CA1295398C - Electroluminescent device with modified thin film luminescent zone - Google Patents

Abstract

ELECTROLUMINESCENT DEVICE WITH
MODIFIED THIN FILM LUMINESCENT ZONE
Abstract of the Disclosure An electroluminescent device is disclosed having a luminescent zone of less than one µm in thickness comprised of an organic host material capa-ble of sustaining hole-electron recombination and a fluorescent material capable of emitting light in response to energy released by hole-electron recombi-nation.

Description

ELECTROLUMINESCENT DEVICE WITH
MODIFIED THIN FILM LUMINESCENT ZONE
Fi eld of the Invention This invention relates to organic electro-luminescent devices. More specifically, this inven-tion relates to devices which emit light from an organic layer positioned between anode and cathode electrodes when a voltage is applied across the elec-trodes.
Background Qf the Invention While organic electroluminescent devices have been known for about two decades, their perfor-mance limitations have represented a barrier to many desirable applications.
Gurnee et al U.S. Patent 3,172,862, issued March 9, 1965, filed Sept. 29, 1960, disclosed an organic electroluminescent device. (For brevity EL, the common acronym for electroluminescent, is some-times substituted.) The EL device was formed of an emitting layer positioned in conductive contact with a transparent electrode and a metal electrode. The emitting layer was formed of a conjugated organic host material, a conjugated organic activating agent having condensed benzene rings, and a finely divided Z5 conductive material. Naphthalene, anthracene, phen~
anthrene, pyrene, benzopyrene, chrysene, picene, car-bazole, fluorene, biphenyl, terphenyls, quaterphen-yls, triphenylene oxide, dihalobiphenyl, ~
stilbene, and 1,4-diphenylbutadiene were offered as examples of organic host materials. Anthracene, tet-racene, and pentacene were named as examples of acti-vating agents, with anthracene being disclosed to impart a green hue and pentacene to impart a red hue. Chrome and brass were disclosed as examples of the metal electrode while the transparent electrode was disclosed to be a conductive glass. The emitting layer was disclosed to be "as thin as possible, about ~' t'`

~3~13 0.0001 inch" - i.e., 2.54 micrometers. Electrolumi-nescence was reported at 800 volts and 2000 hertz.
Recognizing the disadvantage of employing high voltages and frequencies Gurnee U.S. Patent 3,173,050 reported electroluminescence at 110 volts d.c. by employing in series with the emitting layer an impedance layer capable of accounting for 5 to 50 percent of the voltage drop across the electrodes.
Until relatively recently the art has reported at best modest performance improvements over Gurnee while resorting to increasingly challenging device constructions, such as those requiring alkali metal cathodes, inert atmospheres, relatively thick monocrystalline anthracene phosphor elements, and/or gpecialized device geometries. Mehl U.S. Patent 3,382,394, Mehl et al U.S. Patent 3,530,325, Roth U.S. Patent 3,359,445, Williams et al U.S. Patent 3,621,321, Williams U.S. Patent 3,772,556, Kawabe et al "Electroluminescence of Green Light Region in Doped Anthracene", l~E~Ln Journal Q~ A~ Q~ Physics, Vol. 10, pp. 527-528, 1971, and Partridge U.S. Patent 3,995,299 are representative.
In 1969 Dresner, "Double Injection Electro-luminescence in Anthracene", ~ Revi~w, Vol. 30, pp.
322-334, independently corroborated the performance levels of then state of the art EL devices emp~oying thick anthracene phosphor elements, alkali metal cathodes, and inert atmospheres to protect the alkali metal from spontaneous oxidation. These EL devices were more than 30 ~m in thickness and required operating potentials of more than 300 volts. In attempting to reduce phosphor layer thickness and thereby achieve operation with potential levels below 50 volts Dresner attempted to coat anthracene powder between a conductive glass anode and a gold, platinum or tellurium grid cathode, but phosphor layer thick-nesses of less than 10 ~m could not be successfully achieved because of pinholes.
Dresner U.S. Patent 3,710,167 reported a more promising EL device employing like Gurnee et al and Gurnee a conjugated organic compound, but as the sole component of an emitting layer of less than 10 ~m (preferably 1 to 5 ~m) in thickness. A tunnel injection cathode consisting of aluminum or degener-ate N+ silicon with a layer of the corresponding aluminum or silicon oxide of less than 10 Angstroms in thickness was employed.
The most recent discoveries in the art of organic EL device construction have resulted from EL
device constructions with two extremely thin layers (<1.0 ~m in combined thickness) separating the anode and cathode, one specifically chosen to trans-port holes and the other specifically chosen to transport electrons and acting as the organic lumin-escent zone of the device. This has allowed applied voltages to be reduced for the first time into ranges approaching compatibility with integrated circuit driver~, such as field effect transistors. At the same time light outputs at these low driving voltages have been sufficient to permit observation under com-mon ambient lighting conditions.
For example, Tang U.S. Patent 4,356,429 dis-closes in Example 1 an EL device formed of a conduc-tive glass transparent anode, a 1000 Angstroms hole transporting layer of copper phthalocyanine, a 1000 Angstroms electron transporting layer of tetraphenyl-butadiene in poly(styrene) also acting as the lumi-nescent zone of the device, and a silver cathode.
The EL device emitted blue light when biased at 20 volts at an average current density in the range of 30 to 40 mA/cm . The brightness of the device was 5 cd/m2. Tang teaches useful cathodes to be those formed from ~53~8 common metals with a low work function, such as indium, silver, tin, and aluminum.
A further ~mprovement in organic layer EL
devices is taught by Van Slyke et al U.S. PHtent 4,539,507. Referring to Example 1, onto a transpar-ent conductive glass anode were vacuum vapor deposited successive 750 Angstrom hole transporting 1,1-bis(4-di-~-tolylaminophenyl)cyclohexane and elec-tron transporting 4,4'-bis(5,7-di-_-pentyl-2-benzoxa-zolyl)stilbene layers, the latter also providing theluminescent zone of the device. Indium was employed AS the cathode. The EL device emitted blue-green light (520 nm peak). The maximum brightness achieved 340 cd/m at a current density of about 140 mA/cm2 when the applied voltage was 22 volts. The maximum power conversion efficiency was about 1.4 X 10 3 watt/watt, and the maximum EL quantum efficiency was about 1.2 X 10 photon/electron when driven at 20 volts. Silver, tin, lead, magnesium, manganese, and aluminum are specifically mentioned for cathode con-struction.
Summary of the Invention The discovery of techniques for fabricating organic EL devices with thin (<1 ~m) luminescent zones has lndicated the potential for more widespread use. Light output ls directly related to current density which is in turn A function of the field gradient (volts/cm) between the electrodes. When thicker organlc layers are employed, higher voltages must be employed to achieve a field gradient consis-tent with acceptable light output level - e.g., light emission 4ufficient to be readily detected in ambient room light. With organic EL devices having thin luminescent zones acceptable luminescence can be achieved at voltage levels conveniently provided by integrated circuits.

~,f~39~3 Although thin film organic EL devices have offered promise, there have remained significant bar-riers to their broader use~ One ~ignificant concern has been that available organic material~ for forming thin luminescent zones have provided a limited choice of light emission wavelengths. For example, it can be readily ~ppreciated that the formation of a full multicolor display requires at least a triad of indi-vidual organic EL devices, one device for emission in a different one of each of the blue, green, and red portions of the spectrum. Further, given the availa-bility of a primary hue of emission, such as blue or green, the desirability of more precisely choosing the hue of emission arises.
In addition to the matter of the hue of light emission by the thin film organic EL devices, the stability of the devices has remained a matter of concern. Most practical applications require llmited voltage input or light output variance over an extended period of time. While the ~romatic tertiary amine layers employed by Van Slyke et al V.S. Patent 4,539,507, cited above, have resulted in highly attractive initial light outputs in organic EL
device~, the Limited stability of thin film organic EL devices containing these layers has remained a deterrent to widespread use. Device degradation results in obtaining progressively lower current den-sitles when a constant voltage is applied. Lower current densities in turn result in lower levels of light output. With a constant applied voltage, prac-tical EL device use terminates when light emission levels drop below acceptable levels. If the applied voltage is progressively increased to hold light emission levels constant, the field across the EL
device is correspondingly increased. Eventually a voltage level is required that cannot be conveniently supplied by the EL device driving circuitry or which ~2 ~ ~ 3~

produces a field gradient (volts/cm) exceeding the dielectric breakdown strength of the layers separat-ing the electrodes, resulting in a catastrophic failure of the EL device.
The present invention offers to the art thin film organic EL devices having a broader range of light output wavelengths and higher stability levels than have heretofore been realized.
In one aspect this invention is directed to an electroluminescent device comprising in sequence, an anode, an organic hole injecting and transporting zone, a luminescent zone, and a cathode. The EL
device is characterized in that the luminescent zone is formed by a thin film of less than 1 ~m in thickness comprised of an organic host material capa-ble of sustaining hole and electron injection and a fluorescent material capable of emitting light in response to hole-electron recombination.
The presence of the fluorescent material permits a choice from among a wide ~atitude of wave-lengths of light emission. By selection of the mate-rials forming the thin film organic EL devices of this invention, including particularly any one or combination of the fluorescent materials, the cathode metals, and the hole injecting and transporting mate-rials, more stable device operation can be achieved than has been heretofore realized.
Brief Description of the Draw~lg~
These and other advantages of the present invention can be better appreciated by reference to the following detailed description considered in con-junction with the drawings, in which Figures 1, 2, and 3 are schematic diagrams of EL devices.
The drawings are necessarily of a schematic nature, since the thicknesses of the individual lay-ers are too thin ard thickness differences of the d ~

various device element~ too great to permit depiction to scale or to permit proportionate scaling.
DescriPtion of Preferred Embodiments An electroluminescent or EL device 100 according to the invention is schematically illus-trated in Figure 1. Anode 102 i8 separated from cathode 104 by an organic luminescent medium 106.
The anode and the cathode are connected to an exter-nal power source 108 by conductors 110 and 112, respectively. The power source can be a continuous direct current or alternating current voltage ~ource or an intermittent current voltage source. Any con-venient conventional power source, including any desired switching circuitry, can be employed which is capable of positlvely biaslng the anode with respect to the cathode. Either the anode or cathode can be at ground potential.
The EL device can be viewed as a diode which is forward biased when the anode is at a higher potential than the cathode. Under these conditions the anode in~ects holes (positive charge carriers), schematically shown at 114, into the luminescent medium while the cathode in~ects electrons, schemati-cally shown at 116, into the luminescent medium. The portion of the luminescent medium adJacent the anode thus forms a hole in~ecting and transporting zone while the portlon of the lumlnescent medium ad~acent the ca,thode forms an electron in~ecting and trans-porting zone. The in~ected holes and electrons each migrate toward the oppositely charged electrode.
This re~ults in hole-electron recombination within the organic luminescent medium. When a migrating electron drops from its conduction potential to a valence band in fill~ng a hole, energy is released as light. Hence the organic luminescent medium form between the electrodes a luminescence zone receiving mobile charge carriers from each electrode. Depend- \
\

~ 9~

ing upon the choice of alternative constructions, the released light can be emitted from the luminescent material through one or more of edges 118 ~eparating the electrodes, through the anode, through the cath-ode, or through any combination of the foregoing.
Reverse biasing of the electrodes rever~es the direction of mobile charge migration, interrupts charge in~ection, and terminates light emission. The most common mode of operating organic EL devices is to employ a forward biasing d.c. power source and to rely on external current interruption or modulation to regulate light emission.
In the organic EL devices of the invention it is possible to maintaln a current density compati-ble with efficient light emission while employing arelatively low voltage across the electrodes by lim-iting the total thickness of the organic luminescent medium to less than 1 ~m (10,000 Angstroms). At a thickness of less than 1 ~m an applied voltage of 20 volts results in a field potential of greater than 2 X 10 volts/cm, which is compatible with efficient light emission. As more specifically noted below, preferred thlcknesses of the organic luminescent medium are in the range of from 0.1 to 0.5 ~m (1,000 to 5,000 Angstroms), allowing further reductions in applied voltage and/or increase in the field poten-tial, are well within device construction capabili-ties.
Since the organic luminescent medium is qulte thln, lt ls usually preferred to emlt light through one of the two electrodes. This is achieved by forming the electrode as a translucent or trans-parent coating, either on the organic luminescent medium or on a separate translucent or transparent support. The thickness of the coating is determined by balancing light transmission (or extinction) and electrical conductance (or resistance). A practical ~ ~3 ~ 3~

balance in forming a light transmissive metallic electrode is typically for the conductive coating to be in the thickness range of from about 50 to 250 Angstromq. Where the electrode is not intended to transmit light, any gre&ter thicknes~ found conve-nient in fabrication can also be employed.
Organic EL device 200 shown in Figure 2 is illu~trative of one preferred embodiment of the invention. Because of the historical development of organic EL devices it is customary to employ a trans-parent anode. This has been achieved by providing a transparent insulative support 201 onto which is deposited a conductive relatively high work function metal or metal oxide transparent layer to form anode 203. Slnce the portion of the organic luminescent medium immediately ad~acent the anode acts as a hole tranqporting zone, the organic luminescent medium is preferably formed by depositing on the anode a layer 205 of an organic material chosen for its hole trans-porting efficiency. In the orientation of the device200 shown, the portion of the organic luminescent medium ad~acent its upper surface constitutes an electron transporting zone and is formed of a layer 207 of an organlc material chosen for its electron transporting efficiency. With preferred choices of materials, described below, forming the layers 205 and 207, the latter also forms the zone in which luminescence occurs. The cathode 209 is conveniently formed by deposition on the upper layer of the organic luminescent medium.
Organic EL device 300 shown in Figure 3 is illustrative of another preferred embodiment of the invention. Contrary to the historical pattern of organic EL device development, light emission from the device 300 is through the light transmissive (e.g., transparent or substantially transparent) cathode 309. While the anode of the device 300 can ~2~5~

