could lift payloads into Earth orbit to computers smaller than human cells.Increasing efficiency through smaller components is the key towards miniaturization of technology. The use of carbonnanotubes could find successful use from sophisticated, niche applications to everyday electronics.

2. Approaches

2.1. Engineering carbon nanotubes and nanotube circuits

A simple and reliable method has been demonstrated for selectively removing single carbon shells from MWCNTsand single-walled nanotube (SWCNT) ropes to tailor the properties of these composite nanotubes is called electricalbreakdown. Different shells of MWCNTs can be characterized individually by removing the shells stepwise. Bychoosing among the shells, we can convert a MWCNT into either a metallic or a semiconducting conductor, as well asdirectly address the issue of multiple-shell transport. With SWNT ropes, similar selectivity allows us to generate entirearrays of nanoscale field-effect transistors based solely on the fraction of semiconducting SWCNTs [2].

2.2. Manipulating CNT

The successful integration of SWCNTs with the current silicon infrastructure is critical to ensuring the economicsuccess of nanotube-based electronics. Prior to large-scale integration, though, the fundamental physical properties of

P. Sharma, P. Ahuja / Materials Research Bulletin 43 (2008) 2517–25262518

SWCNTs in direct contact with silicon need to be investigated via nanoscale probing techniques such as ultra-high-vacuum scanning tunneling microscopy (UHV STM). These advances have enabled electronic characterization ofSWCNT/silicon interfaces via STM spectroscopy and physical manipulation/translation of SWCNTs on the substratesurface, both of which are vital steps toward successfully creating a transistor device that directly integrates SWCNTswith silicon. Even the axial twist of the nanotubes, when manipulated, can vary their conductivity significantly. Byintroducing structural defects in an ordered way in CNTs, could expand their electronic properties and open the path tonano-electronics.

2.3. Large-scale integration

A large-scale assembly method to deposit discrete multi-walled carbon nanotubes (MWCNTs) across gaps in anelectrode array has been developed with 90% yield. A composite electric field combining an ac with a dc electric fieldcan be introduced to deposit MWCNTs on electrodes; it can be shown that the ac field serves to selectively attract, andthe dc field to guide individual deposition. Such developed experimented method is scalable, and can be extended toultra-high-density (nanoscale) deposition [3].

2.4. Collating nanotubes

Devices for IR detection and imaging require large area, densely packed detectors in a well-defined array structurewith efficient but also easy light coupling, high responsivity, good pixel to pixel uniformity, low dead pixel count andexcellent stability. The CNT array IR detector designs have a number of technical advantages over the quantum wallinfrared photodetectors (QWIP). Therefore large area, ordered and well aligned high-density of uniform nanotubeswould be very desirable [4].

Taking into account all above approaches, we define CNT-electronics as a field involving integration of CNTs intonovel devices for which, otherwise, miniaturization is a tough challenge. Here we define ‘devices’ abstractly so as toaccommodate the various diverse but overlapping approaches mentioned above.

3. Current initiatives by the community

Researchers worldwide are using the above-mentioned approaches to explore the scope of CNT-electronics andsimultaneously develop foundational tools and techniques in this field. In this section, we discuss various initiativestaken by the researchers.

3.1. Integrated carbon nanotube sensors

According to MITs Microsystems Technology Laboratories annual research report CNTs grown through carbonvapor deposition and devices are fabricated for an ultra-low-power wireless sensing system. The main goals are tobuild a CNT sensor array with high yield of semiconducting SWCNTs, high sensitivity and selectivity of gases, andlow variability in the performance of the device. In addition, chemical functionalization of CNT sensors will allow thetarget application to detect different types of toxic gases for environmental and industrial applications. Arrays andnetworks of metallic and semiconducting carbon nanotubes are finding application in flexible electronics, as chemicaland biological sensors, and as electronic interconnects. However, these systems demonstrate an environmentalsensitivity and chemical reactivity that complicates processing, and facile incorporation into devices. An ultra-high-sensitivity down to a single electron charge can be achieved by a CNT sensing channel at room temperature. In suchapplications electrical response of the sensor strongly depends on the position of the sensing charge because of the nearballistic transport and quantum interference in a submicron CNT channel.

