Polymer Journal, Vol. 19, No. I, pp 147-156 (1987)

Electronic Devices from Conducting Organics and Polymers

Richard S. POTEMBER, Robert C. HOFFMAN, Henry S. HU, James E. COCCHIARO, Carla A. VIANDS, and Theodore 0. POEHLER

The Johns Hopkins University, Applied Physics Laboratory, Laurel, MD 20707, U.S.A.

(Received August 29, 1986)

ABSTRACT: In the past 20 years, organic molecular crystals and conducting organic polymers have been prepared in a variety of forms where electrical conductivity can be systematically controlled over a range of 10 orders of magnitude. Recently, a number of potential applications have emerged from this research. This paper provides a brief overview of current research into the possibility of using organic polymers and charge-transfer complexes to fabricate electronic and optical devices. KEY WORDS Molecular Electronics I Batteries I Conductive Polymers I Molecular Crystals I Polyacetylene I Polypyrrole I Optical Recording · Disk I Photochromics I Tetracyanoquinodimethane I Electrochromics I

The rapid development in the semiconduc• polymeric, or even biological materials has tor industry has been marked by the increasing recently been termed Molecular Electronics. density and complexity of semiconductor chip This new interdisciplinary field, combining circuitry. Considerable attention has been organic chemistry, solid-state physics, and given to the inherent problems associated microelectronic engineering, offers the poten• with increasing the density of integrated cir• tial of revolutionizing electronic and compu• cuitry. In the future, fundamental physical ter speed and capacity by making devices considerations will set limits on the size and from complex molecular structures. properties of integrated circuits that can be Organic compounds and polymers offer a manufactured by conventional techniques. viable alternative to the traditional inorganics Some of the limitations arise from process• in many applications because of their ex• ing considerations, materials properties and tremely small size, abundance, diversity, ease operational constraints. of fabrication, and potential low cost. For Although semiconductor technology has not example, the typical dimensions of organic yet reached its full potential, future gener• molecular structures (1 0-100 Angstroms) are ations of computers may require revolu• two to three orders of magnitude smaller than tionary new technologies based on totally dif• existing and proposed devices developed by ferent types of materials to increase memory current state-of-the-art lithographic tech• and processing capabilities. One possible class niques. Organic semiconductors offer an ad• of materials which may find applications in the ditional feature in that it is 'possible to control electronics industry is carbon based (organic) the electronic and optical properties of an semiconductors and conductors. organic device by altering or tailoring the The replacement of inorganic semiconduc• organicmolecular structure before fabricating tors and metals by organic macromolecular, the device. In molecular electronic devices

147 R. S. POTEMBER eta/.

rently divided into two areas: (1) electro-active polymers sometimes called plastic metals or synmetals (synthetic metals); (2) molecular crystals which include monomeric compounds and organic charge-transfer complexes. Figure l shows the structural formula of several com• mon electroactive polymers ana monomeric TCNE TCNQ TCNQ (0-ipr)2 TNAP compounds discussed in this paper.