be formed identically as the device 200, thereby per-mitting light emission through both anode and cath-ode, in the preferred form shown the device 300 employs an opaque charge conducting element to form the anode 301, such as a relatively high work func-tion metallic substrate. The hole and electron transporting layers 305 and 307 can be identical to the corresponding layers 205 and 207 of the device 200 and require no further description. The signifi-cant difference between devices 200 and 300 is thatthe latter employs a thin, light transmissive (e.g., transparent or substantially transparent) cathode in place of the opaque cathode customarily included in organic EL devices.
Viewing organic EL devices 200 and 300 together, it i9 apparent that the present invention offers the option of mounting the devices on either a positive or negative polarity opaque substrate.
While the organic luminescent medium of the EL
devices 200 ~nd 300 are described above as being com-prised of a single organic hole in~ecting and trans-porting layer and a single electron in~ecting and transporting layer, further elaboration of each of these layerq into multiple layers, as more specifi-cally descrlbed below, can result in further enhance-ment of device performance. When multlple electron inJecting and transporting layers are present, the layer receiving hole~ is the layer in which hole=
electron recombination occurs and therefore forms the luminescent zone of the device.
In the practice of the present invention the luminescent zone is in every instance formed by a thin film (herein employed to mean less than 1 ~m in thickness) compri~ed of an organic host material capable of ~ustaining hole and electron inJection and a fluorescent material capable of emitting light in response to hole-electron recombination. It is pre-3~3 ferred that the luminescent zone be maintained in athickness range of from 50 ~o 5000 Angstroms and, optimally, lO0 to lO00 Angstroms, so that the entire organic luminescent medium can be less thsn l ~m and preferably less than lO00 Angstroms in thickness.
The host material can be conveniently formed of any material heretofore employed as the active component o$ a thin film luminescent zone of an organic EL device. Among ho t materials suitable for use in forming thin films are diarylbutadienes and ctilbenes, such as those disclo-~ed by Tang U.S. Pat-ent 4,356,429, cited above.
Still other thin film forming host materials which can be employed are optical brighteners, par-ticularly those disclosed by Van Slyke et al U.S.
Patent 4,539,507, cited above. U~eful optical brighteners include those satisfying structural for-mulae (I) and (II):

20 (I) DZ ~ ~ D4 or (II) D3 ~ D5 ~ / \z~

wherein Dl, D , D3, and D are individually hydrogen; saturated aliphatic of from l to 10 cArbon atoms, for example, propyl, t-butyl, heptyl, and the like; aryl of from 6 to lO carbon atoms, for example, phenyl and naphthyl; or halo such a~ chloro, fluoro, and the like; or D and D or D and D4 taken together comprise the atoms necessary to complete a fused aromatic ring optionally bearing at least one P ~ 9 saturated ~liphatic of from 1 to 10 carbon ~toms, such as methyl, ethyl, propyl and the like;
D is a ~aturated ~liphatic of from 1 to 20 csrbon atoms, such as methyl, ethyl, n-eicosyl, ~nd the like; aryl of from 6 to 10 carbon atoms, for example, phenyl and naphthyl; carboxyl; hydrogen;
cyano; or halo, for example, chloro, fluoro and the like; provided that in formula (II) at least two of D , D4 and D are saturated aliphatic of from 3 to 10 carbon atoms, e.g., propyl, butyl, heptyl and the like;
Z is -O-, -N(D )--, or -S-; and Y is D ~CH CH ~ D , ¦ \ =./ ¦ , ~CH=CH ~D7tCH=CH ~. or ~' ~ ~-wherein m is an integer of from 0 to 4;
n i~ arylene of from 6 to 10 carbon atoms, for example, phenylene and naphthylene; and D is hydrogen; a s~turated Aliphatic 9ub-stituent of from 1 to 10 carbon atoms, such as an alkyl substituent; aryl of from 6 to 10 carbon atoms, such as phenyl or naphthyl; or a halo substituent, such as chloro or fluoro;
D i~ arylene of from 6 to 10 carbon atoms, such as phenyl or naphthyl;
Z' and Z" Are individually N or CH.
As used herein "aliphatic" includes substituted ali-phatic as well as unsubstituted aliphatic. The sub-stituents in the case of substituted aliphatic include alkyl of from 1 to 5 carbon atoms, for exam-ple, methyl, ethyl, propyl and the like; aryl of from 6 to 10 carbon atoms, for example, phenyl and naph-thyl; halo, such as chloro, fluoro and the like;
nitro; and alkoxy having 1 to 5 carbon stoms, for ~ 2~3913 example, methoxy, ethoxy, propoxy, and the like.
Still other optical brighteners that are contemplated to be useful are listed in Vol. 5 of ChemistrY of S m thetic DYes, 1971, pages 618-~37 and 640. Those that are not already thin-film-forming can be rendered so by attaching an aliphatic moiety to one or both end rings.
Particularly preferred host materials for forming the luminescent zone of the organic EL
devices of this inventions are metal chelated oxinoid compounds, including chelates of oxine (also commonly referred to as 8-quinolinol or 8-hydroxyquinoline).
Such compounds exhibit both high levels of perfor-mance and are readily fabricated in the form of thin films. Exemplary of contemplated oxinoid compounds are those satisfying structural formula (III):
(III) \ M / In t+n ~ \ N

\ M
t+n whereln -Mt represents a metal;
n is an integer of from 1 to 3; and 129~39 Z independently in each occurrence represents the atoms completing a nucleus having at least two fused aromatic rings.
From the foregoing it is app~rent that the metal can be monovalent, divalent, or trivalent metal. The metal can, for example, be an alkali metal, such as lithium, sodium, or potassium; an alkaline earth metal, such a-~ magne~ium or calcium;
or an earth metal, such as boron or aluminum. Gener-ally any monovalent, divalent, or trivalent metalknown to be a useful chelating metfll can be employed.
Z completes fl heterocyclic nucleus con-taining at least two fused aromatic rings, at one of which is an azole or azine ring. Additional rings, including both aliphfltic and aromatic rings, can be fused with the two required rings, lf required. To avoid adding molecular bulk without improving on function the number of ring atoms is preferably maln-tained at 18 or less.
Illustrative of useful ho~t materials capa-ble of being used to form thin films are the follow-ing:
HM-l AluMinum trisoxine [a.k.a., tris(8-quinolinol) aluminum]
HM-2 Magnesium bisoxlne [a.k.a., bis(8-quinolinol) magnesium]
HM-3 ~is~benzo{f)-8-quinolinol] zinc HM-4 Bls(2-methyl-8-quinolinolato) aluminum oxide HM-5 Indium tri~oxine [~.k.a., tri~(8-quinolinol) indium]
HM-6 Aluminum tris(5-methyloxine) [a.k.a., tris(5-methyl-8-quinolinol) alumi-num HM-7 Lithium oxine [a.k.a., 8-quinolinol lithium]
HM-8 Gallium trisoxine [a.k.a, tris(5-chloro-8-quinolinol) gallium]

~Z~i3~?1 3 HM-9 Calcium bis(5-chlorooxine) [a.k.a, bis(5-chloro-8-quinolinol) calcium]
HM-10 Poly[zinc (II)-bis(8-hydroxy-5-quinolinyl)-methane]
S HM-11 Dilithium epindolidione HM-12 1,4-Diphenylbutadiene HM-13 1,1,4,4-Tetraphenylbutadiene HM-14 4,4'-Bis[5,7-di(t-pentyl-2-benzoxazolyl~-stilbene HM-15 2,5-Bis[5,7-di(t-pentyl-2-benzoxazolyl]-thiophene HM-16 2,2'-(1,4-phenylenedivinylene)bisbenzothia-zole HM-17 4,4'-(2,2'-Bisthlazolyl)biphenyl HM-18 2,5-Bis[5-(a,u-dlmethylbenzyl)-2-ben-zox3zolyl]thiophene HM-19 2,5-Bis[5,7-di(t-pentyl)-2-benzoxazolyl]-3,4-diphenylthiophene HM-20 Trans-stilbene All of the host materials listed above are known to emit light in response to hole and electron in~ection. By blending with the host material a minor amount of a fluorescent material capable of emitting light in response to hole-electron recombi-nation the hue of light emitted from the lumlnescent zone can be modified. In theory, lf a host material and a fluorescent material could be found for blend-ing which have exactly the same affinity for hole=
electron recombination each material should emit light upon in~ection of holes and electrons in the luminescent zone. The perceived hue of light emis-sion would be the visual integration of both emis-sions.
Since imposing such a balance of host and fluorescent materials is highly limiting, it is pre-ferred to choose the fluorescent material so that it provides the favored sites for light emission. When ~2caS~

only a small proportion of fluorescent material pro-viding favored sites for light emission is present, peak intensity wavelength emissions typical of the host material can be entirely eliminated in favor of a new peak intensity wavelength emission ~ttributable to the fluore~cent material. While the minimum pro-portion of fluorescent material sufficient to achieve this effect varies by the specific choice of host and fluorescent materials, in no instance is it necessary to employ more than about 10 mole percent fluorescent material, based on moles of host material and seldom is it necessary to employ more than 1 mole percent of the fluorescent material. On the other hand, for any host materlal capable of emitting light in the absence of fluorescent material, limiting the fluo-rescent material present to extremely small amounts, typically less than about 10 3 mole percent, based on host material, can result in retaining emission at wavelengths characteristic of the host material.
Thus, by choosing the proportion of a fluorescent material capable of providing favored sites for light emission, either a full or partial shifting of emis-sion wavelengths can be realized. This allows the spectral emissions of the EL devices of thi~ inven-tion to be seLected and balanced to suit the ~ppLica-tion to be served.
Choosing fluorescent materials capable of providing favored sites for light emission neces-sarily involves relating the properties of the fluo-rescent material to those of the host material. Thehost material can be viewed as a collector for in~ected holes and electrons with the fluorescent material providing the molecular sites for light emission. One important relationship for choosing a fluorescent material capable of modifylng the hue of light emission when present in a host material is a comparison of the reduction potentials of the two ~Z~398 materials. The fluorescent materials demonstrated to shift the wavelength of light emission have exhibited a less negative reduction potential than that of the host material. Reduction potentials, measured in electron volts, have been widely reported in the lit-erature along with varied techniques for their mea-surement. Since it is a comparison of reduction potentials rather than their absolute values which is desired, it is apparent that any accepted technique for reduction potential measurement can be employed, provided both the fluorescent and host material reduction potential~ are similarly measured. A pre-ferred oxidation and reduction potential measurement technique is reported by R. J. Cox, Photo~raPhic Sen-sltivitY, Academic Press, 1973, Chapter 15.
A second important relationship for choosinga fluorescent material capable of modifying the hue of light emission when present in a host material is a comparison of the bandgap potentials of the two materials. The fluorescent materials demonstrated to shift the wavelength of light emission have exhibited a lower bandgap potential than that of the host mate-rial. The bandgap potential of a molecule is taken as the potential difference in electron volts (eV) ~eparating its ground state ~nd flr~t singlet state.
Bandgap potentials and techniques for their measure-ment have been widely reported in the literature.
The bandgap potentials herein reported are those mea-~ured in electron volts (eV) at an absorption wave-length which i9 bathochromic to the absorption peakand of a magnitude one tenth that of the magnitude of the absorption peak. Since it is a comparison of bandgap potentials rather than their absolute values which is desired, it is apparent that any accepted technique for bandgap measurement can be employed, provided both the fluorescent and host material band gaps are ~imilarly measured. One illustrative mea-~ Z~39~

surement technique is disclosed by F. Gutman and L.
E. Lyons, Or~anic Semiconductors, Wiley, 1967, Chap-ter 5.
Where a host material is chosen which is itself cspable of emitting light in the absence of the fluorescent msterial, it has been observed that suppression of light emission at the wavelengths of emission characteristic of the host material alone and enhancement of emission at wavelengths character-istic of the fluorescent material occurs when spec-tral coupling of the host and fluorescent materials is achieved. By spectral coupling it is meant that an overlap exists between the wavelengths of emission characteristic of the host msterial alone and the wavelengths of light absorption of the fluorescent materisl in the sbsence o~ the host material. Opti-mal spectrsl coupling occurs when the msximum emis-sion of the host msterisl slone substantislly mstches within + 25 nm the maximum absorption of the fluore~-cent material alone. In practice sdvsntageous spec-tral coupling can occur with peak emission and absorption wavelengths differing by up to 100 nm or more, depending on the width of the peaks and their hypsochromic and bathochromlc ~lopes. Where less than optimum spectral coupling between the host snd fluorescent materlals is contemplated, a bathochromic as compared to a hypsochromic displscement of the fluorescent material produces more efficient results.
Although the foregoing discussion ha~ been undertaken by reference to host materials which are known to themselves emit light in response to hole snd electron in~ection, in fact light emission by the host material itself can entirely cease where light emission by the fluorescent material is favored by any one or combination of the various relationships noted above. It is appreciated that shifting the role of light emission to the fluorescent materisl 9~3 allows a still broader range of choices of host mate-rials. For exsmple, one fundamental requirement of a material chosen to emit light is that it must exhibit a low extinction coefficient for light of the wave-length it emits to avoid internal absorption. Thepresent invention permits use of host materials which are capable of sustaining the in~ection of holes and electrons, but are themselves incapable of effi-ciently emitting light.
Useful fluorescent materials are those capa-ble of being blended with the host material and fab-ricated into thin films satisfying the thickness ranges described above forming the luminescent zones of the EL devices of this invention. While crystal-line host materials do not lend themselves to thin film formation, the limited amounts of fluorescent materials present in the host materials permits the use of fluorescent materials which are alone incapa-ble of thin film formation. Preferred ~luorescent materials are those which form a common phase with the host material. Fluorescent dyes constitute a preferred class of fluorescent materials, since dyes lend themselves to molecular level distribution in the host material. Although any convenient technlque for dlspersing the fluorescent dyes ln the host mate-rials can be undertaken, preferred fluorescent dyes are those which can be vacuum vapor deposited along with the host materials. As5uming other criteria, noted above, are satisfied, fluorescent laser dyes are recognized to be particularly useful fluorescent materials for use in the organic EL devices of this invention.
One preferred class of fluorescent dyes are fluorescent coumarin dyes. Among specifically pre-ferred fluorescent coumarin dyes are those satisfyingformula IV:

