CN1809935B - Fabrication of polymer devices - Google Patents

Abstract

A method of forming a polymer device includes the steps of: (i) depositing on a substrate a solution comprising a polymer or oligomer and a crosslinking moiety, to form a layer; (ii) curing the layer formed in step (i) under conditions to form an insoluble crosslinked polymer; characterised in that the crosslinking moiety is present in step (i) in an amount in the range of from 0.05 to 5 mol % based on the total number of moles of repeat units of the polymer or oligomer and the crosslinking moiety in the solution.

Description

Fabrication of polymer devices

The present invention relates to methods of fabricating polymeric devices. In particular, the present invention relates to methods of depositing polymer layers in methods of fabricating polymer devices. The invention also relates to devices, such as electronic and optoelectronic devices, prepared by the method of the invention.

Polymer devices include polymer light-emitting diodes (LEDs), photodetectors, photovoltaics (PVs) and field-effect transistors (FETs). Such devices typically include one or more semiconducting polymer layers positioned between electrodes. Semiconducting polymers are characterized by partial or substantial π-conjugation in the backbone or side chains.

Semiconducting polymers are now commonly used in many optical devices, such as for polymer light emitting diodes ("PLEDs") disclosed in WO 90/13148; field effect transistors ("FETs"); light emitting diodes disclosed in WO 96/16449. a potentiotropic device; and a photodetector disclosed in US 5,523,555.

A typical PLED includes a substrate, an anode carried on the substrate, a cathode, and an organic electroluminescent layer disposed within the anode and cathode and comprising at least one polymeric electroluminescent material. In operation, holes are injected into the device through the anode, and electrons are injected into the device through the cathode. The holes and electrons combine in the organic electroluminescent layer to form excitons, which then undergo radiative decay to give light. Other layers may be present in PLEDs. For example a layer of organic hole injection material such as poly(ethylenedioxythiophene)/polystyrene sulfonate (PEDT/PSS) can be provided between the anode and the organic electroluminescent layer to assist hole injection from the anode into the organic electroluminescent layer.

Transistors, specifically field effect transistors (FETs), are three-terminal devices that include a source contact, a drain contact, and a gate contact. A semiconductor layer (channel) bridges the source and drain contacts and is itself spaced from the gate contact by an insulating layer called a gate dielectric. In polymer transistors, the semiconducting layer is fabricated from a semiconducting polymer, typically a π-conjugated organic polymer. This layer can be deposited in the device via the precursor route or directly via solution-processing.

A voltage is applied between the source and drain contacts. Furthermore, in field effect transistors, a voltage is applied to the gate contact. This voltage creates an electric field that causes the accumulation or dissipation of charge carriers in the semiconductor layer directly beneath the gate dielectric. This in turn controls the current flow from the source to the drain contact at a given source-drain voltage.

For a phototransistor, light of the appropriate wavelength is shone on this channel. The photons can generate hole-electron pairs that split and contribute to the generation of current flowing between the source and drain, thereby modulating the source-drain conductivity.

As described in WO 96/16499, a typical photovoltaic device includes a photoresponsive section having first and second major surfaces and providing the first and second electrodes. The photoresponsive segment includes a first semiconducting polymer optionally blended with a second semiconducting polymer that is phase separated from the first semiconducting polymer. In the short circuit state, an internal electric field exists in the photoresponsive section. The orientation of the internal electric field causes electrons to migrate to and be collected on the junction with the lowest work function (typically an aluminum, magnesium, or calcium electrode), while holes move toward an electrode with a higher work function, such as an ITO electrode . Thus, the photocurrent can be detected and used, for example, to provide electrical energy (as is the case for example with solar cells), or to allow the detection of a light pattern such as part of an image for use in an image sensor.

As discussed in US 5,523,555, a typical photodetector device includes a photoresponsive layer arranged between first and second electrode layers having different work functions. The photoresponsive layer comprises a semiconducting polymer or a blend of such polymers. There can be multiple photoresponsive layers. A bias circuit is connected to apply a bias voltage between the first and second electrode layers. A sensing circuit is connected to detect a photocurrent flowing across the polymer layer between the first and second electrode layers due to radiation incident on the polymer layer while a bias voltage is applied. The bias voltage is selected in dependence on the distance between the electrodes.