s1\ s CONDUCTING ORGANIC POLYMERS: AN OVERVIEW >=< F CN F I I The study of electronic transport in organic F F polymers is more than thirty years old. The :x: first organic polymers prepared were electri• NC CN \>=< __ / cally insulating with conductivities as low as 10- 14 (ohm-centimeters)- 1 . The insulating TTF TMTSF BEDT properties are the result of all of the electrons N02 in the polymer being localized in the hybrid• atom molecular orbital bonds i.e., the satu• rated carbon framework of the polymer. These insulators, which include polymers such as 0 poly(n-vinylcarbozole), or polyethylene, have TNF Poly (p-phenylene) (PPP) extremely large band gaps with energy as high as 10 electron volts required to excite electrons from the valence to the conduction band. OcOI Electrical applications of insulating organic Poly(p-phen•1lene sulfide) CH=CH2 (PPS) N . vinylcarbazole polymers are limited to insulating or support• ing materials where low weight and excellent Figure l. Structural formulas of organic donor and acceptor molecules. processing and mechanical properties are desirable. Semiconducting organic polymers, some• the electronic and/or optical properties are times called electroactive polymers, exhibit locked into the molecular structure instead conductivities from 10-9 to 100 (ohm• of produced by the processing technique. centimeters) -I. These polymers can be clas• This feature called molecular architecture, sified into three groups based on their chemi• or molecular engineering, may greatly sim• cal structure: (1) composite polymers which plify future manufacturing processes by re• combine graphite-like structures with metal ducing the number of fabrication steps, particles. The concentration of metal particles hence increasing density, speed, and pro• may often reach 85% by weight in the com• duction yield. posite polymers; (2) polymeric charge-transfer Current research in molecular electronics complexes in which a cationic species, often a centers on the design and fabrication of new metal ion or an anionic species such as the synthetic molecular materials, evaluating these iodide ion (13)- or an organic acceptor such as materials, and applying them to novel techni• TCNQ or TNF (trinitrofluorenone) interacts cal applications. Research in this field is cur- with the polymer main chain by altering the

148 Polymer J., Vol. 19, No. I, 1987 Electronic Devices from Conducting Organics and Polymers

Organic and Inorganic Organic polymers molecular crystals materials

TMTSF2PF 6 (superconducting)

Cu

Graphite Hg Pyropolymers Polyacetylene Bi Poly(p-phenylene) TTF-TCNQ Si Poly(ph&nylene sulphide) ZnO Polypyrrole Ge (doped)

Polythiophene Cu-TCNQ Si

I E 1Q-5 u Boron E: ..c Polyphthalocyanine Ag-TCNQ ..£ H20 > ·;;:... Polyacetylene (undoped) Iodine ·;:; 10-10 ZnO "0 c 0 (,) Polydiacetylene

Polyth iophene Cu-phthalocyanine Polypyrrole (undoped) Anthracene Nylon Poly(p-phenylene) (undoped) Poly(phenylene sulphide) (undoped) PVC Polystyrene Polyimide (Kapton) PTFE (Teflon) Diamond Si02 1Q-20

Figure 2. Compilation of electrical conductivities of organic polymers. molecular orbital arrangement through elec• ylene was first prepared some time ago, it was tron interactions, and; (3) conjugated poly• not until 1977 that this polyene was modified mers which show semiconducting properties by combining the carbon chain with iodine because the n-orbital overlap along the un• and other molecular acceptors to produce a saturated polymer chain allows conduction material with metallic conductivity. 2 electron to interact with neighboring n-orbitals The actual nature of the electrical con• on adjacent carbon atoms. ductivity in doped and undoped polyacetylene High electrical conductivity has been ob• is still highly speculative. However, several served in several conjugated polymer or poly• possible explanations have been developed to ene systems (Figure 2). The first and simplest explain the properties of these polymers. In an organic polymer to show high conductivity ideal infinite chain length of polyacetylene, the was "doped" polyacetylene (Figure 3). In the pi-electrons would be expected to form a half• "doped" form its conductivity is in excess of filled energy band leading to one-dimensional 200 (ohm-centimeters)- 1 . Although polyacet- metallic behavior. In reality, however, a finite

Polymer J., Vol. 19, No. I, 1987 149 R. S. POTEMBER et a/. H\ /H H\ /H c=c c=c -c/ \c_c/ \c_ -\ ;-\ ;- H H H H (a)

H H H H I I I I

HI HI HI HI (b) Figure 3. (a) cis-Forms in idealized polyacetylene chain. (b) trans-Forms in idealized polyacetylene chain.