(IV) R
R~O'~ I' R4 t~ ~

where Rl is chosen from the group consisting of hydro-gen, carboxy, alkanoyl, alkoxycarbonyl, cyano, aryl, and a heterocylic aromatic group, R2 is chosen from the group consisting of hydro-gen, alkyl, haloalkyl, carboxy, alkanoyl, and alkoxy-carbonyl, R is chosen from the group consisting of hydro-gen and alkyl, R4 is an amino group, and R5 is hydrogen, or Rl and R2 together form a fused carbocyclic ring, and/or the amino group forming R4 completes with at least one of R4 and R6 a fused ring.
The alkyl moieties in each instance contain from 1 to 5 carbon atoms, preferably 1 to 3 carbon atoms. The aryl moieties are preferably phenyl groups. The fused carbocyclic rings are preferably five, six, Ol seven membered rings. The heterocyclic aromatic groups contain 5 or 6 membered heterocyclic rings containing carbon atoms and one or two hetero atoms chosen from the group consisting of oxygen, sulfur, and nitrogen. The amino group can be a pri-mary, secondary, or tertiary amino group. When theamino nitrogen completes a fused ring with an adja-cent substituent, the ring is preferably a five- or six-membered ring. For examplej R4 can take the form of a pyran ring when the nitrogen atom forms a single ring with one adjacent substituent (R3 or R5) or a julolidine ring (including the fused benzo , ring of the coumarin) when the nitrogen atom forms rings with both ad~acent subst~tuents R3 and R .
The following fire illustrative fluorescent coumarin dyes known to be useful AS la~er dye~:
FD-1 7-Diethylamino-4-methylcoumarin FD-2 4,6-Dimethyl-7-ethylaminocoumarin FD-3 4-Methylumbelliferone FD-4 3-(2'-Benzothiazolyl)-7-diethylaminocoumarin FD-5 3-(2'-Benzimidazolyl)-7-N,N-diethylaminocou-msrin FD-6 7-Amino-3-phenylcoumarin FD-7 3-(2'-N-Methylbenzimidazolyl)-7-N,Ndiethyl-sminocoumarin FD-8 7-Diethylamino-4-trifluoromethylcoumarin FD-9 2,3,5,6-lH,4H-Tetrahydro-8-methylquinola-zino[9,9a,1-gh]coumarin FD-10 Cyclopenta[c]~ulolindino[9,10-3]-llH-pyran-ll-one FD-ll 7-Amino-4-methylcoumarin FD-12 7-Dimethylaminocyclopenta~c~coumarin FD-13 7-Amino-4-tri~luoromethylcoumarin FD-14 7-Dimethylamino-4-trifluoromethylcoumarin FD-15 1~2,4,5,3H,6H,lOH-Tetrahydro-8-trifluoro-methyl[l]benzopyrano[9,9a,1-gh]qulnolizln-10-one FD-16 4-Methyl-7-(sulfomethylamino)coumarin sodium salt FD-17 7-Ethylamino-6-methyl-4-trifluoromethylcou-marin FD-18 7-Dimethylamlno-4-methylcoumarin FD-l9 1,2,4,5,3H,6H,lOH-Tetrahydro-carbethoxy[1]-benzopyrano[9,9a,1-gh~quinolizino-10-one FD-20 9-Acetyl-1,2,4,5,3H,6H,lOH-tetrahydro[l]ben-zopyrano[9,9a,1-gh]quinolizino-lO-one FD-21 9-Cyano-1,2,4,5,3H,6H,lOH-tetrahydro[l]ben-zopyrano[9,9a,1-gh]quinolizino-10-one FD-22 9-~t-Butoxycarbonyl)-1,2,4,5,3~,6H,lOH-tet-rahyro[l]benzopyrano[9,9a,1-gh]quinolizino-10-one FD-23 4-Methylpiperidino[3,2-g]coumarin FD-24 4-Trifluoromethylpiperidino[3,2-g]coumarin FD-25 9-Carboxy-1,2,4,5,3H,6H,lOH-tetrahydro[l]ben-zopyrano[9,9a,1-gh]quinolizino-10-one FD-26 N-Ethyl-4-trifluoromethylpiperidino[3,2-g]-coumarin Another preferred class of fluorescent dyes are fluorescent 4-dicyanomethylene-4H-pyrans and 4-dicyanomethylene-4H-thiopyrans, hereinafter referred to as fluorescent dicyanomethylenepyran and thiopyran dyes. Preferred fluorescent dyes of this class are those satisfying formula (V):
( V ) C
/~\
R7-~ O-R6 X
wherein X represents oxygen or sulfur;
R6 represents a 2-(4-aminostyryl) group; and R7 represents a second R6 group, an alkyl group, or an aryl group.
Although X most conveniently represents oxy-gen or sulfur, it is appreciated that higher atomic number chalcogens shou.ld provide similar, though bathochromically shifted, response. The amino group can be a primary, secondary, or tertiary amino group. In one specifically preferred form the amino group can form at least one additional fused ring with the styryl phenyl ring. For example, the styryl phenyl ring and the amino group can together form a julolidine ring or the amino group can form an five or six membered ring fused with the styryl phenyl ring. The alkyl group forming R7 typically contains 39~

from 1 to 5 carbon atoms, preferably 1 to 3 carbon atoms. The aryl group forming R6 is preferably phenyl. When both R6 and R7 form a 2-(4-aminosty-ryl) group, the groups can be the same or different, but symmetrical compounds are more conveniently syn-thesized.
The following are illustrative fluorescent dicyanomethylenepyran and thiopyran dyes:
FD-27 4-(Dicyanomethylene)-2-methyl-6-(~-dimethyl-aminostyryl)-4H-pyran FD-28 4-(Dicyanomethylene)-2-methyl-6-~2-(9-julolidyl)ethenyl]-4H-pyran FD-29 4-(Dicyanomethylene)-2-phenyl-6-[2-(9-julolidyl)ethenyl]-4H-pyran FD-30 4-(Dicyanomethylene)-2,6-[2-(9-julolidyl)-ethenyl~-4H-pyran FD-31 4-(Dicyanomethylene)-2-methyl-6-[2-(9-julolidyl)ethenyl~-4H-thiopyran Useful fluorescent dyes can also be selected from among known polymethine dyes, which include the cyanines, merocyanines, complex cyanines and merocya-nines (i.e., tri-, tetra- and poly-nuclear cyanines and merocyanines), oxonols, hemioxonols, styryls, merostyryls, and streptocyanines.
The cyanine dyes include, joined by a meth-ine linkage, two basic heterocyclic nuclei, such as azolium or azinium nuclei, for example, those deri~ed from pyridinium, quinolinium, isoquinolinium, oxazo-lium, thiazolium, selenazolium, indazolium, pyrazo-lium, pyrrolium, indolium, 3H-indolium, imidazolium, oxadiazolium, thiadioxazolium, benzoxazolium, benzo-thiazolium, benzoselenazolium, benzotellurazolium, benzimidazolium, 3H- or lH-benzoindolium, naphthoxa-zolium, naphthothiazolium, naphthoselenazolium, naphthotellurazolium, carbazolium, pyrrolopyridinium, phenanthrothiazolium, and acenaphthothiazolium qua-ternary salts.

5~

Exemplary of the basic heterocyclic nuclei sre those satisfying Formulae VI and VII.
(VI) l~ ~ Z - ~1 =C - (LCL)q-N-R
z3 -C -(L-L)q=N -R

10 (VII) 1- - Q'- - ~l -C=L- (LsL)q-N-R
I

_ _ Q-_ _ _ -C-L==(L-L)q=N -R
where Z represents the elements needed to com-plete a cyclic nucleus derived ~rom basic heterocy-clic nitrogen compounds such as oxazoline, oxazole, benzoxazole, the naphthoxazoles (e.g., naphthl2,1-d3-oxazole, naphthl2,3-d]oxazole, and naphth~l,2-d~oxa-zole), oxadiazole, thiazoline, thiazole, benzothia-zole, the naphthothiazoles (e.g., naphthot2,1-d]thia-zole), the thiazoloquinolines (e.g., thiazolot4,5-b]-qulnoline), phenanthrothlazole, acenaphthothiazole,thladloxazole, selerlRzollne, selenRzole, benzoselena-zole, the naphtho-~elenazoles (e.g., naphthotl,2-d]-selenazole), benzotellurazole, naphthotellurazoles (e.g., napthotl,2-d]tellurazole), imldazollne, imlda-zole, benzlmldazole, the naphthlmidazoles (e.g.,naphtht2,3-d]imldazole~, 2- or 4-pyridlne, 2- or 4-qulnollne, 1- or 3-lsoqulnollne, benzoquinoline, 3H-indole, lH- or 3H-benzoindole, and pyrazole, which nuclei may be substituted on the ring by one or more of a wlde varlety of substltuents such as hydroxy, the halogens (e.g., fluoro, chloro, bromo, and lodo), alkyl groups or substltuted alkyl groups (e.g., ~z~

methyl, ethyl, propyl, lsopropyl, butyl, octyl, dodecyl, octadecyl, 2-hydroxyethyl, 3-sulfopropyl, carboxymethyl, 2-cyanoethyl, and trifluoromethyl), aryl groups or substituted aryl group~ (e.g., phenyl, l-naphthyl, 2-naphthyl, 4-sulfophenyl, 3-carboxy-phenyl, and 4-biphenylyl), aralkyl groups (e.g., ben-zyl and phenethyl), alkoxy groups (e.g., methoxy, ethoxy, and isopropoxy), aryloxy groups (e.g., phe-noxy and l-naphthoxy), alkylthio groups (e.g., meth-ylthio and ethylthio), arylthio groups (e.g., phenyl-thio, ~-tolylthio, and 2-naphthylthio), methylenedi-oxy, cyano, 2-thienyl, styryl, amino or substituted amino groups (e.g., anilino, dimethylamino, diethyl-amino, and morpholino), acyl groups, (e.g., formyl, acetyl, benzoyl, and benzenesulfonyl);
Q' repre~ents the elements needed to com-plete a cyclic nucleus derived from basic heterocy-clic nitrogen compounds such as pyrrole, indole, car-bazole, benzindole, pyrazole, indazole, and pyrrolo-pyridine;
R represents alkyl groups, aryl groups,alkenyl groups, or aralkyl groups, with or without substituent3, (e.g., carboxy, hydroxy, sulfo, alkoxy, sulfato, thio~ulfato, phosphono, chloro, and bromo substituents);
L is in each occurrence independently selected to represent a substituted or unsubstituted methine group -e.g., -CR = groups, where R repre-sents hydrogen when the methine group is unsubsti-tuted and most commonly represents alkyl of from l to4 carbon atoms or phenyl when the methine group i~
substituted; and q is 0 or l.
Cyanine dyes can contain two heterocyclic nuclei of the type shown in Formula VI ~oined by a methine linkage containing an uneven number of meth-ine groups or can contain a heterocyclic nucleus 3~

according to each of Formulae VI and VII ~oined by a methine linkage containing an even number of methine groups, where the methine groups can take the for~
-CR = as described above. The greater the number of the methine groups linking nuclei in the polymethine dyes in general and the cyanine dyes in particular the longer the absorption wavelengths of the dyes.
For example, dicarbocyanine dyes (cyanine dyes con-taining five methine groups linking two basic hetero-cyclic nuclei) exhibit longer absorption wavelengthsthan carbocyanine dyes (cyanine dyes containing three methine groups linking two basic heterocyclic nuclei) which ln turn exhibit longer absorption wavelengths than simple cyanine dyes (cyanine dyes containing a single methine group linking two basic heterocyclic nuclei). Carbocyanine and dicarbocyanine dyes are longer wavelength dyes while simple cyanine dyes are typically yellow dyes, but can exhibit absorption maxima up to about 550 nm in wavelength with proper choice of nuclei and other components capable of bathochromically shifting absorption.
Preferred polymethine dyes, particularly cyanine dyes, for use as fluorescent dyes are 90=
called rigidized dyes. These dyes are constructed to restrict the ~lovement of one nucleus in relation to another. This avoids radiationless, kinetic dissipa-tion of the excited state energy. One approach to rigidizing the dye structure i9 to incorporate a separate bridging group providing Q separate linkage in addition to the methine chain linkage ~oining the terminal nuclei of the dye. Bridged polymethine dyes are illustrated by Brooker et al U.S. Patent 2,478,367, Brooker U.S. Patent 2,479,152, Gilbert U.S. Patent 4,490,463, and Tredwell et al, "Picosec-ond Time Resolved Fluorescence Lifetimes of the Poly-methine and Related Dyes", Chemical Phvsics, Vol. 43 (1979) pp.307-316.

ii.3g~3 The methine chain ~oining polymethine dye nuclei can be rigidized by including the methine chain as part of a cyclic nucleus joining the termi-nal basic nuclei of the dye. One of the techniques for both rigidizing and bathochromically shifting the absorption maxlma of polymethine dyes in general and cyanine dyes in particular is to include in the meth-ine linkage an oxocarbon bridging nucleus. Exemplary oxocarbon bridging nuclei can take any of the forms indicated by Formula YIII.
(VIII) O
Il 11 C- (C) \C~

O--I

o o Il 11 /C - (C) ~:/
I

wherein n is the integer 0, l, or 2.
Merocyanine dyes link one of the cyanine dye type basic heterocycllc nuclei described above to an acidic keto methylene nucleus through a methlne llnk-age as described above, but containing zero, two, or a higher even number of methine groups. Zero methlne dyes, those containlng no methine groups in the link-age between nuclei, exhibit a double bond linkage between the nuclel ln one resonance form and a single bound llnkage ln another resonance form. In elther resonance form the llnkage sltes in the nuclel are formed by methine groups formlng a part of each nucleus. Zero methine polymethine dyes are yellow dyes.