3.2. CNT rings

Carbon nanotube rings have potential applications for both complementary metal-oxide semiconductor (CMOS)interconnects and devices. As copper interconnect dimensions shrink, the copper bulk resistivity is expected toincrease due to surface scattering effects. An alternative to standard copper interconnects is to use carbon nanotubes,

P. Sharma, P. Ahuja / Materials Research Bulletin 43 (2008) 2517–2526 2519

which have demonstrated ballistic transport properties and high achievable current densities. Wunsch et al. of MIT arecurrently investigating a method by which pre-grown carbon nanotubes are assembled onto a surface to form aninterconnect structure. In addition, the possibility of using these carbon nanotube rings as inductors in CMOS circuitsis also being investigated. It has been observed carbon nanotube rings with diameters in the range of hundreds ofnanometers. Because current state-of-art CMOS metal inductors are generally tens to hundreds of microns in size, ifthese nanotube rings are suitable for inductor applications, we can achieve a dramatic improvement in device density.

3.3. Carbon nanotube—CMOS chemical sensor integration

The CNT changes its conductance when exposed to certain chemicals [5], and thus we can effectively utilize CNTsas resistive chemical sensors. A new method of growing carbon nanotubes is predicted to revolutionize theimplementation of nanotechnology and the future of electronics. Researchers at the University of Cambridge havesuccessfully grown nanotubes at a temperature, which permits their full integration into present complementary metal-oxide semiconductor (CMOS) technology (350 8C). It has been proposed that carbon nanotube growth is governed bythe catalyst surface without the necessity of catalyst liquefaction [6]. The stochastic nature of CNT chemical sensorscalls for multiple deployments of CNT sensors in one sensor node. This constraint, in turn, requires an efficientalgorithm to infer the concentration of the chemical. Thus an energy efficient algorithm that can be operated in realtime is also being developed.

3.4. Light-induced electron transfer

Scientists have recently shown that altering carbon nanotubes results in electrons supply when exposed to light.This was done by having two flat rings of carbon molecules sandwich a ferrocene molecule. Ferrocene is known for itstendency to relinquish electrons. When exposed to visible light, the carbon atoms accepted the ferrocene molecule.This is the first time that carbon nanotubes have been hybridized to undergo light-induced electron transfer.Researchers say that these modified carbon nanotubes are the first step in building solar cells using this technology.

3.5. Nano-oscillator

The need for motors and oscillators is there in the smallest devices, assembled at the molecular level. According toUC Riverside Mechanical Engineering Professor Qing Jiang, bundling groups of carbon nanotubes together couldmake an ultra-efficient and accurate nano-oscillator. The schematics of a CNT bundle oscillator, which could beinitiated by an electrostatic capacitive force, have been presented. While the capacitive force acting on a CNToscillator extruded it, the force exerted on the CNT oscillator by the excess van der Waals energy sucked it into thebundle. Therefore, the CNT oscillator could be oscillated by both Coulomb and the van der Waals interactions. Theoperation frequency of a CNT bundle oscillator could be controlled by both the size and the length of the bundle [7].

3.6. Ring oscillator

A complete electronic circuit around a SWCNT has been built by the IBM researchers. This is the first complexintegrated circuit implementation that is entirely done on a single molecule. The IBM scientists will now use the ringoscillator to test improved CNT transistors and circuits, and to benchmark their performance in complete chip designs.However, scientists have focused so far on fabricating and optimizing individual carbon nanotube transistors. Now,they can evaluate the potential of carbon nanotube electronics in complete circuits—a critical step toward theintegration of the technology with existing chip-making techniques [8].

3.7. Carbon nanotube field-effect transistors (CN-FET)

Carbon nanotube (CNT) is one of the candidates for a quantum wire for the molecular-FET. Multi-channel CN-FETs have been realized by depositing a large number of CNTs onto a metallic back gate. A current gain cut-offfrequency of 8 GHz has been obtained. This work clearly demonstrates that CN-FETs are promising components forhigh-frequency (HF) applications. Recently, several papers have been reported on the CNTs for FETs and CNT-logic

P. Sharma, P. Ahuja / Materials Research Bulletin 43 (2008) 2517–25262520

applications [9]. IBM has fabricated carbon nanotube field-effect transistors (CNT-FETs) based on the technique ofalteration by electrical breakdown (Fig. 1).