H H

Figure 4. Chemical structure of conductive polypyrrole. length of (CH)x chains would be an intrinsic elasticity that make polymers unique semiconductor with an electron band gap of materials. approximately 2 electron volts because the One of the most scientifically and tech• alternating double bond-single bond structure nologically studied new ·conductive organic has two atoms per unit cell instead of one; polymers is polypyrrole (Figure 4). The highly hence a semiconductor band gap is formed stable, flexible films of polypyrrole produced through a Peierls metfll-to-semiconductor by the one-step electro-oxidation have room transition. temperature p-type conductivities ranging In the ten years since the discovery of the from 10 to 100 (ohm-centimeters)- 1 . No ad• first highly conductive organic polymer, work ditional dopants are required to produce elec• on numerous other conductive polyene sys• trical conductivity. 3 tems (Figure 1). including poly(p-phenylene) In addition to showing high electrical con• PPP, poly(p-phenylene sulfide) PPS, polypyr• ductivity, polypyrrole films can be repeatedly role PP, and poly(p-phenylene vinylene) PPV electrochemically driven or "switched" be• has proved that, as a class of materials, the tween a black highly conductive 100 (ohm• unsaturated conjugated organic polymers can centimeters) -l oxidized form and a yellow be in an electrically conducting form and still nonconducting neutral. The switching rate for maintain the low density, processibility and thin films is approximately one per second.4

!50 Polymer J., Vol. 19, No. I, 1987 Electronic Devices from Conducting Organics and Polymers

Electrolyte

Polymer electrode (cathode) Polymer separator Polymer electrode (anode)

Figure 5. Schematic of "plastic" battery.

APPLICATIONS OF MOLECULAR lyte (lithium perchlorate), and completely en• ELECTRONIC MATERIALS capsulated in a plastic casing a "plastic bat• tery" can be made5 (Figure 5). The two sheets Molecular electronic materials show great of polyacetylene or poly(p-phenylene) act as promise in a wide range of applications. Many both anode and cathode for the battery. of the uses will be in direct competition with These novel polymeric batteries have several existing inorganic materials, while the novel n.otable features. They are rechargeable, ten properties reported in many molecular crystals times lighter than a conventional lead-acid and polymers assure that entirely new appli• battery, and flexible enough to fit into a variety cations and technologies will emerge from this of design configurations. field. A partial list of applications currently under PHOTO VOLTAIC APPLICATIONS development includes capacitors, transistors, batteries, memory devices, chemical sensors, Several conductive polymers are being test• transducers, and plastic wiring. A complete ed for use as coatings on photovoltaic and description of each application and material is photoelectroactive semiconductors in solar en• lengthy; however, five novel areas of appli• ergy conversion. As an example, a film of cation are described briefly below to illustrate conductive polypyrrole used as an overcoat on the role these materials may play in future the activated semiconductor, cadmium sulfide, technologies. can enhance the evolution of oxygen at the electrode/electrolyte surface and increase the LIGHTWEIGHT RECHARGEABLE lifetime of the cell by a million fold. 6 BATTERIES In another example, a photovoltaic cell has been fabricated using polyacetylene as the Highly conductive organic polymers are actual active photoelectrode. In a Schottky being evaluated for electrode materials in barrier configuration, this solar cell delivered both disposable and rechargeable batteries. an open circuit voltage of 0.3 volts and a short When two sheets of polyacetylene or poly(p• circuit current of 40 microamperes per square phenylene) are separated by an insulating film centimeter under an illumination equivalent of polycarbonate saturated in an electro- to one sun. 7 Polyacetylene may be an attrac-

Polymer J., Vol. 19, No. I, 1987 151 R. S. POTEMBER et a/.