Exemplary acidic nuclei are those which sat-isfy Formula IX.
(IX) 0 I!_ =./

where Gl represents an alkyl 8rouP or substituted alkyl group, an sryl or substituted aryl group, an aralkyl group, an alkoxy group, an aryloxy group, a hydroxy group, an amino group, or a substituted amino group, wherein exemplary substituents can take the various forms noted in connection with Formulae VI
and VII;
G2 can represent any one of the groups listed for Gl and in addltion can represent a cyano group, an alkyl, or arylsulfonyl group, or a group represented by -C-Gl, or G2 taken together with can represent the elements needed to complete a cyclic acidic nucleus such as those derived from 2,4-oxazolidinone (e.g., 3-ethyl-2,4-oxazolidindione), 2,4-thiazolidindione (e.g., 3-methyl-2,4-thiazolidin-dione), 2-thio-2,4-oxazolidindione (e.g., 3-phenyl-2-thio-2,4-oxazolidlndione), rhodanine, such as 3-eth-ylrhodanine, 3-phenylrhodanine, 3-(3-dimethylamino-propyl)rhodanine, and 3-carboxymethylrhodanine, hydantoin (e.g., 1,3-diethylhydantoin and 3-ethyl-1-phenylhydantoin~, 2-thiohydantoin (e.g., 1-ethyl-3-phenyl-2-thiohydantoin, 3-heptyl-1-phenyl-2-thiohy-dantoin, and arylsulfonyl-2-thiohydantoin), 2-pyrazo-lin-5-one, such as 3-methyl-1-phenyl-2-pyrazolin-5-one, 3-methyl-1-(4-carboxybutyl)-2-pyrazolin-5-one, and 3-methyl-2-(4-sulfophenyl)-2-pyrazolin-5-one, 2-isoxazolin-5-one (e.g., 3-phenyl-2-i~oxazolin-5-one), 3,5-pyrazolidindione (e.g., 1,2-diethyl-3,5-pyrazoli-dindione and 1,2-diphenyl-3,5-pyrazolidindione), 1,3-2 $ ~

indandione, 1,3-dioxane-4,6-dione, 1,3-cyclohexanedi-one, barbituric acid (e.g., l-ethylbarbituric acid snd 1,3-diethy;barbituric acid), and 2-thiobarbituric acid (e.g., 1,3-diethyl-2-thiobarbituric ~cid and 1,3-~is(2-methoxyethyl)-2-thiobarbituric acid).
Useful hemicyanine dyes are essentially similar to the merocyanine dyes described above, dif-fering only in substituting for the keto methylene group of Formula IX the group shown below in Formula 10 X.
(X) G3 - ~G4 where G3 and G4 may be the ssme or different and may represent alkyl, substituted alXyl, aryl, substituted aryl, or aralkyl, as illustrated for ring substitu-ents in Formula VI or G3 and G taken together complete a ring sy~tem derived from a cyclic secon-dary amine, such as pyrrolidine, 3-pyrroline, piperi-dine, piperazine (e.g., 4-methylpiperazine ~nd 4-phenylpiperazine), morpholine, 1,2,3,4-tetrahydro-quinoline, deahydroquinoline, 3-azabicyclo[3,2,2]-nonane, indoline, azetidine, snd hexahydroazepine.
Useful hemioxonol dyes exhibit a keto meth-ylene nucleus as shown in Formula IX and a nucleus as shown in Formula X ~oined by a methine linkage as previously described containing one or a higher uneven number of methine groups.
Useful merostyryl dyes exhibit a keto meth-ylene nucleus as shown in Formula IX and a nucleus as shown in Formula XI ~oined by a methine linkage as described above containing one or a higher uneven number of methine groups.
35 (XI) - - \ G3 ~ G4 'S~

where G and G are as previously defined.
The longer wavelength cyanine, merocyanine, hemicyanine, hemioxonol, and merostyryl dyes described above are intended to be ~llustrative of the simpler structural forms of useful longer wave-length polymethine dyes. It is generally recognized that substituents can ~oin the nuclei and methine linkages to form additional cyclic structures. Fur-ther, the dyes can contain three or more nuclei. Forexample, by substituting a merocyanine dye in methine linkage with a second ba~ic heterocyclic nucleus of the cyanlne dye type an allopolar cyanine dye can be formed. Further, the varlous substituents not form-ing a part of the dye chromophore can be varied asdesired to tailor dye physical properties, particu-larly hydrophobicity and hydrophilicity, to suit the particular film forming components employed. By choosing as the aliphatic moieties of the dyes hydro-carbon groups having more carbon atoms ~e.g., fromabout 6 to 20 carbon atoms) the dyes can be rendered more oleophilic while hydrocarbon groups containing fewer numbers of carbon atoms (e.g., 1 to 5 carbon atoms) and particularly those bearing polar ~ubstitu-ents render the dyes more hydrophilic. The aromaticmoieties of the dyes typically contain from 6 to 10 carbon atoms.
The Eollowing are illustrative of polymeth-ine dyes capable of maximum light absorption at shorter (<550 nm) wavelengths:

~S39 ~, ~s\ ~s~
!\N+~---=-\~U\ ,~! X-R R
R X
FD-32C16H33 Cl FD-34-CH2CH=CH2 Cl PTS = P-toluene sulfon~te T'~---= --=-/ 'I Cl-FD-36/S \ / S\
il +~ ._.=._.=. \ i Cl !~ ,I!,+N~l-CH=!,N,I!, ~!
X
R R
_ R X

FD-39 -C5Hll BF4 FD-40 C!H3 FD-41 , _.

I~ /0\N ~

FD-42 ~-=-~

I~ ,O~+~---=---\ ,I

FD-43 ~-~ ~0 ~ ~ 0~ ~-~
N ~ \ N/ \ ~
l l Cl-\ o/-~ Cl 1 ~ R

_Ra_~R_ .__Rb _ FD-47 ~~ ~-~

CH2CH3 \ ~ ~ I

~539~3 R~ CH=CH ~ ~ (NRC)2 X

n R Rc X
FD-48 l -CH3 C2H5 PTS-FD-49 1 (CH2)3S03- C5Hl1 FD-50 1 (CH2)4S03- C5Hl1 FD--51 2 (CH2)5S03-- C2H5 lC FD-52 3,3'-Ethylenethiacyanine ~-toluenesulfonate FD-53 1',3-Ethylenethia-2'-cyanine chloride FD-54 1,1'-Ethylene-2,2'-cyanine chloride FD-55 3,3'-Ethyleneoxacyanine chloride FD-56 1,1'-Dlethyl-3,3'-Ethylenebenzimidazolocya-nine ~-toluenesulfonate FD-57 1,1'-Dlethyl-3,3'-methylenebenzimidazolocya-nlne chloride FD-58 1,1'-Ethylenecyanine chloride FD-59 1,1'-Methylenecyanine chloride FD-60 5,5',6,6'-Tetrachloro-1,1'-diethyl-3,3'-eth-anediylbenzimldazolocyanine chloride FD-61 5,5',6,6'-Tetrachloro-l,l'-ethanediyl-3,3'-dimethylbenzimidazolocyanine chloride FD-62 Anhydro-5,5',6,6'-tetrachloro-1,1'-ethandiyl-3,3'-bls(3-sulfopropyl)benzimidazolocyanine hydroxlde, sodlum salt FD-63 2,2'-Methanediylbis-(5,6-dichloro-1-methyl-benzlmidazole FD-64 5,5',6,6'-Tetrachloro-l,l'-dimethyl-3,3'-pro-panediylbenzlmidazolocyanine P-toluenesulfo-nate FD-65 5,5',6,6'-Tetrachloro-l,1'-dimethyl-3,3'-methanediylbenzimidazolocyanine ~-toluene-sulfonate FD-66 5,5',6,6'-Tetrachloro-1,1'-ethanediyl-3,3'-bis(2,2,2-trifluoroethyl)benzimidazolocyanine ~-toluenesulfonate ~ ;~Pr ~3~

FD-67 5,5'~6,6'-Tetrachloro-l,l'-ethanediyl-3,3',8-trimethylbenzimidazolocyanine ~-toluenesul-fon~te Many polymethine dyes are capable of maximum light absorption at longer visible (> 550 nm) wave-lengths, with maximum fluorescence wavelengths gener-~lly lying in the red ~nd ne~r infrared portions of the ~pectrum. The following are illustrative of polymethine dyes capable of m~ximum light absorption at longer visible wavelengths:

R` ~ \; / \ -tCH CH ~ CH= / \l ~

n Rd X

FD-71 3 tCH=CH~ 2 C104-CH - ~ \ =------ / ~ -CH

FD-73 /S\ / S\
ll +~ ~-=-~ J

CH2CH3 CH2CH3 Cl04-FD-74 /S \ / S\
il N~ -- = -- = \ il CH2CH3CH2CH3 Cl N =i\ ~IU\ ~i Cl .9~ ~.~S\ IR ~S~.~ ~.

R R X
R 8 Re X
R

FD-78 -C4Hg -CH3 - C1 FD-79 -C5Hll -CH3 C1 FD-80 -i-C3H7 -CH3 Cl FD-81 -C3H7 -C2H5 Cl FD--83 --C2H5 C6Hl1 tCH=CHt2 Cl (cyclohexyl) FD--84 C2H5 C15H31 tCH=CHt2 Cl 29~S391 ff R8~ R8d i8a\
x R R
R R8a R8b R3c R8 R X
FD-85 -CH2CH3 H H H H H Cl FD-86 -CH2CH3 H H H H OCH3 PTS _ (CH3)2 (CH3)2 ;f \0/ \ ' ' f'\ f'/ ~ ~; Cl FD-93 C6 ~ C6H5 C104-~~ -CH=CH-i~ i=CH-CH=- ~ 1l 35 (CH2)3cOOCH3 (CH2)3Co 3 39~3 F~94 ~-~ ~0\ ~0~ ~-~

F~95 ~ 0~ ~0~ ~-~

li + T~
X
R R
R X
F~96 --CH2CH2C6H5 BF4 F~97 CH2CH3 Cl IU
c~3\.~

`F

FD-101CH3CH2~ ~ CH2CH3 5 0\ ~I I O Cl-FD-102 ~-\ /S\
O ~ --I O

Rh Rb o Rh Rh O-R R
Rh R Rg 20 FD-104 -CH3 -CH3 ~cH=cHt2 I~ O
T

FD-107 2 adjacent Rh = ~CH2~5 i.e., spirocyclohexyl 30 FD-108 C 3\ CH3 0 I O
I~ ,0~+~- ` ~- ~ /~\

~ H3 FD-llO R = -CloH7, i.e.
a-nsphthyl 25C12 - \i~ \C H

FD-112 _ 0 \._./ I_ \._./
FD-113~ 3 /CH3 C ~ /CH2CH3 CH3- W / ~- ~ D - CH3 O
H H
Another useful class of fluore~cent dyes flre 4-oxo-4H -benz--[d,e]~nthrscenes, hereinafter referred to a~ oxobenz~nthrscene dyes. Preferred fluorescent oxobenzanthracene dyes are those represented by for-muls XII:
(XII) yl ~ 9 i!

~ ~i li i h li ./-\./
o ~2~5~8 ~o-ln this structure, R is hydrogen, substi-tuted or unsubstituted alkyl (prefer~bly of 1 to 12 carbon atoms, e.g. methyl, ethyl, isopropyl, benzyl, phenethyl, etc.), substituted or unsubstituted hydroxyalkyl (preferably of 1 to 12 carbon atoms, e.g. hydroxymethyl, 2-hydroxyethyl, 2-hydroxyisopro-pyl, etc.), or substituted or unsubstituted alkoxy-carbonyl (preferably of 2 to 12 carbon atoms, e.g.
methoxycarbonyl, ethoxycarbonyl, n-propoxycarbonyl, etc.). Preferably, R9 is hydrogen, substituted or unsubstituted alkyl or substituted or unsubstituted alkoxycarbonyl, and more preferably, it is substi-tuted or unsubstituted alkoxycarbonyl.
W is hydrogen or an electron withdrawing group as that term is understood in the art (i.e. a group generally having a positive Hammett sigma value RS determined by standard procedures). Particul~rly useful electron withdrawing groups include, but Qre not limited to, halo (e.g. fluoro, chloro, bromo), cyano, carboxy, acyl, substituted or unsubst~tuted arylsulfonyl (preferably of 6 to 10 carbon atoms, e.g. phenylsulfonyl, tolylsulfonyl, etc.), substi-tuted or unsubstituted alkylsulfonyl (preferably of 1 to 6 carbon atoms, e.g. methylsulfonyl, ethylsulfo-nyl, etc.), substituted and unsubstituted dialkyl-phosphinyl (preferably where each alkyl group inde-pendently has 1 to 10 carbon atoms, e.g. methyl, ethyl, butyl, decyl, etc.) and substituted or unsub-stituted dialkyl phosphono (preferably where each alkyl group independently has 1 to 10 carbon atoms as defined above). Preferably, W is hydrogen or halo.
Y is hydrogen, or Q group comprised of a hetero atom having a lone pair of electrons or a negative charge with an associated cation, e.g.
hydroxy, mercapto or amino (-NR"R"'). R" and R"' are independently substituted or unsubstituted alkyl (preferably of 1 to 10 carbons, e.g., methyl, ethyl, D3~l3 decyl, etc.), substituted or unsubstituted aryl (preferably of 6 to 10 carbons, e.g., phenyl, naph-thyl, etc.), or R" and R"', taken together, can rep-resent the atoms necessary to complete a substituted or unsubstituted heterocyclic ring (preferably of 5 to 10 carbon, nitrogen or oxygen atoms, e.g. a mor-pholino, pyrrolidinyl, pyridyl, piperidino, etc.
ring). yl can also be substituted or unsubstituted alkoxy (preferably of 1 to 10 carbon atoms, e.g.
methoxy, ethoxy, 2-chloro-1-propoxy, etc.), substi-tuted or unsubstituted carbamyloxy o (-0-C-NR"R"'), wherein R" and R"' are defined above, -0 M+ or -S M+, wherein M+ is a monovalent cation, e.g. Na+, K+, Li+, NH4+, etc. Preferably Ylis hydroxy or -0 M~.
R9 W yl FD-114 Methyl Hydrogen Hydroxy FD-115 Methyl Hydrogen -0 Na+
20 FD-116 Methyl ChloroHydroxy FD-117 Methyl Chloro-0 Na+
FD-118 Methyl ChloroN-methyl-N-phenylcarbamyloxy FD-ll9 Methyl Hydrogen Pyrrolidinyl 25 FD-120 Butoxy- Hydrogen Hydroxy carbonyl FD-121 Butoxy- Hydrogen -0 Na~
carbonyl 30 FD-122 Butoxy- Chloro-0 Na+
carbonyl The oxobenzanthracene dyes illustrated above can have one or more substituents other than those specifically illustrated in the structure as long as the substituents do not adversely affect the fluores-cence of the compound, such as alkyl (e.g., alkyl of1 to 5 carbon atoms), aryl (e.g., phenyl), and other groups.