3.8. Diodes

A nanotube formed by joining nanotubes of two different diameters end to end can act as a diode, suggesting thepossibility of constructing electronic computer circuits entirely out of nanotubes. A ‘p–n junction’ is simulated alongthe single-walled carbon nanotube channel using two separate gates close to the source and drain of the CNTFET,respectively. The Schottky barrier field-effect transistor mechanism-based calculations of the current–voltagecharacteristics of the double-gated CNTFET shows a good rectification performance of the p–n junction [11].

3.9. Millipede technology

Millipedes are non-volatile memories stored on nanoscale pits burned into the layer of a thin polymer, where data isread and written using an array of NEMS-based probes. This technique of data storage offers a density of more than 1TB per square inch, about 20 times the density of the best magnetic storage available today. The technology offersgreat potential for replacing magnetic storage hard drives, and also reduction in the form-factor in case of Flash drives.IBM has already demoed a millipede prototype, and is promising to make the technology commercially available by2007. The disadvantage that it will be costly when launched will be offset by the storage capacity that it would offer(Fig. 2).

P. Sharma, P. Ahuja / Materials Research Bulletin 43 (2008) 2517–2526 2521

Fig. 1. Scanning electron microscopy of a CN-FET with two-finger back gate [10].

Fig. 2. Millipede-a nanomechanical storage device (courtesy of IBM [12]).

3.10. Quantum capacitance

Recently Lugli et al. have exploited and reported the quantum capacitance effects in CNT-based devices [13]. Thetransport properties of co-axially gated CNTs were studied. Their work introduces both limitations and new interestingoperational modes for ultra-short one-dimensional field-effect devices, and also a particular attention in the design ofthe gate electrostatics is required. In a recent work by Ilani et al. they have used field-effect transistor geometry toobtain the first direct capacitance measurement of individual carbon nanotubes, as a function of the carrier density[14].

Complex molecules with moving parts that can cross-surfaces and be steered by electrical or magnetic fields knownas nanocars have been developed at the Rice University. They can be employed to ferry the loose end of a carbonnanotube to a point where it can connect to make a circuit.

The first complete electronic integrated circuit around a single carbon nanotube molecule has been developed byIBM. The significance of this achievement lies in the standard semiconductor processes used for building thisintegrated circuit.

4. Applications of CNT in electronics

CNT has the potential to make the process of development of electronics comprehensible to us as well asconquering many of the size limitations of the circuits with possible applications in integrated circuits and energyconservation. It is believed that CNT-electronics shares the potential, together with Biotechnology and ArtificialIntelligence to improve current devices. Such advances can then be used to solve problems not possible in presentscenario.

Conductive and high-strength composites; energy conversion and energy storage devices; sensors; field emissiondisplays and radiation sources; hydrogen storage media; and nanometer-sized semiconductor devices, probes, andinterconnects are some of the many potential applications based on carbon nanotubes. Some of these applications arenow realized in products. Others are demonstrated in early to advanced devices, and one, hydrogen storage, is cloudedby controversy. Nanotube cost, polydispersity in nanotube type, and limitations in processing and assembly methodsare important barriers for some applications of single-walled nanotubes.

Few possible applications of CNT in electronics are discussed below:

4.1. CNT-enhanced ultracapacitors

Ultracapacitors or double layer capacitors (DLCs) are energy storage devices whose operation is based on thedouble layer effect [15]. It has been shown in a project undertaken by laboratory of electromagnetic and electronicsystems of MIT that utilization of a matrix of vertically aligned CNTs as electrode structure, can lead to anultracapacitor characterized by a power density greater than almost four orders of magnitude higher than batteries, alifetime longer than 300,000 cycles, and an energy density higher than 60 W/kg [16].

The significant enhancement in the achievable DLC power density derives from the high conductivity obtainablewith CNTs, which in the limit of a few microns in length present ballistic conduction. The energy density improvementof a ‘‘nanotube-enhanced electrode’’ is due to the higher effective surface area obtainable with a structure based onvertically aligned nanotubes over activated carbon (Fig. 3).