(a)

Glass

(b)

Electrode Polymeric film Electrode

D = Neutral donor o+ = Oxidized donor

Figure 6. (a) Cell structure of electrochromic display. (b) Coloring electrochromic mechanism in polymeric film. tive solar cell material because studies suggest 6a. It consists of a transparent glass electrode, it is a direct band gap semiconductor with an the electrochromic polymer system and a thin optical band gap matching the solar spectrum. gold semitransparent counterelectrode. The electronic structure is analogous to tradi• When the structure is electrically biased tional inorganic semiconductors in that the (1.0-3.5 volts), the optical density of the film strong n-electron overlap along the polymer changes, i.e., the color changes, because the chain leads to large energy band widths. organic complex undergoes a reversible elec• tron transfer reaction from a neutral form to ELECTROCHROMIC POLYMER an oxidized form (Figure 6b). Using a TTF DISPLAYS redox compound, the color of the display changes from yellow to a bright red, while An electrochromic display is a thin solid pyrazoline switches from transparent to yel• state device that changes color reversibly when low. Multicolor displays are also being studied subjected to a small electrical potential. This which use combinations of organic com• new display technology has several advantages pounds that exhibit multiple coloration at over liquid crystal displays currently in use. different applied potentials. Electrochromic displays have low power con• sumption, good optical contrast, a wide view• AN ERASABLE COMPACT DISK ing angle, and an all solid state construction. In a unique application of charge-transfer The interaction of laser radiation with mat• organic materials, electrochromism has been ter for use in optical storage systems has reported in a thin film system consisting of a received considerable attention. A variety of polymer, polymethacrylonitrile (PMCN), and different media, including photographic films, organic redox materials such as TTF or pyra• photoresists, photosensitive polymers, photo• zoline.8 An electrochromic display cell struc• chromics, thin amorphous films, and elec• ture using these materials is shown in Figure trooptic materials have been proposed as opti-

152 Polymer J., Vol. 19, No. I, 1987 Electronic Devices from Conducting Organics and Polymers cal storage systems. Present optical storage plex phase (right side of equation) of the salt. systems have been applied in the fields of Various diagnostic techniques including in• document storage, audio and video reproduc• frared, Auger, X-ray photoelectron, and tion, and direct data collection. However, opti• Raman spectroscopy have been applied to help cal storage systems have not seen wide-spread understand this reaction. These experiments applications in computer technology because suggest that the effect of the applied electric most optical media are not erasable. field (laser) on an organometallic charge• We have recently developed an erasable transfer salt (e.g., copper TCNQ) is to induce a optical recording medium using the electric phase transition to a non-stoichiometric com• field induced optical switching effect in films plex salt containing neutral acceptor molecules of either copper or silver complexed with (TCNQ0 ). the electron acceptors tetracyanoethylene Figure 7 shows a cross section of a typical (TCNE), tetracyanonaphthoquinodimethane erasable optical disk medium using the TCNQ (TNAP), tetracyanoquinodimethane (TCNQ), family of organometallic materials. An ap• or other derivatives of TCNQ (Figure 1). 9 - 15 proximately 2000 Angstrom-thick organic The optical change in these materials is re• charge-transfer salt is deposited on aluminum, versible and fast, with switching times of less glass or polycarbonate supporting base using a than lO nanoseconds observed in static switch• solid-state diffusion process. This solid-state ing experiments. diffusion/deposition guarantees that the or• The observed changes in the optical proper• ganic film will be uniform over a wide area and ties of films fabricated from these organic that the size of the individual crystallites (less materials is a direct consequence of the laser than 1 micron) will be minimized. The disk is induced reaction shown in eq l. completed when a thin protective polymer overcoat is cast over the photoactive organic n[M+(TCNQ 7 )] xMo heat layer. The disk can then be placed in an optical recording, playback and erasure system. A where M can be silver, copper, sodium, po• block diagram of a typical laser diode optical tassium, or lithium. recording system is shown in Figure 8. The reaction shown in eq 1 is reversible and Information is recorded, read and erased on the thermodynamically favored initial phase the spinning disk by modulating the intensity can be readily reformed by heating the com- of the laser. To record information, the in-

Transparent protective coating

Supporting base material

Figure 7. Schematic of optical storage disk using organic charge-transfer complexes.