, ;~

39~

The oxobenzanthracene dyes can be prepared generally using the following procedure. The detail of seversl preparations are provided in APPENDIX I
below. The general preparatory procedure includes:
(1) preparation of a dihydrophenalenone by the proce-dure described by Cooke et al, Australian J. Chem., 11, pp. 230-235 (1958), (2) preparation of the lithium enolate of the dihydrophenalenone, (3) reac-tion of the lithium enolate with the appropriate phosphonium iodide reagent, and (4) reaction of this product with cupric chloride and lithium chloride to produce the chlorinated or unchlorinated dye.
Another useful class of fluorescent dyes are xanthene dyes. One particularly preferred clas~ of xanthene dyes are rhodamine dyes. Preferred fluores-cent rhodamine dyes are those represented by formula XIII:
(XIII) ~ / \R10 Rl5~., ,I~ . Rl2 Rl N~ +N~Rl 6 I

Rll ~ / \R10 Rl 59~ Rl 2 R ~ ~./ ~ \.~ \ ~R
l18 R14 R R17 where Rl and R are independently hydrogen, car-boxyl, sulfonyl, alkanoyl, or alkoxycarbonyl groups;
R12 R13 R14 and R15 are hydrogen;
R16 R17 R13 and Rl9 are alkyl group~;
and X is an anion; or any one of or all of following substituent pairs:
12 d R16 R13 and R17. R14 and R , and R and R19, complete a five or six membered ring containing nitrogen as the sole heteroatom.
The alkyl moieties in each instance contain from 1 to 5 carbon atoms, preferably 1 to 3 carbon atoms. When substituent pairs complete a fused ring, the ring csn, for example, take the form of a pyran ring when a single fused rin8 including a formula nitrogen atom is formed or a ~ulolidene ring (including a formula fused benzo ring) when two fused rings each including the ~ame nitrogen atom of the formula are formed.
The following are illustrative of rhodamine dyes known to be useful laser dyes:
FD-123 [9-(o-Carboxyphenyl)-6-(diethylamino)-3H-xanthen-3-ylidene]diethyl ammonium chlo-rlde [a.k.a rhodamine B]
FD-124 N--{6-[Diethylamino]-9-[2-(ethoxycarbon-yL)phenyl]-3H-xanthen-3-ylidene}-N-eth-ylethanaminium perchlorate FD-125 Ethyl o-[6-(Ethylamino)-3-(ethylimino)-2,7-dimethyl-3H-xanthenyl]benzoate chlo-ride FD-126 Ethyl o-[6-(ethylamino)-3-(ethylimino)-2,7-dimethyl-3H-xanthenyl]benzoate per-chlorate FD-127 Ethyl o-[6-(ethylamino)-3-(ethylimino)-2,7-dimethyl-3H-xanthenyl]benzoate tetra-fluoroborate FD-128 o-[6-(Ethylamino~-3-(ethylimino)-2,7-dimethyl-3H-xanthenyl]benzoic acid FD-129 o-(6-Amino-3-imino-3H-xanthenyl)benzoic acid hydrochloride FD-130 o-[6-(Methylamino)-3-methylimino)-3H-xan-then-9-yl]benzoic acid perchlorate FD-131 Methyl o-(6-amino-3'-imino-3H-xanthen-9-yl)benzoate monohydrochloride FD-132 8-(2,4-Disulfophenyl)-2,3,5,6,11,12,14,15-lH,4H,lOH,13H-octahydroquinolizinol-[9,9a,1-bc;9,9a,1-hi]xanthylium hydroxide inner salt FD-133 Sulforhodamine B
FD-134 Q-[6-(Dimethylamino)-3-(dimethylamino)-3H-3H-xanthen-9-yl]benzoic acid perchlorate Another specifically preferred class of xan-thene dyes are fluorescein dyes. Preferred fluores-cein dyes are those represented by formula XIV:
(XIV) I~ ~o\Rlo R~~ o~ ~ R
where R10 and Rll are as previously defined and R20 and R21 are hydrogen, alkyl, aryl, or halo substituents. Preferred alkyl groups contain from 1 to 5, optimally from 1 to 3 carbon atoms while phenyl iB a preferred aryl group.
An illustrative fluorescein dye is FD-135 9-(Q-Carboxyphenyl)-6-hydroxy-3H-xanthen-3-one FD-136 9-(o-Carboxyphenyl)-2,7-dichloro-6-hydroxy-3H~xanthen-3-one ., ~ .

3g~

Another useful group of fluorescent dyes are pyrylium, thiapyrylium, selenapyrylium, and telluro-pyrylium dyes. Dyes from the first three of these classes are disclosed by Light U.S. Patent 3,615,414 while dyes of the latter class are disclosed by Detty U.S. Patent 4,584,258. Since the latter two classes of dyes are bathochromically shifted toward the infrared the former two classes of dyes are preferred for achieving visible light emissions.
Illustrative preferred fluorescent pyrylium and thiapyrylium dyes are represented by formula XV:
(XV) R22 0\ ~I
~ _ ~ \o +~ -oR23 where R22 is hydrogen, methyl, or a tertiary amino group, optimally a -NR R group;
R23 is an alkyl group;
X is an anion; and J is oxygen or sulfur.
The alkyl group preferably contains from 1 to 5 car-bon atoms and optimally from 1 to 3 carbon atoms.
Illustrative pyrylium and thiapyrylium fluorescent dyes satisfying formula XV are the following:
FD-137 4-(4-dimethylaminophenyl)-2-(4-methoxy-phenyl)-6-phenylpyrylium perchlorate FD-138 4,6-diphenyl-2-(4-ethoxyphenyl)thiapyry-lium ~-toluenesulfonate FD-139 2-(4-methoxyphenyl)-6-phenyl-4-(~-tolyl)-pyrylium tetrafluoroborate Another useful class of fluorescent dyes are fluorescent carbostyril dyes. These dyes are charac-terized by a 2-quinolinol or isoquinolinol ring structure, often fused with other rings. The wave-length of maximum fluorescence generally increases with the presence of other fused rings. Typical of simple carbostyril dyes, which fluoresce in the blue portion of the spectrum, are the following:
FD-140 7-Amino-4-methyl-2-quinolinol [a.k.a. 7-amino-4-methylcarbostyril]
FD-141 7-Dimethylamino-2-hydroxy-4-methylquino-line [a.k.a. 7-dimethylamino-4-methylcarbosty-ril]
FD-142 3,3'-Bis[N-phenylisoquinoline]
Examples are more complex fused ring carbostyril dyes are provided by Kadhim and Peters, "New Intermediates and Dyes for Synthetic Polymer Fibres Substituted Benzimidazolothioxanthenoisoquinolines for Polyester Fibres~, JSDC, June 1974, pp. 199-201, and Arient et al, "Imidazole Dyes XX -Colouring Properties of 1,2-Napthooxylenebenzimidazole Derivatives", JSDC, June 1968, pp. 246-251. Illustrative of these more com-plex carbostyril dyes are the following:
FD-143 Benzimidazo[1,2-b]thioxantheno-[2,1,9,d,e,f]isoquinolin-7-one and its stereo isomer Benzimidazo[1,2-a]thioxantheno~
[2,1,9,d,e,f]isoquinolin-7-one Among other fused ring ~luorescent dyes the perylene dyes, characterized by a dinapthylene nucleus. A variety of useful fluorescent perylene dyes are known, such as, for example those disclosed by Rademacher et al, "Soluble Perylene Fluorescent Dyes with Photostability", Chem. Ber., Vol. 115, pp.
2927-2934, 1982, and European Patent Application 553,363Al, published July 7, 1982 One preferred perylene dye is illustrated by formula XVI:

~o~

(XVI) (R24 ~ \.~ R25)3 where R24 and R25 are independently selected from the group consisting of alkyl, halo, and haloalkyl substituents. Preferred alkyl groups having from 1 to 5 carbon atoms, optimally from 1 to 3 carbon atoms.
Another preferred group Of perylene dyes are the 3,4,9,10-perylenebis(dicarboximides), hereinafter referred to a perylenebis(dicarboximide) dyes. Pre-ferred dyes of this class are represented by formula XVII:
(XVII) (R26~ 0 0 (R27 o where R26 and R27 are independently selected from the group consisting of alkyl, halo, and haloalkyl sub8tituents. Preferred alkyl groups having from 1 to 5 carbon atoms, optimally from 1 to 3 carbon atoms.
Illustrative of preferred peryle~e dyes are the following:
FD-144 Perylene FD-145 1,2-Bis(5,6-Q-phenylenenapthalene) FD-146 N,N'-diphenyl-3,4,9,10-perylenebis(dicar-boximide) FD-147 N,N'-di(p..-tolyl)-3,4,9,10-perylenebis-(dicarboximide) FD-148 N,N'-di(2,6-di-t-butyl)-3,4,9,10-perylene-bis(dicarboximide) The foregoing listing of preferred fluores-cent dyes useful in combination with the host materi-als, though lengthy, iS recognized to be only exem-J~'J~

plary of known fluorescent dyes, both in the classes specifically identified and in still other dye classes. For example, many other classes of known fluorescent dyes, such as acridine dyes; bis(styryl)-benzene dyes; pyrene dyes; oxazine dyes; and phenyl-eneoxide dyes, sometimes referred to as POPOP dyes;
are useful, specific illustrative dyes from these classes including the following:
FD-149 9-Aminoacridine hydrochloride FD-150 ~-Bis(o-methylstyryl)benzene FD-151 2,2'-p-Phenylenebis(4-methyl-5-phenyloxa-zole) FD-152 5,9-Diaminobenzo[a]phenoxazonium perchlo-rate FD-153 5-Amino-9-diethylaminobenz[a]phenoxazonium perchlorate FD-154 3,7-Bis(diethylamino)phenoxazonium per-chlorate FD-155 3,7-Bis(ethylamino)-2,8-dimethylphenoxa-zin-5-ium perchlorate FD-156 9-Ethylamino-5-ethylimino-10-methyl-5H-benzo[a]phenoxazonium perchl.orate FD-157 8-Hydroxy-1,3,6-pyrene-trisulfonic acid trisodium salt Not only are there many available classes of fluorescent dyes to choose from, there are wide choices of individual dye properties wlthin any given class. The absorption maxima and reduction poten-tials of individual dyes can be varied through the choice of substituents. As the conjugation forming the chromophore of the dye is increased the absorp-tion maximum of a dye can be shifted bathochromically.
Emission maxima are bathochromic to the absorption maxima. Although the degree of bathochro-mic shifting can vary as a function of the dye class,usually the wavelength of maximum emission is from 25 to 125 nm bathochromically shifted as compared to the ~39 wavelength of maximum absorption. Thus, dyes which exhibit absorption maxima in the near ultraviolet in almost all cases exhibit maximum emissions in the blue portion of the spectrum. Dyes which exhibit absorption maxima in the blue portion of the spectrum exhibit emission maxima in the green portion of the spectrum, and, similarly, dyes with absorption maxima in the red portion of the spectra tend to exhibit emission maxima in the near infrared portion of the spectrum.
In one form of the invention the material forming the luminescent zone can be one uniform layer interposed between and contacting both the cathode and the hole injection zone of the EL device. As an alternative construction a separate layer containing the host material, but lacking the fluorescent mate-rial, can be interposed between the luminescent zone and the cathode. Although the additional interposed organic electron injection layer can be of any con-ventional form, it is preferred that both the elec-tron injection layer and the layer forming the lumi-nescent zone be present in the form of a thin film (< 1 ~m in thickness) and most preferred that these layers have a combined thickness no greater than those thicknesses indicated above ~or the lumi-nescent zone The organic luminescent medium of the EL
devices of this invention preferably contains at least two separate organic layers, at least one layer forming a zone for transporting electrons injected from the cathode and at least one layer forming a zone for transporting holes injected from the anode.
The latter zone is in turn preferably formed of at least two layers, one, located in contact with the anode, providing a hole injecting zone and the remaining layer, interposed between the layer forming the hole injecting zone and the layer providing the 39~3 electron tr~nsporting zone, providing a hole trans-porting zone. While the description which follows is directed to the preferred embodiments of organic EL
device~ according to this invention which employ at least three separate organic lsyers, as taught by Van Slyke et al, it is appreciated that either the layer forming the hole in~ecting zone or the layer forming the hole transporting zone can be omitted and the remaining layer will perform both functions. Higher initial and sustained performance levels of the organic EL devices of this invention are realized when the separate hole in~ecting and hole transport-ing layers described below are employed in combina-tion.
A layer containing a porphyrinic compound forms the hole in~ecting zone of the organic EL
device. A porphyrinic compound is any compound, natural or synthetic, which is derived from or includes a porphyrin ~tructure, including porphine itself. Any of the porphyrinic compounds di~closed by Adler U.S. Patent 3,935,031 or Tang U.S. Patent 4,356,429, can be employed.
Preferred porphyrinic compounds are those of structural formula (XVIII):
25 (XVIII) T\ /T