Semiconducting polymers can exhibit a wide variety of photophysical properties (such as π-π* bandgap and photoluminescence yield); optical properties (such as refractive index and its dispersion); electronic properties (such as hole- and electron- - transport energy levels, and hole- and electron-mobility); and processing properties (such as solvent solubility, phase transition temperature, crystallinity and phase transition temperature). These properties are mainly controlled by the chemical structure of the polymer. In this regard, these properties can basically be controlled within a certain range by suitable selection of the backbone units and side chains of the polymer.

One or more polymers in the aforementioned polymeric devices are preferably soluble in common organic solvents in order to facilitate their deposition during device fabrication. One of the key advantages of this solubility is that polymer layers can be fabricated by solution processing methods such as by spin-casting, inkjet printing, screen printing, dip coating and the like. Examples of such polymers are disclosed, for example, in Adv. Mater. 2000 12 (23) 1737-1750 and include those having aromatic or heteroaromatic units such as fluorene, indenofluorene, phenylenes, arylene Polymers of vinylidene, thiophene, pyrrole, quinoxaline, benzothiadiazole, oxadiazole, thiophene, and arylamine containing solubilizing groups at least partially conjugated backbone, and polymers with non-conjugated backbones such as poly(vinylcarbazole). Polyarylenes such as polyfluorenes have good film-forming properties and can be easily formed by Suzuki or Yamamoto polymerization methods, allowing a high degree of control over the regioregularity of the resulting polymer.

In some devices it is desirable to cast multiple layers of different polymers (ie, laminates) on a single substrate surface. This enables, for example, the optimization of different functions, such as electron or hole charge transport, luminescence control, photon-confinement, exciton-confinement, light-induced charge generation, and charge confinement or storage.

In this regard, it can be used to fabricate multilayers of such polymers to control, for example, electrical and optical properties across the polymer stack. This is useful for optimal device performance. Optimal device performance can be achieved, for example, by careful design of electron and hole transport level shifts, optical refractive index mismatches, and energy gap mismatches across the interface. Such hybrid structures can, for example, facilitate the injection of one carrier but block the extraction of the opposite carrier and/or prevent the diffusion of excitons to the quenching interface. Therefore, such hybrid structures can provide useful carrier and photon confinement effects.

It can also be used to fabricate multiple layers to provide a protective layer over the device structure. In this regard, as an example, PEDT/PSS is believed to have deleterious effects on the electroluminescent layer of a PLED. Without wishing to be bound by theory, it is thought that this may be due to an electrochemical reaction between the PEDT:PSS layer and the electroluminescent layer (ie, the layer in which holes and electrons combine to form excitons). This is generally believed to result in a quenching of the luminescence and an incremental increase in the required drive voltage. Therefore, it is desirable to provide a protective layer between the PEDT:PSS and the electroluminescent layer.

However, the preparation of the polymer laminate is generally not trivial. In particular, the solubility of the initially cast or deposited layer in the solvent used in the subsequent layer can pose problems. This is because solution deposition of subsequent polymer layers can dissolve and destroy the integrity of previous layers.

One option to overcome this problem is to use precursor polymer systems. Precursor systems for PPV (polyphenylene vinylene) and PTV (polythienyl vinylene) are known in the art.

Layers of semiconducting polymers can be formed by depositing a soluble polymer precursor followed by chemical conversion to an insoluble electroluminescent form. For example, WO94/03030 discloses a process in which an insoluble, electroluminescent poly(phenylene vinylene) is formed from a soluble precursor and further layers are then deposited from solution onto this insoluble layer.

However, it is clearly not desirable to limit the polymers in a polymer device to those types of polymers formed from insoluble precursor polymers. Furthermore, the required chemical conversion process of the precursor polymers involves extreme processing conditions and reactive by-products that can impair the performance of the preceding layers in the finished device.