Polymer J., Vol. 19, No. I, 1987 153 R. S. PoTEMBER eta/.

Optical disk

A./4 coat plate

GaAIAs diode laser U

Figure 8. Block diagram of typical optical recording system. tensity of the modulated laser is increased above the power threshold of the particular organic charge-transfer complex, thus produc• ing a series of high contrast spots on the surface of the rotating recording medium. The reading of stored information is accomplished by illuminating each bit location with a con• tinuous reduced power beam (below write threshold) and monitoring the reflected in• tensity with a suitable photodetector which converts the light into electrical impulses. The information can subsequently be erased by Disk adjusting the power level and pulse duration to cause localized heating on the media. Figure 9. Magnified surface of AgTCNQ optical disk An example of optical information storage showing high contrast between optically switched and using this organic medium is shown in Figure unswitched regions for various laser intensities. 9. The figure shows the magnified surface of a AgTCNQ optical disk. Note the high contrast complexes with the 458, 488 and 514-nano• between the optically switched regions (spots) meter lines from an argon ion laser, the 532 and the background. The spots were made and 1064-nanometer lines from a Nd: YAG with the 532 nanometer line from a pulsed laser, the 633 nanometer line from a He: Ne Nd: YAG laser; laser energy was about 5 laser, the 780 nanometer line from a GaAlAs microjoules per pulse. Spot size was approx• laser and the I 0600 nanometer line from a imately 20 microns. carbon dioxide laser. In the majority of ma• We have irradiated a number of other cop• terials studied, the write threshold power is per and silver TCNQ type charge-transfer relatively independent of wavelength throug-

!54 Polymer J., Vol. 19, No. I, 1987 Electronic Devices from Conducting Organics and Polymers hout the visible and infrared region of the spectrum. The threshold writing power varies between 3 and 150 milliwatts depending upon the specific choice of metal and acceptor com• plex. These power levels are well within the capabilities of most moderately powered com• mercial laser systems.

PHOTOCHROMIC PROPERTIES OF METAL-ORGANIC CHARGE• Wavelength (nm) TRANSFER COMPLEXES Figure lOa. Typical CuTCNQ transmission switch- 30 millijoule pulses from frequency doubled Nd: Y AG An area of technological interest is the laser (532 nanometers). change in the macroscopic optical properties of these organometallic materials in an elec• tromagnetic field. The potential for large changes in transmission and reflection in CuTCNQ and other members of this family make these compounds prime candidates as high speed photochromic filters. 14•16 We have observed that these films undergo a change from their respective blue and violet colors to a rather pale yellow color characteristic of neu• tral TCNQ. As shown in Figure lOa, a typical Wavelength (nm) CuTCNQ film formed by solid state diffusion Figure lOb. Typical AgTCNQ transmission switch- is rather poorly transmitting from the mid• 45 millijoule pulses from frequency doubled Nd: Y AG visible into the near infrared in the vicinity of laser (532 nanometers). 1100 nanometers, and there is a substantial increase in transmission extending into the comes strongly transmitting in a band center• infrared. Upon irradiating the CuTCNQ film ed at 500 nanometers, while the unswitched with a Nd: Y AG laser at 532 nanometers, film is poorly transmitting in the same re• one observes a significant increase in trans• gion. mission throughout the spectrum ranging from the midvisible to the infrared. Dramatic CONCLUDING REMARKS increases are observed in the red end of the spectrum and in the near· infrared, those re• It may be premature to evaluate the tech• gions which are of particular interest with re• nological significance of this new field of sci• spect to many of the principal laser sources. ence to the electronics industry. However, this The transmission properties of an AgTCNQ paper provides concrete examples of organic film are shown in Figure lOb. In contrast to the macromolecules and polymers with specific changes observed in CuTCNQ, the switched chemical and physical properties useful in de• and unswitched films are quite similar in the vice applications such as lightweight recharge• infrared portion of the spectrum. Large able batteries, photovoltaic solar cells, junc• changes in transmission are noted in the tion and electro-optic devices, photochromic visible part of the spectrum between 400 and filters, and information storage and processing 600 nanometers. The switched AgTCNQ be- media.