Tl / ~ N ~ . Tl [I ~ _ M - 5 i1 .=., T~ T2 wherein Q is -N= or -C(R)=;
M is a metal, metal oxide, or metal halide;

R is hydrogen, alkyl, aralkyl, aryl, or alkaryl, and Tl and T2 represent hydrogen or together com-plete a unsaturated 6 membered ring, which can include substituents, such as alkyl or halogen. Pre-ferred 6 membered rings are those formed of carbon, sulfur, and nitrogen ring atoms. Preferred alkyl moieties contain from about 1 to 6 carbon atoms while phenyl constitutes a preferred aryl moiety.
In an alternative preferred form the por-phyrinic compounds differ from those of structural formula (XVIII) by substitution of two hydrogen for the metal atom, as indicated by formula (IXX):
(IXX) \.=./

0~ / \T2 Highly preferred examples of u~eful porphyr-inic compound~ are metal free phthalocyanines and metal containing phthalocysnines. While the porphyr-inic compounds in general and the phthalocyanines in particular can contain any metal, the metal prefera-bly has a positive valence of two or higher. Exem-plary preferred metals are cobalt, magnesium, zinc, palladium, nickel, and, particularly, copper, lead, and platinum.
Illustrative of useful porphyrinic compounds are the following:
PC-l Porphine PC-2 1,10,15,20-Tetraphenyl-21H,23H-porphine copper (II) 3~1~

PC-3 1,10,15,20-Tetraphenyl-21H,23H-porphine zinc (II) PC-4 5,10,15,20-Tetrakis(pentsfluorophenyl)-21H,23H-porphine PC-5 Silicon phthalocyanine oxide PC-6 Aluminum phthalocyanine chloride PC-7 Phthalocyanine (metal free) PC-8 Dilithium phthalocyanine PC-9 Copper tetrsmethylphthalocyanine PC-10 Copper phthalocyanine PC-ll Chromium phthalocyanine fluoride PC-12 Zinc phthalocyanine PC-13 Lead phthalocyanine PC-14 Titanium phthalocyanine oxide PC-15 Magnesium phthalocyanine PC-16 Copper octamethylphthalocyanine The hole transporting layer of the organic EL device contains at least one hole transporting aromatic tertiary amine, where the latter is under-stood to be a compound containing at least one triva-lent nitrogen atom that is bonded only to carbon atoms, at least one of which is a member of an aro-matic ring. In one form the aromatic tertiary amine can be an arylamlne, such as a monoarylamine, diaryl-amine, triarylamine, or a polymeric arylamine. Exem-plary monomeric trlarylamines are illustrated by Klupfel et al U.S. Patent 3,180,730. Other suitable triQrylamines substituted with vinyl or vinylene radicals and/or containing at least one active hydro-gen containlng group are disclosed by Brantley et alU.S. Patents 3,567,450 and 3,658,520.
A preferred class of aromatic tertiary amines are those which include at least two aromatic tertiary amine moieties. Such compounds include those represented by structural formula (XX):
(XX) QlG/Q2 ~Z~ ~ 3 wherein Q and Q are independently aromstic tertiary amine moieties and G is a linking group such an arylene, cycloalkyl-ene, or alkylene group or a carbon to carbon bond.
A particularly preferred class of class oftriarylamines satisfying structural formula (XX) and containing two triarylamine moieties are those satis-fying structural formula (XXI):
(XXI) R24 where R and R25 each independently represents a hydrogen atom, an aryl group, or an alkyl group or R24 and R25 together represent the atoms complet-ing a cycloalkyl group and R2 and R27 each independently represents an aryl group which is in turn substituted with a diaryl substituted amino group, as indicated by structural formula (XXII):
(XXII) ,R28 \R29 wherein R2a and R29 are independently selected aryl groups.
Another preferred class of aromatic tertiary amines are tetraaryldiamines. Preferred tetraaryldi-amines lnclude two diarylamino groups, such 8S indi-cated by formula (XXII), linked through an arylene group. Preferred tetraaryldiamines include those represented by formula (XXIII).
(XXIII) R30 ,R31 ~ Aren Nr\ 32 wherein Are is an arylene group, n i5 an integer of from 1 to 4, and Ar R30 R31 and R32 are independentlY
selected aryl groups.
The various alkyl, alkylene, aryl, and aryl-ene moieties of the foregoing structural formulae (XX), (XXI), (XXII), and (XXIII) can each in turn be substituted. Typical substituents including alXyl groups, alkoxy groups, eryl groups, aryloxy groups, and halogen such as fluoride, chloride, and bromide.
The various alkyl and alkylene moieties typically contain from about 1 to 6 carbon atoms. The cycloal-kyl moieties can contain from 3 to about 10 carbon atoms, but typically contain five, six, or seven ring carbon atoms-e.g., cyclopentyl, cyclohexyl, and cycloheptyl ring structures. The aryl and arylene moieties are preferably phenyl and phenylene moieties.
Whlle the entire hole transporting layer of the organic electroluminesce medium can be formed of a single aromatic tertiary amine, it is a further recognition of this invention that increased st~-bility can be realized by employing a combination of aromatlc tertiary amines. Specifically, as demon-strated in the examples below, it has been observed that employlng a triarylamine, such as a triarylamine satisfying formula (XXI), in combination with a tet-raaryldiamine, such as indicated by ~ormula (XXIII), can be advant~geous. When a triarylamlne is employed in combination with a tetraaryldiamine, the latter is positioned ~s a layer interposed between the triaryl-amine and the electron in~ecting and transportinglayer.
Representative useful aromatic tertiary amines are disclosed by Berwick et al V.S. Patent 4,175,960 and Van Slyke et al U.S. Patent 4,539,507.
Berwick et al in addition discloses as useful hole transporting compounds N substituted carbazoles, ~L ~'id~ 5 3~8 which can be viewed as ring bridged variants of the diaryl and triarylamines disclosed above.
Illustrative of useful aromatic tertiary amine~ are the following:
ATA-l 1,1-Bis(4-di-~-tolylaminophenyl)cyclohexane ATA-2 1,1-Bis(4-di-~-tolylaminophenyl)-4-phenyl-cyclohexane ATA-3 4,4'-Bis(diphenylamino)quadriphenyl ATA-4 Bis(4-dimethylamino-2-methylphenyl)phenyl-methane ATA-5 N,N,N-Tri(~-tolyl)amine ATA-6 4-(di-~-tolylamino)-4'-[4(di-~-tolylamino)-styryl]stilbene ATA-7 N,N,N',N'-Tetra-~-tolyl-4,4'-diaminobi-phenyl ATA-8 N,N,N',N'-Tetraphenyl-4,4'-diaminobiphenyl ATA-9 N-Phenylcarbazole ATA-10 Poly(N-vinylcarbazole) Any conventional electron in~ecting and tran~porting compound or compounds can be employed in forming the l~yer of the organic luminescent medium ad~acent the cathode. This layer can be formed by historically taught luminescent materials, such as anthracene, n~lphthalene, phenanthrene, pyrene, chrys-ene, and perylene and other fu~ed ring luminescentmaterials containlng up to about 8 fu~ed rings a~
illustrated by Gurnee et al U.S. Patent 3,172,862, Gurnee U.S. Patent 3,173,050, Dresner, "Double In~ec-tion Electroluminescence in Anthracene", RCA Review, Vol. 30, pp. 322-334, 1969; snd Dresner U.S. Patent 3,710,167, cited above. Although such fused ring luminescent materlals do not lend themselves to form-lng thin (<1 ~m) films and therefore do not lend themselves to achieving the highest attainable EL
device performance levels, organic EL devices incor-porating such luminescent materials when constructed according to the invention show improvements in per-formance and stability over otherwise comparable prior art EL devices.
In the organic EL devices of the invention it is possible to maintain a current density compati-ble with efficient light emission while employing arelatively low voltage across the electrodes by lim-iting the total thickness of the organic luminescent medium to less than 1 ~m (10,000 Angstroms). At a thickness of less than 1 ~m an applied voltage of 20 volts results in a field potential of greater than 2 X 105 volts/cm, which is compatible with efficient light emission. An order of magnitude reduction (to 0.1 ~m or 1000 Angstroms) in thickness of the organic luminescent medium, allowing further reduc-tions in applied voltage and/or increase in the fieldpotential and hence current density, are well within device construction capabilities.
One function which the organic lumlnescent medium performs is to provide a dielectric barrier to prevent shorting of the electrodes on electrical biasing of the EL device. Even a single pin hole extending through the organic luminescent medium will allow shorting to occur. Unlike conventional EL
devices employing a single highly crystalline lumi-nescent material, such as anthracene, for example,the EL devicea of this invention sre capable of fab--rication at very low overall organic luminescent medium thicknesses without shorting. One reason is that the presence of three superimpo~ed layers greetly reduces the chance of pin holes in the layers being aligned to provide a continuous conduction path between the electrodes. This in itself permits one or even two of the layers of the organic luminescent medium to be formed of materials which are not ideally suited for film formation on coating while still achieving acceptable ET. device performance and reliability.

~Z~3~;3~3~

The preferred materials for forming the organic luminescent medium are each capable of fabri-cation in the form of a thin film- that is, capable of being fabricated as a continuou~ layer having a thickness of less than 0.5 ~m or 5000 Angstroms.
When one or more of the layers of the organic luminescent medium are solvent coated, a film forming polymeric binder can be conveniently code-posited with the active material to a~sure a continu-ous layer free of structural defects, such as pinholes. If employed, a binder must, of course, itself exhibit a high dielectric strength, preferably at least about 2 X 106 volt/cm. Suitable polymers can be chosen from a wide varlety of known solvent cast addition and condensation polymers. Illustrative of suitable addltion polymers are polymers and copoly-mers (including terpolymers) of styrene, t-butylsty-rene, N-vinyl carbazole, vinyltoluene, methyl methac-rylate, methyl acrylate, acrylonitrile, and vinyl acetate. Illustrative of suitable condensation poly-mers are polyesters, polycarbonates, polyimides, and polysulfones. To ~void unnecessary dilution of the active material, binders are preferably limited to less than 50 percent by weight, based on the total weight of the m~terlal forming the layer.
The preferred active material~ formlng the organlc luminescent medium are each film forming materials and capable of vacuum vapor deposition.
Extremely thin defect free continuous layers can be formed by v~cuum vRpor depositlon. Speclflcally, lndividual layer thicknesses as low as about 50 Ang-stroms can be present while still realizing satisfac-tory EL device performance. Employing a vacuum vapor deposi~ed porphorinlc compound as a hole in~ecting layer, a film forming aromatic tertiary amine as a hole transporting layer (which can in turn be com-prised of s triarylamine layer and a tetraaryldiamine ~ 39 layer), and a chelated oxinoid compound as an elec-tron in~ecting and transportlng layer, individual layer thicknesses in the range of from about 50 to 5000 Angstroms are contemplated, with layer thick-nesses in the range of from 100 to 2000 Angstromsbeing preferred. It is generally preferred that the overall thickness of the organic luminescent medium be at least about 1000 Angstroms.
The anode and cathode of the organic EL
device can each take any convenient conventional form. Where it is intended to tran~mit light from the organic EL device through the anode, this can be conveniently achieved by coating a thin conductive layer onto a light transmissive substrate -e.g~, a transparent or substantially transparent glass plate or plastic film. In one form the organic EL devices of this invention can follow the historical practice of including a light transmissive anode formed of tin oxide or indium tin oxide coated on a glass plate, as disclosed by Gurnee et al U.S. Patent 3,172,862, Gurnee U.S. Patent 3,173,050, Dresner, "Double In~ec-tion Electroluminescence in Anthracene", RCA Review, Vol. 30, pp. 322-334, 1969; and Dresner U.S. Patent 3,710,167, cited above. While any light transmissive polymeric film can be employed as a substrate, Gillson U.S. Patent 2,733,367 and Swindells U.S. Pat-ent 2,941,104 disclose polymeric films specifically selected for this purpose.
As employed herein the term "light transmis-sive" means simply that the layer or element under dlscussion transmits greater than 50 percent of the light of at least one wavelength it receives and preferably over at least a 100 nm interval. Since both ~pecular (unscattered) and diffused (scattered) emitted light are desirable device outputs, both translucent and transparent or substantially trans-parent materials are useful. In most instances the ~ ~3 light transmissive layers or elements of the organic EL device are also colorless or of neutral optical density - that is, exhibiting no markedly higher absorption of light in one wavelength range as com-pared to another. However, it is, of course, recog-nized that the light transmissive electrode supports or separate superimposed films or elements can be tailored in their light absorption properties to act as emission trimming filters, if desired. Such an electrode construction is disclosed, for example, by Fleming U.S. Patent 4,035,686. The light transmis-sive conductive layers of the electrodes, where fab-ricated of thicknesses approximating the wavelengths or multiples of the light wavelengths received can lS act aR interference filters.
Contrary to historical practice, in one pre-ferred form the organic EL devices of this invention emit light through the cathode rather than the anode. This relieves the anode of any requirement that it be light transmissive, and it is, in fact, preferably opaque to light in this form of the inven-tion. Opaque anodes can be formed of any metal or combination of metals having a suitably high work function for anode construction. Preferred anode metals have a work function of greater than 4 elec-tron volts (eV). Suitable anode metals can be chosen from among the high (> 4 eV) work function metals listed below. An opaque anode can be formed of an opaque metal layer on a support or as a separate metal foil or sheet.
The organic EL devices of this invention can employ a cathode constructed of any metal, including any high or low work function metal, heretofore taught to be useful for this purpose. Unexpected fabrication, performance, and stability advantages have been realized by forming the cathode of a combi-nation of fl low work function metal and at least one other metal. A low work function metal i8 herein defined as a metal having a work function of less than 4 eV. Generally the lower the work function of the metal, the lower the volt~ge required for elec-tron in~ection into the organic luminescent medium.
However, alkali metal~, the lowest work function met-al~, are too reactive to achieve stable EL device performance with simple device constructions and con-struction procedures and are excluded (apart from impurity concentrations) from the preferred cathodes of this invention.
Available low work function metal choices for the cathode (other alkali metals) are listed below by periods of the Periodic Table of Elements and categorized into 0.5 eV work function groups.
All work functions provided are taken Sze, PhYsics of Semiconductor Devices, Wiley, N.Y., 1969, p. 366.
Work Function Period ElementBY eV GrouP
2 Beryllium3.5 - 4.0 3 Magnesium3.5 - 4.0 4 Calclum2.5 - 3.0 Scandium3.0 - 3.5 Titanium3.5 - 4.0 Manganese3.5 - 4.0 Gallium3.5 - 4.0 Strontlum2.0 - 2.5 Yttrium3.0 - 3.5 Indium 3.5 - 4.0 6 Barium ~2.5 Lanthanum3.0 - 3.5 Cerlum 2.5 - 3.0 Praseodymium 2.5 - 3.0 Neodymium 3.0 - 3.5 Promethium 3.0 - 3.5 Samarium 3.0 - 3.5 Europium2.5 - 3.0 Gadollnium3.0 - 3.5 Terbium 3.0 - 3.5 Dysprosium3.0 - 3.5 Holmium 3.0 - 3.5 Erbium 3.0 - 3.5 Thulium 3.0 - 3.5 Ytterbium2.5 - 3.0 Lutetium3.0 - 3.5 Hflfnium ~3.5 7 ~adium 3.0 - 3.5 Actinium2.5 - 3.0 Thorium 3.0 - 3.5 Uranium 3.0 - 3.5 From the foregoing listing lt is apparent that the availabl0 low work function metsls for the most part belong to the Group IIfl or alkaline earth group of metals, the Group III group of met~ls (including the rare earth metals- i.e. yttrium and the lanthanides, but excluding boron and aluminum), ~0 and the actinide groups of metals. The alkaline earth metals, owlng to their ready availability, low cost, ease of handling, and minimal adverse environ-mental impact potential, constitute a preferred class of low work function metals for use in the cathodes of EL devices of this lnvention. Magneslum and cal-clum are partlcularly preferred. Though slgnifi-cantly more expensive, the included Group III metals, partlcularly the rare earth metals, possess slmilar sdvantages and are specifically contemplated as pre-ferred low work function metals. The low work func-tlon metal~ exhiblting work functlons in the range of from 3.0 to 4.0 eV are generally more stable than metals exhibiting lower work functions ~nd are there-fore generally preferred.
~ second metal included in the construction of the cathode has as one primary purpose to increase the stability (both ~torage and operational) of the ~539~