4.2. Bio-chips

Selective covalent functionalization of probe oligonucleotides has been achieved through the formation of amidebonds at the exposed end of MWCNT arrays which have been successfully fabricated on micropatterns. Directelectrochemical detection of the oxidation signal of inherent guanine bases in the target nucleic acids has beendemonstrated with both oligonucleotide and polymerase chain reaction (PCR) amplicon targets. Low-cost disposablechips for rapid molecular analysis using handheld devices are ideal for space applications. By incorporating nanoscaleelements in diagnostics devices a miniaturized electronics chip will detect a specific biomarker signature withextremely high sensitivity and simplified sample preparation. Gene mutations are known to be the major causes for thedevelopment of cancer and other genetic diseases. The high radiation and microgravity in outer space environments

P. Sharma, P. Ahuja / Materials Research Bulletin 43 (2008) 2517–25262522

would greatly increase such risks. Simple, low cost, but accurate methods for rapid gene analysis are highly demandedfor health and bioenvironmental monitoring in long-duration human flights. This electronic detection platform can beintegrated with microfluidics and microelectronics as highly automated disposable chips for human flights [17].

4.3. Nonvolatile electro-fluidic memory devices

Choi et al. of Chung-Ang University, Korea, working for Samsung have investigated the schematics and operationsof several electro-fluidic shuttle (EFS) memory devices based on carbon capsules and carbon peapods using moleculardynamics simulations. The system proposed by them can operate a non-volatile memory device.

4.4. Field emission

What is the state-of-art after years of research on field emission from carbon nanotubes? It is proven that the nanotubesare excellent electron sources, providing a very stable current at very low fields and capable of operating at moderate

P. Sharma, P. Ahuja / Materials Research Bulletin 43 (2008) 2517–2526 2523

Fig. 3. Matrix of vertically aligned CNTs (courtesy of MIT [16]).

Fig. 4. A field-emission display using CNTs (courtesy of Samsung [18]).

vacuum. Different methods are available to deposit various types of nanotubes on surfaces. Techniques have beendeveloped to pattern the films, to vary the density of nanotubes and their orientation on the substrate and therefore tocontrol their emission properties. It is further possible to place and manipulate one single nanotube on a support. Differentproperties have been measured, the workfunction have been estimated and emission models have been proposed (Fig. 4).

4.5. Constructive destructive technique

IBM has successfully developed a technique to construct a dense array of semiconductor CNT. The techniqueconsists of first depositing both the metallic and semiconducting nanotubes on the silicon oxide substrate and thendestroying the metallic nanotubes by applying appropriate voltage thus leaving only the semiconducting nanotubesbecause the semiconducting nanotubes did not allow any current to pass through them.

4.6. Transistors

IBM has made CNFET resembling to conventional MOSFET having conductional channel beneath the gateelectrodes separated by a thin dielectric. The top gate devices exhibited excellent electrical characteristics, includingsteep sub-threshold slope and high transconductance at low voltages by reducing the gate-to-channel separation.Furthermore, the IBM scientists were able to fabricate both hole (p-type) and electron (n-type) transistors. The top-gate design allows independent gating of each transistor, making it possible to generate CMOS (complementary metal-oxide semiconductor) circuits that have a simpler design and consume less power. The nanotube devices in this caseoutperformed the prototype silicon transistor.

4.7. Logic circuits

The first inter-molecular logic circuit has been created by IBM. The circuit is a voltage inverter made by using twonanotube field-effect transistors.

4.8. Organic light emitting diodes (OLEDs)

Yang et al. of Zhejiang University have successfully synthesized OLEDs based on MWNTs modified electrode[19]. In order to investigate the role of SWCNTs in a hole conducting polymer, OLEDs were fabricated with aconjugated emissive copolymer, poly(3,6-N-2-ethylhexyl carbazolyl cyanoterephthalidence) (PECCP) and SWCNTsdispersed in a hole conducting buffer polymer, polyethylene dioxythiophene (PEDOT).

5. Obstacles to success

Many problems confront CNT-electronics before it goes into full-fledged applications. These problems come indiverse colors; ranging from fundamental understanding to commercial production.

5.1. Thermal conductivity

It has been experimentally found that the critical radius below which the thermal conductivity of the nanocompositeis less than the conductivity of the matrix varies between 10 nm and 20 nm. It is because of this reason that carbonnanotube-based composite have not been able to achieve high-thermal conductivity (k). Researchers have been tryingto enhance the thermal conductivity using a mixture of micro- and nano-sized particles by providing a percolatingchain between the larger particles, but without much success.

5.2. Reproducibility

It has been difficult to have precise control over the process to replicate nanotubes having identical diameter andchirality. Development of large scale, high-productivity synthesis methods and long range order assembly processesare still not viable.