Polymer J., Vol. 19, No. I, 1987 !55 R. S. POTEMBER et a/.

In each of the various devices and appli• Soc., 106, 2294 (1984). cations discussed in this paper, the novel con• 5. P. J. Nigrey, D. Macinnes, D. P. Nairns, and A. G. cept of molecular architecture is demonstrated MacDiarmid, J. Electrochem. Soc., 128, 1651 (1981); R. Somano, Appl. Phys. Commun., l, 179 (1981-82). i.e., the precise design and synthesis of chemi• 6. R. A. Bull, F. R. Fan, and A. J. Bard, J. E/ectrochem. cal structure controls the specific physical Soc., 130, 1636 (1983); A. J. Frank and K. J. Honda, properties observed in a par!icular device J. Phys. Chern., 86, 1933 (1982). application. Current research in this field 7. S. N. Chen, A. J. Heeger, Z. Kiss, A. G. Mac• Diarmid, S. C. Gau, and D. L. Peebles, App/. Phys. has also established a foundation of mate• Lett., 36, 96 (1980). rials, and fundamental scientific principles 8. Y. Hirai and C. Tani, Appl. Phys. Lett., 43, 704 amenable to the design of new molecular elec• (1983). tronic devices. 9. R. S. Potember, T. 0. Poehler, and D. 0. Cowan, Appl. Phys. Lett., 34, 405 (1979). 10. R. S. Potember, T. 0. Poehler, A. Rappa, D. 0. Acknowledgments. Work at the Applied Cowan, and A. N. Bloch, J. Am. Chern. Soc., 102, Physics Laboratory was supported in part by 3659 (1980). DARPA/AFOSR, Wright-Patterson Air II. R. S. Potember, T. 0. Poehler, D. 0. Cowan, A. N. Bloch, P. Brant, and F. L. Carter, Chern. Scripta, 17, Force Base, and U. S. Naval Air Systems 219 (1981). Command under contract No. N00024-85-C- 12. R. S. Potember, T. 0. Poehler, D. 0. Cowan, and A. 530l. N. Bloch, Proceedings of the NATO Conference on Chemistry and Physics of One-Dimensional Materials, L. Alcacer, Ed., Reidel, Boston, 1980, p REFERENCES 419. 13. R. S. Potember, T. 0. Poehler, and R. C. Benson, l. Y. W. Park, A. J. Heeger, M. A. Druy, and A. G. Appl. Phys. Lett., 41, 548 (1982). MacDiarmid, J. Chern. Phys., 73, 946 (1980). 14. R. S. Potember, R. C. Hoffman, R. C. Benson, and 2. C. K. Chiang, C. R. Fincher, Jr., Y. W. Park, A. J. T. 0. Poehler, J. De Physique, 44, C3-1597 (1983). Heeger, H. Shirakawa, E. J. Louis, S.C. Gau, and A. 15. R. C. Benson, R. C. Hoffman, R. S. Potember, E. G. MacDiarmid, Phys. Rev. Lett., 39, 1098 (1977). Bourkoff, and T. 0. Poehler, Appl. Phys. Lett., 42, 3. K. K. Kanazawa, A. F. Diaz, W. D. Gill, P. M. 855 (1983). Grant, G. B. Street, G. P. Gardini, and J. F. Kwak, 16. T. Hirono, M. Fukuma, and T. Yamada, J. Appl. Synthetic Metals, l, 329 (1980). Phys., 57, 2267 (1985). 4. P. G. Pickup and R. A. Osteryoung, J. Am. Chern.

156 Polymer J., Vol. 19, No. I, 1987