cathode. It can be chosen from among any metal other than an alkali metal. The second metal can itself be a low work function metal and thus be chosen from the metals listed above having a work function of less than 4 eV, with the same preferences above discu~sed being fully applicable. To the extent that the sec-ond metal exhibits a low work function it can, of course, supplement the first metal in facilitating electron in~ection.
Alternatively, the second metal can be cho-sen from any of the various metals having a work function greater than 4 eV, which includes the ele-ments more resistant to oxidation and therefore more commonly fabricated as metallic elements. To the extent the second metal remains invariant in the organic EL device as fabricated, it contributes to the ~tability of the device.
Available higher work function (4 eV or greater) metal choices for the cathode are listed below by periods of the Periodic Table of Elements and categorized into 0.5 eV work function groups.
Work Function Period ElementBY eV GrouP
2 Boron ~4.5 Carbon4.5 - 5.0 3 Aluminum4.0 - 4.5 4 Vanadium4.0 - 4.5 Chromium4.5 - 5.0 Iron 4.0 - 4.5 Cobalt4.0 - 4.5 Nickel ~4.5 Copper4.0 - 4.5 Zinc 4.0 - 4.5 Germanium4.5 - 5.0 Arsenic5.0 - 5.5 Selenium4.5 - 5.0 53~3 Molybdenum4.0 - 4.5 Technetium4.0 - 4.5 Ruthenium 4.5 - 5.0 Rhodium 4.5 - 5.0 Palladium 4.5 - 5.0 Silver 4.0 - 4.5 Cadmium 4.0 - 4.5 Tin 4.0 - 4.5 Antimony 4.0 - 4.5 . Tellurium 4.5 - 5.0 6 Tantalum 4.0 - 4.5 Tungsten ~4.5 Rhenium ~5.0 Osmium 4.5 - 5 0 Irldium 5.5 - 6.0 Platinum 5.5 - 6.0 Gold 4.5 - 5.0 Mercury ~4.5 Lead ~4 0 Bismuth 4.0 - 4.5 Polonium 4.5 - 5.0 From the foregoing listlng of available met-als having a work function of 4 eV or greater attrac-tive h$gher worX function metals for the most part are accounted for aluminum, the Group Ib metals (cop-per, silver, ~nd gold), the metals in Groups IV, V, and VI, and the Group VIII tr~nsltion metals, par-ticul~rly the noble metals from this group. Alumi-num, copper, ~ilver, gold, tin, lead, bismuth, tellu-rium, and antimony are particularly preferred higherwork function second metals for incorporation in the cathode.
There are several reasons for not restrict-ing the choice of the second metal based on either its work function or oxidative stability. The second metal is only a minor component of the cathode. One of it~ primary function~ is to stabilize the first, low work function metal, and, surprisingly, it accom-plishes this ob~ective independent of its own work function and susceptibility to oxidation.
A second valuable function which the second metal performs is to reduce the sheet resistance of the cathode as a function of the thickness of the cathode. Since acceptably low sheet resistance lev-els (less than 100 ohms per square) can be realized at low cathode ~hicknesses (less than 250 Angstroms), cathodes can be formed which exhibit high levels of light tran~mission. This permits highly stable, thin, transparent cathodes of acceptably low resis-tance levels and high electron in~ecting efficiencies to be achieved for the flrst time. This in turn per-lS mits (but does not require) the organic EL devices of this invention to be constructed with light transmis-sive cathodes and frees the organic EL devices of any necessity of having a light transmissive anode to achieve light emission through an electrode area.
A third valuable function which the second metal hus been observed to perform is to facilitate vacuum vapor deposition of a first metal onto the organic luminescent medium of the EL device. In vapor deposition less metal is deposited on the walls of the vacuum chamber and more metal is deposited on the organic luminescent medium when a second metal is also deposited. The efficacy of the second metal in stabilizing organic EL device, reducing the sheet resistance of thin cathodes, and in improving accep-tance of the first metal by the organic luminescencemedium is demonstrated by the examples below.
Only a very small proportion of a second metal need be present to achieve these advantages.
Only about 0.1 percent of the total metal atoms of the cathode need be accounted for by the second metal to achieve a substantial improvement. Where the sec-ond metal is itself a low work function metal, both 3~

the first and second me~als are low work function metals, and it is immaterial which is regarded as the first metal and which is regarded as the second metal. For example, the cathode composition can range from about 0.1 percent of the metal ~toms of the cathode being accounted for by one low work func-tion metal to about 0.1 percent of the total metal atoms being accounted for by a second low work func-tion metal. Preferably one of the two metals account for at least 1 percent and optimally at least 2 per-cent of the total metal pre~ent.
When the second metal is a relatively higher (at least 4.0 eV) work function metal, the low work function metal preferably accounts for greater than lS 50 percent of the total met~l atoms of the cathode.
This is to avoid reduction in electron in~ection efficiency by the cathode, but it is also predicated on the observation that the benefits of adding a sec-ond metal are essentially realized when the second metal accounts for less than 20 percent of the total metal atoms of the cathode.
Although the foregoing discussion hus been in terms of a binary combination of metals forming the cathode, it is, of course, appreciated that com-binations of three, four, or even higher numbers ofmetsls are possible and can be employed, if desired.
The proportions of the first metal noted above can be accounted for by any convenient combination of low work function metals and the proportions of the sec-ond metal can be accounted for any combination ofhigh and/or low work function metals.
While the second metal or metals can be relied upon to enhance electric~l conductivity, their minor proportion of the total cathode metal renders it unnecessary that these metals be present in an electrically conducting form. The second metal or metals can be present as compounds (e.g., lead, tin, 3~

or antimony telluride) or in an oxidized form, such as in the form of one or more metal oxides or salts.
Since the first, low work function metal or metals account for the major proportion of the cathode metal content and are relied upon for electron conduction, they are preferably employed in their elemental form, although some oxidation may occur on aging.
In depositing the first metal alone onto a substrate or onto the organic luminescent medium, whether from solution or, preferably, from the vapor 35~9B

phase, initial, spatially separated deposits of the fir~t metal form nuclei for subsequent deposition.
Subsequent deposition leads to the growth of these nuclei into microcrystals. The result is an uneven and random distribution of microcrystals, leading to a non-uniform cathode. By presenting fl Recond metal during at least one of the nucleation and growth stages and, preferably, both, the high degree of sym-metry which a single element affords is reduced.
Since no two substances form crystal cells of exactly the same habit and size, any second metal reduces the degree of symmetry and at least to some extent acts to retard microcrystal growth. Where the first snd second metals have distinctive crystal habits, spa-tial symmetry is further reduced and microcrystal growth is further retarded. Retarding microcrystal growth favors the formation of additional nucleation sites. In this way the number of deposition sites is increased and a more uniform coating is achieved.
Depending upon the specific choice of metals, the second metal, where more compatible with the substrate, can produce a disproportionate number of the nucleation sites, with the first metal then depositing at these nucleation sites. Such a mecha-nism may, if fact, account for the observation that, with a second metal present, the efficiency with which the first metal is accepted by a substrate is significantly enhanced. It has been observed, for example, that less deposition of the firRt metal occurs on vacuum chamber walls when a second metal is being codepo~lted.
The first and second metals of the csthode are intimately intermingled, being codeposited. That is, the deposltion of neither the first nor second metals is corllpleted before at least a portion of the remaining me~Al is deposited. Simultaneous deposi-tion of the first and second metals is generally pre-S39~

ferred. Alternatively, successive incremental depo-sitions of the first and second metals can be under-taken, which at their limit may approximate concur-rent deposition.
While not required, the cathode, once formed can be given post treatments. For example, the cath-ode may be heated within the stability limits of the substrate in a reducing atmosphere. Other action on the cathode can be undertaken as a conventionally attendant feature of lead bonding or device encapsu-lation.
ExamPles The invention and its advantages are further illustrated by the specific examples which follow.
The term "atomic percent" indicates the percentage of a particular metal present, based on the total number of metal atoms present. In other words, it is ana-logous to mole percent, but is based on atoms rsther than molecules. The term "cell" as employed in the examples denotes an organic EL devicec ExamPleq 1-6 Hue Modification An EL device containing an organic lumines-cent medium satisfying the requirements of the inven-tion was constructed in the following manner:
a) A transparent anode of indium tin oxide coated glass was polished with 0.05 ~m alumina abra-sive for a few minutes, followed by ultrasonic clean-ing in a 1:1 ~volume) mixture of isopropyl alcohol and distilled water. It was rinsed with isopropyl alcohol and then immersed in toluene vapor for about 5 minutes.
b) A hole in~ecting and transporting ATA-l (750 A) layer was then deposited on the anode. ATA-l was evaporated from a quartz boat using a tungsten flla-ment.c) An electron in~ecting and transporting layer (750 A) forming the luminescent zone was then ~ 3 deposited on top of the ATA-l layer. HM-l was employed as a host material and was evaporated from a quartz boat. The fluorescent material to be incorpo-rated in the luminescent zone as a dopant was concur-rently evaporated from a separate quartz boat. Inone instance no fluorescent material was incorpo-rated. Tungsten filaments were employed for both evaporations.
d) On top of the luminescent zone was deposited a 2000 ~ cathode formed of a 10:1 atomic ratio of Mg and Ag.
The shift in hue of light emitted by the organic EL device attributable to the presence of differing fluorescent materials is summarized below in Table I. The power conversion of the organic EL
device (hereinafter also referred to simply as effi-ciency) was measured as the ratio of the power of the light emitted to the electrical power supplied, mea-sured at a light output level of 0.05 mW/cm2. The relative efficiency was determined by dividing the efficiency of the EL device being examined by the efficiency of the corresponding EL device lacking a dopant.
Table I
EL Conc. Rel, Power Device Dopant MolQ_~ Co~v. ~
Control None 0 1 Green Ex. 1 FD-273x10 1 1,5 Orange Ex. 2 FD-281.6xlO 1 2 Red-orange Ex. 3 FD-290.5x10 0.5 Red-orange Ex. 4 FD-302x10 1 0.8 Red-orange Ex. 5 FD-316x10 1 0.6 Red-orange Ex. 6 FD-ll90.3x10 1 o 9 Orange-green From Table I it is apparent that in all instances the presence of the fluorescent dye as a dopant shifted light emission to longer wavelengths.
By compaxing spectra of emitted light with and with-39~3-70-out FD-28 present it was determined that the peak emission was shifted from about 540 nm to 610 nm by the addition of the fluorescent dye. The presence of FD-27 and FD-28 had the further beneficial effect of markedly increasing the power conversion efficiencies of the organic EL devices. The control EL device has an absolute efficiency of 5 X 10 3 W/W.
Examples 7-13 Hue as a Function of Concentration A series of organic EL devices were prepared as described in Examples 1 through 6, using FD-31 at differing concentration levels. The results are sum-marized in Table II.
Table II
Conc. Mole % Rel. Eff. (W/W) Emission ~max 0 1.0 535 2.2x10-2 0.9 648 6.2xlO 2 0 4 640 l.lxlO 1 0.8 645 3.2xlO-1 0.35 666 204.5xlO 1 0.27 665 4.4 0.14 690 From Table II it is apparent that the peak wavelength of emission was shifted over the range of 155 nm. Output efficiencies declined somewhat as the level of fluorescent material was increased. How-ever, output efficiencies measured at 690 nm were actually enhanced as compared to that of the EL
devices containing the undoped HM-l.
Example 14 Stability An EL device containing an organic lumines-cent medium satisfying the requirements of the inven-tion was constructed in the following manner:
a) A transparent anode of indium tin oxide coated glass was polished with 0.05 ~m alumina abra-sive for a few minutes, followed by ultrasonic clean-ing in a 1:1 (volume) mixture of isopropyl alcohol and distilled water. It was rinsed with isopropyl alcohol and then immersed in toluene vapor for about 5 minutes.
b) A hole injecting PC-10 (300 A) was deposited on the anode by vacuum vapor deposition.
PC-10 was evaporated from a quartz boat using a tung-sten filament.
c) A hole transporting ATA-l ~350 A) layer was then deposited on the hole injecting layer. ATA-l was evaporated from a quartz boat using a tungsten filament.
d) An electron injecting and transporting layer (750 A) forming the luminescent zone was then deposited on top of the ATA-l layer. HM-l was employed as a host material and was evaporated from a quartz boat. FD-28 was incorporated in the lumines-cent zone as a dopant at a concentration of 2 mole percent, based on HM-l, by concurrently evaporation from a separate quartz boat. Tungsten filaments were employed for both evaporations.
e) On top of the luminescent zone was deposited a 2000 A cathode formed of a 10:1 atomic ratio of Mg and Ag.
The cell was run at a constant current of 20 mA/cm in a dry argon atmosphere. The initial light output was 0.45 mW/cm2. After continuous operation for 500 hours the light output was 0.15 mWtcm2.
When a cell was constructed and operated as described above, but with FD-2~ omitted, the light output of the cell dropped below 0.15 mW/cm2 after only 250 hours of operation.
~am~ 8 R~ ction Poten~ials and Ba~ga~
A series of EL devices were constructed as described in Example 14, but with differing fluores-cent dyes present. The reduction potentials and bandgaps of the fluorescent dyes successfully employed and the host material HM-l are compared in Table III.

