Electroluminescent Device
The invention relates to an electroluminescent device in particular such a device including a luminescent molecule such as a polymer.
Known electroluminescent devices comprise organic light-emitting diodes. These arrangements comprise films of molecular semiconductors or semiconducting polymers sandwiched between electrodes in solid state. The devices are discussed in, for example, J.H. Burroughs, D.D.C. Bradley, A.R. Brown, R.N. Marks, K. Mackay, R.H. Friend, P.L. Burns and A.B. Holmes, Nature (London) 347,539 (1990).
Luminescent polymers have attracted widespread scientific and commercial interest owing primarily to their potential applications in thin-film solid-state emissive devices. Films of luminescent polymers are highly promising from a commercial perspective because they may be deposited directly from solution. Additionally, as these are in solid-state, they provide intrinsic strength and rigidity thereby obviating the need for additional support structures within devices.
There are various problems known with the existing devices, however. The closed structures of the known solid state devices trap significant amounts of heat. Localised heating has a detrimental effect on device stability as a result of thermal cycling effects and accelerated chemical degradation. Furthermore when the active layer in a solid state device degrades for whatever reason, or otherwise requires replacement, the device as a whole has to be removed and
replaced which is costly, wasteful, and gives rise to an unnecessarily long down time.
Limited research has previously been undertaken into solution-based electrochemiluminescent devices using emissive molecules. However in these arrangements an additional electrolyte is needed to enable charge injection. Examples of these are electrochemical cells as reported previously by for example [L. R. Faulkner and A. J. Bard, "Electrogenerated Chemiluminescence. I. Mechanism of Anthracene Chemiluminescence in N,N-Dimethylformamide Solution," J. Am. Chem. Soc, 90, 6284 (1968)]. The requirement for an ionic electrolyte in these devices significantly constrains freedom in the choice of solvent and requires dilution of the luminescor. Furthermore the electrolyte materials required tend to be reactive which can affect operating stabilities. Yet further, because the charge carrier mechanism relies on ion mobility, the relative slowness of ion motion gives rise to slow turn-on speeds.
An alternative form of electrochemiluminescent devices are solvent- mediated/solution phase electrochemical devices (SPECLDs) using conjugated polymers in which transport of the majority carrier is achieved by using an easily oxidised or reduced solvent of low molecular mass (sometimes termed a "redox-active" solvent). The polymer layer of a conventional solid state device is replaced by a high concentration solution of the emissive polymer dispersed in the solvent. SPECLDs exhibit relatively high quantum efficiencies (approximately 1% photons/electrons) at modest applied fields of approximately 106 Vm"1 but are found to be highly sensitive to the choice of solvent, in particular requiring the use of easily oxidisable (or reducable) solvent molecules for their operation. In S.C. Chang, Y. Li, Y. Yang, Journal of
Physical Chemistry B104 (2000) Chang et al an electrochemiluminescence (ECL) based mechanism for device operation is proposed in which injection and transport of the majority carriers is mediated by the solvent. The solvent is assumed to undergo oxidation at the anode to form small radical cations with the polymer undergoing reduction at the cathode to form bulky (and relatively immobile) radical anions (negatively charged polarons). The mobile cations are assumed to drift through the bulk of the device and on encountering a radical anion of the luminescent polymer transfer their charge producing an excited state chain segment capable of radiative decay.
In these devices the role of the solvent as a charge transport medium is fundamental to the operation of the devices. The solvent must form radical ions readily which constrains freedom in the choice of solvent . Furthermore the electrolyte solvent is by necessity reactive which can affect operating stabilities. Again, because the charge carrier mechanism relies on ion mobility, the relative slowness of ion motion gives rise to slow turn-on speeds.
Although electrochemiluminescent devices have been reported based on emissive metal chelates which do not require a redox-active solvent or an external electrolyte, these devices still rely on ion mobility. This is because the metal chelates are salts which dissociate to form ionically conductive solutions, hence serving as intrinsic electrolytes.
According to the invention there is provided a luminescent device comprising an active electroluminescent medium in liquid phase acting as a charge transport medium and a method as substantially as described herein and as illustrated in the figures. Preferred features are set out in the dependent claims.