An additional option to overcome this problem is to use polymers that differ to a large extent in their dissolution behavior. For example, the use of polymers soluble in hydrocarbon solvents in combination with polymers soluble in water or acetate solvents allows the preparation of limited bilayer or trilayer stacks. An important example in this regard is the deposition of conjugated polymers from aromatic hydrocarbon solvents onto first formed conductive PEDT:PSS coatings that are insoluble in aromatic hydrocarbon solvents. Again, this severely limits the types of polymers that can be used in multilayer stacks. This is because most conjugated polymer systems are characterized by solubility in the same set of aromatic hydrocarbon solvents (such as xylenes and other substituted benzenes) and slightly polar hydrocarbon solvents (such as tetrahydrofuran and halogenated solvents).

It should be appreciated that limitations on the polymers that can be used in laminates mean that many concepts of device structures cannot be studied or implemented. As such, further development of the device hierarchy can be severely hampered.

WO 96/20253 generally describes luminescent, film-forming, solvent-processable polymers containing crosslinks. It is claimed that because the film is resistant to dissolution in common solvents, this allows for the deposition of additional layers, thereby facilitating device fabrication. The use of azido groups attached to the polymer backbone is mentioned as an example of thermal crosslinking. In the general formula shown on page 7, the polythiophene copolymer contains a repeating unit having a moiety for crosslinking in a content between 5 mol % and 66 mol %.

US 6107452 discloses a method of forming multilayer devices in which fluorene-containing oligomers comprising terminal vinyl groups are deposited from solution and cross-linked to form an insoluble polymer requiring deposition of additional layers on top of it. The vinyl units must be present in a molar ratio higher than at least 25%. This can be determined by the mechanism in US6107452 via which crosslinking occurs.

Similarly, Kim et al., Synthetic Metals 122 (2001), 363-368 disclose polymers comprising triarylamine groups and ethynyl groups, which can be crosslinked after deposition of the polymer. This document discloses the presence of acetylene groups present as part of the repeating unit of the polymer at 100 mol%.

In addition to being used to form polymer laminates, making the polymer insoluble after deposition also allows for negative-tone lithography. According to standard negative-tone photoresist lithography, a polymer film, usually polystyrene or poly(methyl methacrylate), containing a critical loading of a crosslinked system is cast onto a substrate. Selected areas are exposed via a mask pattern. The exposed areas become insoluble through a cross-linking reaction. The unexposed areas remain soluble and can be subsequently removed by washing with the developer solvent system, resulting in the transfer of the negative image of the mask onto the photoresist coating.

In this regard, it is desirable to form full-color displays by patterning red, green and blue electroluminescent materials. It is also desirable to pattern and separate the individual FET or photodetector elements formed on the substrate. Known methods include patterned deposition of solution processable substances, especially polymers, by means of inkjet printing as disclosed in eg EP 0880303 or by evaporation of vaporizable substances through shadow masks.

One approach is to introduce polymerizable moieties in the side chains of the polymer and initiate the polymerization (crosslinking) reaction via a radiation-sensitive initiator. An example of this approach is described in Nature 421, 829-833, 2003, which discloses the formation of full-color displays by deposition of layers of red, green and blue electroluminescent polymers containing pendant oxetane groups In this method, the polymer is crosslinked after deposition by exposure to suitable radiation via a photoacid generator. Oxetane-containing monomers were incorporated into the exemplified polymers at a level of 25 mol%. Exposing only selected areas of each layer to UV radiation through a mask causes the selected areas to crosslink. The polymer in the unexposed areas remains soluble and can therefore be washed away leaving a patterned layer. However, this disclosed method results in polymers with reduced electron and hole mobility. In addition oxetanes are prone to self-polymerize during storage in the presence of traces of acid catalysts.

Another approach is the use of low molecular weight radiation-sensitive crosslinkers in polymer-solvent formulations.

In this regard, the use of bisaryl azides is disclosed in T Iwayanagi, T Kohashi, S Nonogaki, T Matsusawa, K Douta, H. Yanazawa, IEEE Transactions on Electronic Devices ED25 (1981), p. 1306. This document discloses a photosensitive composition of an aromatic azide compound and a phenolic resin as a negative working deep UV photoresist for lithography. The azide compound is mixed with the phenolic resin in an amount between 5-30 wt%.