P. Sharma, P. Ahuja / Materials Research Bulletin 43 (2008) 2517–25262524

5.3. Chemical complications

Satisfactory ways to solve the problem of entanglement of CNT and achieve stable dispersions, posed by mutualattraction is yet to be solved. Because of this mutual attraction the nanotubes tend to stick together in hairball-likeclumps. To date, detergent and water solutions have been the preferred medium that contain less than 1% of dispersednanotubes by volume and are processed with polymer solutions. Such concentrations are not suitable for commercialfabrication of large CNT fibers. Also, removal of the entire polymer and soap and the conversion of the nanotubes backinto their pure form are quite difficult.

5.4. Expensive

Presently, incorporation of CNTs in electronic applications is an expensive process. It is expensive because CNTproduction is quite costly. Taking the current average price of SWNTs around $75,000 per kg, it would take a fortune tosynthesize just a minimal circuit. Additionally, techniques for purifying them – sorting good tubes from bad ones – andways to incorporate tubes into other products need to be perfected. So, currently, mass production of applicationsbased on CNT-electronics is not yet feasible.

5.5. Limitations of CNT field emission electron sources

It has been found that the emission degradation becomes apparent for emission currents in the mA range for a singleemitter and the experimentally obtained field emission I–V characteristics suggests that power dissipation due highcontact or intra-CNT resistance is the cause of the emitter degradation. Therefore, in spite of knowing that thefundamental properties of CNTs are very favorable for the use as field emission tips, these properties alone will notguarantee their success in this area. Hence, at least for high-current applications, a perfect control of the catalytic CNTgrowth process is needed for successful CNT field emitter technology. Related questions such as the problem ofelectrical contacts to nanotubes also need to be addressed. The optimization of the emitter density, which is still poorlycharacterized, as well as the incorporation of nanotubes in gated devices, are also important milestones for theadvancement of the field.

5.6. Friction

The friction and heat limitations that are encountered by CNTs also pose a big problem. This leads to a big hurdle insuccessfully implementing CNTs in electronic circuits.

6. Future of CNT-electronics

CNTs have been evinced as the material of future which will assist in extending the Moore’s law, which says thatnumber of transistors per integrated circuit doubles every 18 months, and it has been the guiding principle for thesemiconductor industry for over 30 years. With the outstanding mechanical, electrical, thermal, and optical propertiesof carbon nanotubes (CNTs), now widely established [20,21], there is an emerging need for materials and methods thatintegrate carbon nanotubes as thin-films in standard microelectronic and micromechanical fabrication processes.

The prime driving force in microelectronic industry is due to the continual miniaturization of electronic devices.Researchers all over the world finally aim to synthesize devices as small as molecules or cluster of atoms. SWCNTsunusual electronic properties, if exploited, could lead to nanometer-sized electronic devices. In spite of havingenormous theoretical possibilities CNTs have not lived up to the hype surrounding their development. The highlylucrative potential of the thriving applications have forced the scientists to look for ways to use them.

The metal filaments in X-ray machines, which tend to burn out quickly, will 1 day be substituted by CNTs. This canlead to development of portable X-ray machines to be used in airport security, ambulances, and customs work. Flat-panel displays, microwave generators, devices for electric surge protection, single-electron transistors,nanolithography systems and high-intensity lamps are some of the great potential applications for the implementationof CNTs. A nanospintronics controlling a manipulation of electron spin would be useful electronics for informationdetection from nanostorage media and quantum spin detection or logic operations for future spin quantum devices.

P. Sharma, P. Ahuja / Materials Research Bulletin 43 (2008) 2517–2526 2525

The CNTs hold a firm promise to efficiently carry signals in optical fibers. The nanoTVs, analogous to CRT(cathode-ray tube) televisions, use CNTs as electron emitters to produce the picture. This idea has been floating aroundfor a couple of years, but the price decline in plasmas and LCDs is making them slightly less attractive. Thecombination of nanotubes flexure hinges and nanoscale sensors, actuators, and electronics forms the core of nextgeneration nanomechanical systems such as nanoscale positioners and nanoscale end effectors [22,23]. The highsensitivity and nonlinearity of CNT-based charge sensors play an important role in the future design of CNT-basednanocrystal memories and biological sensors.