The invention gives rise to a range of advantages. Because it is an electroluminescent (as opposed to electrochemiluminescent) device, neither an electrolyte nor a redox-active solvent is required for its operation. Obviation of the electrolyte provides considerable freedom in the choice of solvent and allows the luminescor to be dissolved to higher concentrations than is otherwise possible (indeed in appropriate circumstances - e.g. if the molecular semiconductors are in liquid phase - a solvent may not be required at all). The host solvent used can be a chemically inert non-polymer aromatic solvent — preferably toluene, which does not in general dissolve ionic salts -, leading to potential improvements in operating stabilities. From a physical perspective, elimination of the electrolyte achieves various advantages. In particular the invention relies on electronic rather than ionic mobilities allowing significantly increased switching speeds enabling the devices to be driven at high frequency in pulsed operation.
Furthermore, by virtue of provision of a liquid phase active medium the medium can be continuously flowed allowing efficient thermal-sinking to ensure rapid elimination of excess heat. The continuous-flow device further provides an alternative means of reducing the local density of electrically and optically induced defects which can otherwise reduce electroluminescence efficiencies. Yet further the active medium can be replaced wholesale simply by draining or pumping the liquid phase away.
Embodiments of the invention will now be described, by way of example, with reference to the figures of which:
Fig. 1 is a perspective view of a first embodiment of the device according to the present invention;
Fig. 2 is a perspective view of the device according to another embodiment of the invention; Fig. 3 is a schematic view of a device including a liquid cell according to the present invention;
Fig. 4a is an exploded view of a multiple flowstream device according to another embodiment of the invention;
Fig. 4b is a perspective view of the multiple flow stream device of Fig. 4a showing, schematically, flow through the capillary;
Fig. 5 shows a schematic view of an analysis implementation of the invention using laminar flow;
Fig. 6 shows a plot of intensity against energy for a photoluminescent and electroluminescent device; and Fig. 7 shows a plot of current density against voltage for two dummy devices containing a redox active and redox inactive solvent and a device according to the present invention.
Fig. 1 shows a device according to the invention in its simplest form. The device 10 includes electrodes 12,14 and an active liquid layer 16 sandwiched therebetween. The electrodes can be of any preferred type for example gold electrodes with lOμm spacing between anode and cathode. The active layer is a one-percent by- weight solution of poly(9,9-dioctyl fluorene) (PFO) in spectroscopic grade toluene. In this solution-phase electroluminescent device the solvent plays no part in the charge transport process (other than as a host for the dissolved polymer). Emission arises from direct bipolar carrier injection from the electrodes into the carrier bands of the dissolved polymer.
Accordingly, in a device according to the invention, the (non-ionic) semiconducting polymer is the only solute and there is no electrolyte present (intrinsic or otherwise). Additionally, and in contrast generally with electrochemiluminescent devices, positive and negative charge carriers do not move freely through the solution but hop between sites on a large and relatively immobile high molecular mass host-polymer. Nor is an easily oxidised or reduced low molecular weight host solvent required.
The device can be driven either under steady state or under pulsed conditions. Pulsed operation of organic LEDs (generally at relatively low duty cycles) is known to minimise non-radiative decay channels (such as polaron induced fluorescence quenching) by lowering the time averaged density of electrically induced defects. For example known solid-state LEDs are able to achieve instantaneous brightnesses of 10 cdm" under pulsed operation and similar brightnesses are believed to be achievable in liquid phase devices of the type according to the present invention with a sub microsecond response time owing to the relatively high electronic mobilities. Any appropriate pulsed drive arrangement may be used as will be known to the skilled reader. It will be noted that the drive voltages required are low as a result of the thinness of active layer available in the liquid phase. As charge is transported only in the form of polarons, drive voltages of- 200 V are needed to achieve appreciable current injection and light emission; broadly in line with the typical field- strengths required for balanced bipolar charge injection in conventional thin- film organic LEDs.
The active medium can be retained in a static cell or can be subjected to flow between the electrodes either in continuous or pulsed mode. In the case of a
static cell, any appropriate inlet/outlet valve can be included to allow the cell to be drained and the active medium to be replaced. As a result the active medium can be changed either when different operating characteristics are required or where it has decayed or otherwise deteriorated operationally.