..'`~

~g~3913 Table III
Material E-red volts Bandgap ev _el. Eff. ~W/W) HM-l -1.79 2.81 l.0 FD-4 -1.46 2.5 2.0 FD-5 -1.58 2.58 2.0 FD-7 -1.77 2.75 0.8 FD-15 -1.69 2.63 l.0 FD-19 -1.68 2.64 FD-20 -1.51 2.61 1.4 FD-21 -1.46 2.63 1.0 FD-22 -1.75 2.69 0.8 FD-25 -1.38 2.63 FD-26 -1.66 2.72 FD-27 -1~32 2.17 1.5 FD-28 -1.34 2.01 2.0 FD-119 -1.39 2 0.9 In every instance the EL device containing the fluorescent dye as a dopant to the HM-l layer exhibited a visually detectable shift in hue. The following dyes also produced improvements in effi-ciency: FD-4, FD-5, FD-15, FD-20, FD-27, and FD-28.
When fluorescent dyes were substituted exhibiting either a more negative reduction potential than host material HM-l or a larger bandgap poten-tial, no useful result attributable to the presenceof the fluorescent dye was observed.
APPENDIX I
Preparation of Fluorescent Compound FD-114 n-Butyllithium (7 mmole in 3.5 ml of hexane) was slowly added to a well-stirred, cold (-70~C~
solution of diisopropylamine (1 ml, 7.2 mmole) in dry tetrahydrofuran under argon atmosphere. After stir-ring for five minutes, a solution of 1.5 g, 7.0 mmole of 6-methoxydihydrophenalenone in 20 ml of tetrahy-drofuran was slowly added.

313~3 The resulting dark-colored solution was stirred at -70~C for 1.5 hours. The solution was then transferred using 8 syringe, to a round-bottomed flask containing 5 g (10 mmole) of (2-ethoxy-1,3-pen-tadienyl)triphenylphosphonium iodide prepared accord-ing to the procedure described by Martin et al in J
Or~. Chem. 43, pp.4673-4676 (1978). The resulting suspension was stirred at room temperature for 1 hour and refluxed for 3 hours. All of the steps were car-ried out under argon and the exclusion of moisture.
The suspension was then cooled to room tem-perature, 50 ml of 1 normal hydrochloric acid added and stirred vigorously for 1 hour. Then 50 ml of ether were added and the layers separated. Three additional ether extractions were combined with the first and the ether solution was washed in turn with saturated sodium bicarbonate solution, water, and saturated sodium chloride solution. The ether 901u -tion was dried and the solvent evaporated to yield about 3 g of ~olid residue. The residue was purified via flash chromatography on silica gel using a 10:45:45 ethyl acetate, dichloromethane, cyclohexane mixture as the eluent. The desired product was seen as an orange band when illuminated with a long wave-length (355 nm) ultraviolet lamp. The b~nds contain-ing the orange fluorescent dye were combined, the ~olvents evaporated yielding 860 mg (44% yield) of 4-methoxy-8-methyl-10-oxo-7,8,9,10-tetrahydrobenzo[d,e]-anthr&cene having a m.p. of 135-136C and m/e of 278(M ). The calculated analysis for ClgH18O
was C, 8Z.0 H, 6.2 with the found of C, 81.7, H, 6.3.
A solution of 530 mg (1.9 mmole) of the com-pound identified above, in N,N - dimethylformamide (15 ml), was slowly added to a solution of 700 mg (4.1 mmole) of cupric chloride hydrate and 200 mg (4.7 mmole) of lithium chloride in N,N - dimethylformamide (30 ml) heated to 90C. The resulting mixture was 12~ S~
~ 74-stirred for 70 minutes. Ice was added to the mixture and the resulting brown solid WA5 separated snd washed ~everal times with cold water, yielding 397 mg (80% yield) of Dye 1. This product was ~hown to be pure by thin layer chromatography on silica gel, but was recrystallized ~rom ethyl acetate-ethanol to give a material with a m.p. of 289-295C, and m/e of 260(M ). The structure of the dye was confirmed by the analysis of its N-phenyl-N-methylcarbamate derivative. The calculated analysis for C26Hlg-N03 was C, 79.4, H, 4.9, N, 3.6 and found was C, 79.2, H, 5.1, N, 3.8.
Preparation of Fluorescent Dye FD-116 Cupric chloride dihydrate (2.45 g, 14.4 -mmole) and lithium chloride (1.0 g, 22.7 mmole) were suspended in 20 ml of N,N-dimethylformamide (DMF) heated to 90C. To this hot mixture was added 8 solution of 650 mg, (2.34 mmole) of the methoxy ketone intermediate of Example 1 dissolved in 10 ml of DMF. The mixture was kept at 90C for 24 hours and then quenched by adding ice and water. The resulting precipitate was wa~hed several times with water and dried. It was purified by triturating with 10~ methanol in 1:1 ethyl acetate/dichloromethane, yielding 300 mg (44%) o~ Dye 3. The product from the trituration step ls pure enough for the lntended appllcatlons. It can be further purlfied by flash chromatography on silica gel using 20% ethyl acetate ln a 1:1 dlchloromethane/cyclohexane mlxture a~ elu-ent, ylelding material wlth m.p. of 238-240C. The structure of Dye 3 was confirmed by the elemental analysls of both its methyl ether and its N-phenyl-N-methylcarbamate derivatives.
Analysis for methyl ether derivative Clg-H13C102: Calcd: C, 73.9, H, 4.2. Found: C, 74.0, H, 4.1.

3~3 Analysis for csrbamate derivative C26-H18ClN03: Calcd: C, 73.0, H, 4.2, N, 3.3. Found:
C, 72.8, H, 4.1, N, 3.1.
PreParation of Fluorescent DYe FD-118 N-phenyl-N-methylcarbamoyl chloride (1.2 equivalents) was added to a mixture of 1.2 equiva-lents each of pyridine and Dye 3 of Example 2 in toluene as solvent and the mixture heated to reflux for 12 hours. The toluene solution was cooled, washed with dilute hydroch~oric acid, then water, and finally brine. The organic phase was separated and the solvent evaporated. The residue WAS purified by flash chromatography on sillca gel using 20% ethyl acetate in 1:1 dichloromethsnetcyclohexane as the eluent. Dye 5 has m.p. of 233-235C and the elemen-tal analysis given for the carbamate in Example 2.
PreParation of Fluorescent DYe FD-119 The methoxy ketone intermediate of Example 1 (400 mg, 1.4 mmole) dissolved in fl small amount of dichloromethane was added to 5 ml, 59.8 mmole of pyr-rolidine in 100 ml of anhydrous methanol and the mix-ture refluxed for four days with stirring. The mix-ture was cooled and the solvents evaporuted under reduced pressure. The residue was purified by flash chromatography on silica gel using 1:1 dichlorometh-anetcyclohexane contsining ethyl scetate ranglng from 0% at the start to 50% at the end of the reaction.
The appropriate fractions were combined and the sol-vents evaporated. The residue was triturated with ethyl acetate and filtered, giving Dye 6 in 44%
yield, with m.p. of 244-246~C, m/e 313 of (M ).
The invention has been described ln detail with particular reference to preferred embodiments thereof, but it will be understood that variations and modifications can be effected within the spirit and scope oF the invention.

Claims (17)

1. An electroluminescent device comprising in sequence, an anode, an organic hole injecting and transporting zone, a luminescent zone, and a cathode, characterized in that said luminescent zone is formed by a thin film of less than 1 µm in thickness comprised of an organic host material forming a layer capable of sustaining both hole and electron injection and located in said layer as a fluorescent material a dye capable of emitting light in response to hole-electron recombination, said dye having a bandgap no greater than that of said host material and a reduction potential less negative than that of said host material.

2. An electroluminescent device according to claim 1 in which said layer is located in contact with said organic hole injecting and transporting zone.

3. An electroluminescent device according to claim 1 in which said cathode is comprised of a metal other than an alkali metal having a work func-tion of less than 4 eV.

4. An electroluminescent device according to claim 1 in which said organic hole injecting and transporting zone is comprised of a layer in contact with said anode containing a hole injecting porphyrinic compound and a layer containing a hole transporting aromatic tertiary amine interposed between said hole injecting layer and said luminescent zone.

5. An electroluminescent device according to claim 1 in which said dye is chosen from the class consisting of coumarin, dicyanomethylenepyrans and thiopyrans, polymethine, oxabenzanthracene, xanthene, pyrylium and thiapyrylium, carbostyril, and perylene fluorescent dyes.

6. An electroluminescent device according to claim 5 in which said fluorescent dye is a cou-marin dye of the formula:
where R1 is chosen from the group consisting of hydro-gen, carboxy, alkanoyl, alkoxycarbonyl, cyano, aryl, and a heterocylic aromatic group, R2 is chosen from the group consisting of hydro-gen, alkyl, haloalkyl, carboxy, alkanoyl, and alkoxy-carbonyl, R3 is chosen from the group consisting of hydro-gen and alkyl, R4 is an amino group, and R5 is hydrogen, or R1 and R2 together form a fused carbocyclic ring, or the amino group forming R5 independently com-pletes with at least one of R4 and R6 a fused ring.

7. An electroluminescent device according to claim 5 in which said fluorescent dye is a dye of the formula:
where X represents oxygen or sulfur;
R6 represents a 2-(4-aminostyryl) group; and R7 represents a second R6 group, an alkyl group, or a phenyl group.

8. An electroluminescent device according to claim 1 in which said organic luminescent zone is less than 5000 Angstroms in thickness.

9. An electroluminescent device according to claim 8 in which said host material is capable of emitting light of a first wavelength in the absence of said fluorescent material and said fluorescent material is capable of absorbing light at the first wavelength.

10. An electroluminescent device according to claim 1 in which said cathode is comprised of a metal other than an alkali metal having a work func-tion of less than 4 eV and is light transmissive.

11. An electroluminescent device according to claim 10 in which said cathode is formed of a com-bination of metals other than alkali metals having a work function of less than 4 eV.

12. An electroluminescent device according to claim 10 in which said cathode additionally includes less than 50 atomic percent of a second metal having a work function of greater than 4 eV.

13. An electroluminescent device according to claim 8 in which said organic hole injecting and transporting zone is comprised of a layer in contact with said anode containing a hole injecting porphyrinic compound and a layer containing a hole transporting aromatic tertiary amine interposed between said hole injecting layer and said luminescence zone.

14. An electroluminescent device according to claim 8 in which said host material is a chelated oxinoid compound.

15. An electroluminescent device according to claim 14 in which said chelated oxinoid compound is represented by the structure:

wherein Mt represents a metal;
n is an integer of from 1 to 3; and Z2 independently in each occurrence represents the atoms completing a nucleus having at least two fused aromatic rings.

16. An electroluminescent device according to claim 1 in which said anode is formed of one or more metals having a work function of greater than 4 eV, said cathode is formed of a combination of metals, with one or more metals other than alkali metals having a work function of less than 4 eV
accounting for greater than 50 percent on an atomic basis of the total cathode metals, and, interposed between said anode and cathode, said organic luminescent medium is less than 5000 Angstroms in thickness, and said hole injecting and transporting zone is comprised of a tetraphenyldiamine layer contacting said electron injecting and transporting zone, a hole injecting porphyrinic compound con-taining layer contacting said anode, and a triphenylamine layer interposed between said tetraphenyldiamine layer and porphyrinic com-pound.

17. An electroluminescent device according to claim 1 in which said cathode contains magnesium and a minor amount of another metal chosen from the class consisting of aluminum, copper, silver, gold, tin, lead, bismuth, tellurium, indium, and antimony, at least one of said anode and said cathode is light transmissive, said organic luminescent medium is less than 5000 Angstroms in thickness and is comprised of a hole injecting layer containing a phthalocya-nine in contact with said anode, a first hole transporting layer containing a bis-triphenylamine, a second hole transporting layer containing a tetraphenyldiamine, and said luminescent zone containing a metal chelated 8-hydroxyquinoline host material and a fluorescent dye capable of absorbing light at the emission wavelength of said metal chelated 8-hydroxyquinoline, having a lower bandgap and a less negative reduction potential than said chelated 8-hydroxyquinoline.