Thus, no harm in believing that they could 1 day will be used to make everything from smaller, faster computerchips to lighter airplanes. Incorporating nanotubes into chips or devices for delivering medicines into the body maywell take up a few years down the line.

7. Conclusion

The introduction of optical storage devices a few years ago has dramatically changed the demand, stability andreliability scenario by superbly outperforming the magnetic storage devices.

CNT today, holds a similar promise to revolutionize the electronics world in the coming years. But to realize thispromise, it needs active participation from researchers form all over the world. Establishments such as IBM are doingan excellent job in promoting this field among young scientists. But more academic, industrial and governmentalparticipation is needed. Recently the US government announced a fund of $435 million for establishment of twoInstitutes for Nanoelectronics Discovery and Exploration (INDEX) in the country to promise a breakthrough in thefield of CNT-electronics and miniaturization of devices.

The field today needs to build a strong foundation. Researchers from all over the world have been continuouslyworking to achieve precision and economic production. The popularity of CNT-electronics is growing with the growthof manipulation techniques’ development. Further technology developments should be encouraged and supported.Commercial interests will drive more efficient and accurate synthesis of full-fledged electronic circuits andapplications.

Along with the efficiency, the cost of CNT synthesis has been remarkably reducing. If the trend continues and isvery likely to continue, then the CNT-based circuits’ synthesis can be made very affordable and viable for massapplications.

References

[1] S. Iijima, Nature 354 (1991) 56–58.[2] P.G. Collins, S. Michael, Arnold, P. Avouris, Science 292 (2001) 706–709.[3] J. Chung, K.H. Lee, J. Lee, R.S. Ruoff, Langmuir (2004).[4] J.M. Xu, Infrared Phys. Technol. 42 (2001) 485–491.[5] J. Kong, N.R. Franklin, C. Zhou, M.G. Chapline, S. Peng, K. Cho, H. Dai, Science 287 (5453) (2000) 622–625.[6] M. Cantoro, S. Hofmann, S. Pisana, V. Scardaci, A. Parvez, C. Ducati, et al. Nano Lett. 6 (2006) 1107–1112.[7] J.W. Kang, K.O. Song, H.J. Hwang, Q. Jiang, Nanotechnology 17 (2006) 2250–2258.[8] M. Kanellos, IBM Researchers Claim Nanotube First, CNET News.com, 23 March, 2006.[9] P.G. Collins, M.S. Arnold, P. Avouris, Science 292 (2001) 706–709.

[10] J.M. Bethoux, H. Happy, G. Dambrine, V. Derycke, M. Goffman, J.P. Bourgoin, Mater. Sci. Eng., B 135 (2006) 294.[11] J. Li, Q. Zhang, B. Mary, Chan-Park, Carbon 44 (2006) 3087–3090.[12] http://pda.physorg.com/lofi-news-data-millipede-tip_88.html.[13] L. Latessa, A. Pecchia, A.D. Carlo, P. Lugli, J. Comput. Electron. 4 (1–2) (2005) 51–55.[14] S. Ilani, L.A.K. Donev, M. Kindermann, P.L. McEuen, Nat. Phys. 2 (2006) 687–691.[15] H.V. Helmholtz, Ann. Phys. (Leipzig) 89 (1853) 211.[16] http://lees.mit.edu/lees/projects/cnt_ultracap_project.htm.[17] J. Koehne, H. Chen, J. Li, A.M. Cassel, Q. Ye, H.T. Ng, et al. Nanotechnology 14 (2003) 1239–1245.[18] http://www.aip.org/png/html/nanodisp.html.[19] M. Yang, Q. Zhang, J. Mater. Sci. 39 (11) (2004) 3777–3778.[20] R.H. Baughman, A.A. Zakhidov, W.A. Heer, Science 297 (2002) 787–792.[21] M. Endo, T. Hayashi, Y.A. Kim, M. Terrones, M.S. Dresselhaus, Philos. Trans. Royal Soc. A 362 (2004) 2223–2238.[22] S. Akita, Y. Nakayama, Jpn. J. Appl. Phys. 41 (Part 16B) (2002) 4242–4245.[23] J. Lee, S. Kim, Sens. Actuators A: Phys. 120 (1) (2005) 193–198.

P. Sharma, P. Ahuja / Materials Research Bulletin 43 (2008) 2517–25262526