Referring to Figs. 2 and 3 a device capable of either static cell or continuous/pulse flow operation is shown. The device comprises a planar surface microfluidic chip device and is again designated generally 10. It includes inter-digitated electrodes 18 on a substrate 20, an inlet valve 24 and outlet valve 26 defining between them a liquid cell 28 in which the active medium is retained. A reservoir 30 contains the active medium to replenish that in the cell 28 as appropriate. A pump (not shown) can be provided to enable replenishment. A drain reservoir 32 receives used or exhausted active medium. In continuous flow operation the valves 24 and 26 are retained open or removed altogether and a continual flow of active medium passes through the flow cell
28. If desired a recycling circuit 34 can be provided between the inlet and drain reservoirs 30,32.
In the continuous or pulsed flow arrangement the polymer solution is passed rapidly through the active device (i.e. between the opposing faces of the electrodes) allowing heat to be dissipated out of it. The continuous or pulsed flow can be governed by any appropriate pump mechanism. Fluid extractors can also be incorporated within the device to provide a further heat sink.
Continuous flow further serves as a method of improving device lifetimes. To the extent that lifetimes are dictated by the production of molecular species in normal devices fluidics allow the introduction of scavenger molecules that can
mop up possible contaminants. Also flow can be useful to reduce surface adsorption processes which may affect LED functioning.
Solution LEDs are of particular interest in the context of microfluidic devices using micro cavities. In these, diametrically opposed SLEDs and photocells are combined to form simple chemical sensors, and can be constructed to offer a plurality of excitation wavelengths. For instance, they may be integrated into electrophoretic capillaries in a substrate 40 as shown in Figs. 4a and 4b. In a simple implementation, the capillary 41 is filled with a variety of emissive polymers 42 in solution phase (which may or may not contain appropriate molecular dopants to modify the emission characteristics), each dissolved in an appropriate solvent. The individual polymer solutions are separated by immiscible solvent spacers 44 to prevent intermixing. The different polymers, which are typically chosen to offer excitation wavelengths over the full visible and near IR spectrum may be sequentially selected by applying an appropriate electric field parallel to the capillary axis using a pixel-type electrode configuration 46. The SLED in this embodiment functions as a highly compact integrated multi-wavelength excitation source for Lab-on-a-Chip and other analytical applications with a photocell bank for example at 48. In general any fluidic approach has the advantage over solid-state devices of flexibility in configuration and operation. In an alternative preferred embodiment the SLEDs are integrated into microfabricated fluidic channels.
In a further embodiment, referring to Fig. 5, laminar flow properties can be exploited such that solvent spacers are not required, using immiscible solvents in an alternating fashion. Laminar flow is almost always observed within micron sized channels due to low Reynolds numbers. This essentially means
that fluid streams remain well defined as they move downstream (i.e. no turbulence is observed). This allows for the possibility of defining light source emission from individual streams within a single channel. Due to laminar flow, even miscible fluidic streams could be used over defined distances. This flow allows the definition of very thin solvent flows, of the order of micron thickness. Using microfluidic assemblies, it is possible to bring the luminescent polymer 60 into direct physical contact with a solution 62 containing an analyte under study. In the first instance, this should permit efficient coupling of the light into the sample, especially if care is taken to match refractive indices. Indeed, by exploiting laminar flow in the microcapillaries, which prevents intermixing of the flow streams on the time-scale of the measurement, coupling efficiencies of 100% are in principle attainable, with the entire forward-moving photon flux entering the sample volume.
Various applications for the invention are available. For example optical confinement effects may be exploited to improve the spectral properties of the excitation source and detector. Semiconductor polymers (SPs) typically have broad emission and absorption properties. For the purposes of optical detection, a narrow excitation source is desirable. In order to improve the monochromaticity of excitation light, the LED is enclosed in a simple vertical microcavity of appropriate optical pathlength. High quality cavities may be fabricated using alternative thermal deposition of materials with sharply differing refractive indices to form simple low-cost Bragg reflectors and tuning capabilities. Microcavity-enclosed SLEDs operating in pulsed mode offer a highly promising means of achieving electrically-induced lasing.
Solution LEDs - both steady and continuous flow versions - are also of importance for applications requiring occasional and wholesale exchange of the emissive medium, e.g. lighting applications. Most street lights for example use relatively short-lived bulbs which are replaced on a periodic basis. In order to replace the bulb, physical access is required (typically via a ladder or crane) and replacement is therefore a slow inconvenient process. In SLED devices, failure of the bulb is analogous to "poisoning" of the emitter. The electrodes and housing can be fabricated from stable materials, allowing the lamp to be restored simply as discussed with reference to Fig. 4 by replacing the luminescent polymer. Moreover, since the polymer is used in solution, pipe work could provide a simple and direct physical connection to the remote lamp. In principle, it is therefore a simple matter to replenish the lamp from street level, lowering maintenance costs markedly. The same principle can be applied to any other appropriate illumination device, for example, illuminated road signs.
The physical characteristics of SLEDs according to the present invention can be seen with reference to Figs. 6 and 7. In particular the device according to the present invention including a solution of PFO in toluene was compared against non-emissive dummy devices using five L[L droplets of spectroscopic grade solvent in place of the polymer solution. This also allows an investigation of the charge transport properties of the host solvent. Emission spectra were recorded using a calibrated spectrograph.
Figure 6 shows in situ photoluminescent (PL) and electroluminescent (EL) spectra for the dissolved polymer in toluene. Similarity between the spectra indicates that emission arises from the same polymer species in both instances;
the differences in the relative magnitudes of the phonon peaks are attributed to cavity effects and differing recombination profiles.
Figure 7 shows current-voltage characteristics for two dummy devices containing respectively toluene and 1,2-dichlorobenzene (DCB) (as used in the SPECLDs), and for an emissive device containing toluene and PFO. As expected, oxidation of toluene is unfavourable and the corresponding device is highly resistive. The resistance of the DCB-based device is considerably lower reflecting the lower ionisation energy of the solvent. When PFO is added to the toluene to form the emissive device, the rate of electronic carrier injection increases rapidly. Because the solvent is unable to sustain a substantial electronic current itself, electronic injection occurs directly from the electrodes into the carrier bands of the polymer. The observation of EL indicates that the injection process is bipolar. The operating mechanism is therefore essentially the same as that for a conventional solid-state organic light-emitting diode.
The drive voltages of the EL devices according to the invention are considerably higher than those reported for ECL devices. In those devices, it is suggested that the lightweight cations of the solvent are free to move through the bulk, whereas the negative polarons formed at the cathode are relatively immobile. The cations therefore drift through the bulk of the device before transferring their charge to a negative polymer chain.
The electric field strengths required for the present invention, of the order of 10 V/m, suggest that polaron mobilities in the solution-phase devices are comparable to the corresponding mobilities in solid-state despite the wider (time-averaged) spacing of individual chains. It is anticipated that there will be
three main contributions to the conductivity: hopping between sites on a single polymer chain; modest drift diffusion, and convection of charged polymer chains; and charge transfer between chains (which depends in turn on their spacing and segmental motion).
In the known electrochemiluminescent arrangement solvents are chosen with low ionisation energies (such as cyclohexanone and 1,2-dichlorobenzene) in order to ensure facile oxidation of the solvent. Toluene by contrast is an extremely resistive solvent with a conductivity of 10" S/m or less and oxidation is highly unfavourable. This, for example, creates considerable difficulties in the field of non-aqueous solution electrochemistry where the low conductivity of toluene gives rise to severely distorted voltammograms because of the large potential drop between working and reference electrodes. It is also worth noting that, because toluene is extremely non-polar, only a very small number of salts will dissolve at room temperature (all of which are likely to be organic in nature). The probability of accidental contamination with an impurity salt is therefore low and, unless a supporting electrolyte is deliberately introduced into the solution, it will remain electrochemically inactive.
It will be appreciated that the device can be altered in various ways without departing from the invention. For example although PFO is discussed as the active medium any appropriate luminescence and semi-conducting polymer or molecule can be used for example other derivatives of the polyfluorene or poly (p-phenylene-vinylene) families. Similarly any appropriate solvent can be selected and in addition to toluene, benzene, heptane or other low permittivity highly resistive solvent can be used.
Similarly, any appropriate electrode configuration and drive/pump arrangement can be adopted.