Also, in S.X.Cai, D.J.Glenn, M.Kanskar, M.N.Wybourne and J.F.W.Keana, "Development of highly efficient deep-uv and electron beam mediated cross-linkers: synthesis and photolysis of bis(perfluorophenyl)azides", Chemistry of 6 Materials (1994), pp. 1822-1829, the use of bis(perfluorophenylene azide) for the purpose of negative resist formulations has been disclosed. In this publication, bis(perfluorophenylene azide) is blended with poly(styrene), which is a common lithographic polymer. This document does not disclose polymer devices.

While the use of cross-linking agents is known in the field of negative-tone-resist e-beam and optical lithography, negative-tone lithographic resist (NLR) formulations and cross-linkable formulations are critical for semiconductor polymerization Specific (SP) requirements are vastly different:

(1) NLR formulations typically include a non-absorbing, non-polymeric matrix and a cross-linker system that is sensitive to the desired radiation. Since the polymer matrix is substantially transparent to the stated wavelengths of radiation, it does not itself impose requirements on the spectral wavelengths required for the crosslinker system to be sensitized. The spectral sensitivity of the crosslinker system is therefore mainly governed by the chosen radiation wavelength. In contrast, SPs are characterized by strong absorption bands in the visible and ultraviolet wavelength ranges. This absorption feature is important for the nature and use of SPs in devices. Because the absorption intensity of their main absorption spectral bands is considerable, the corresponding absorption depth is shallow and SP films are practically opaque at these wavelengths. For example, the 1/e absorption depth (or equivalently, the radiation penetration depth) is typically below 50 nm at the band maximum for most SPs. The cross-linked system thus ideally has its spectral sensitivity matched to a limited transparency window between the strong absorption bands of the individual SPs. Crosslinking systems that do not meet this requirement can only crosslink the uppermost layer of the membrane, thus limiting their use in SP. Furthermore, it would also be desirable to be able to use wavelengths at which the SP is substantially non-absorbing to avoid concomitant light-induced oxidation and other reactions.

(2) NLR formulations are not used as active layers in semiconductor devices. They serve as a sacrificial masking layer and are therefore not subject to the stringent requirements of cross-linkable SP formulations required for use in the device. In particular, in SP formulations, the crosslinker system should not introduce large concentrations of electron traps, hole traps or excitonic traps, which may degrade device performance. For example, some well-known photogenerated acid (PGA) chemically amplified photoresist systems leave PGA residues that can seriously interfere with the charge transport and light emitting properties of conjugated polymers.

The use of non-fluorinated bis(phenylazides) as radiation sensitive crosslinkers to formulate negative photoresists based on the conductive polymer PEDT:PSS has recently been disclosed in F.J.Touwslager, N.P.Willard and D.M.de Leeuw, "I-line lithography of poly(3,4-ethylenedioxythiophene) electrodes and application in all-polymer integrated circuits", Applied Physics Letters, 81(2002), pp. 4556-4558. This describes the photolithographic patterning of conductive (as opposed to semiconducting) polymer lines to form electrodes and interconnects. The PEDT:PSS was crosslinked with bis(phenylazide) after exposure to 365-nm I-ray radiation. The main feature of this method (like the azide method of WO 96/20253) is that phenyl azides as a class of species are rather inefficient as crosslinkers, requiring their introduction in high concentrations. Furthermore, there are many by-products of the cross-linking reaction, many of which are toxic to device performance.

A similar approach is described in US 2002/106529, where the electroluminescent material is blended with or bound to a polymeric binder. The binder polymer may be crosslinked. Some binder polymers having photocrosslinkable groups have repeating units represented by the general formula (9) disclosed on page 5. Clearly, these materials have a high mol% content of photocrosslinkable groups. Alternatively, the photocrosslinking reactive moiety may be mixed with the binder polymer, and in this embodiment, it has been taught that the agent should be mixed in an amount of 5-50 parts by weight to polymerize with 100 parts by weight of the binder thing-based.

The systems disclosed in the prior art, especially those disclosed in the above Applied Letters Physics disclosure, US 2002/106529, WO96/20253, and US 6107452, are not particularly suitable for use in devices. The disclosed crosslinking processes are based on polymerization or on specific couplings between two crosslinking moieties. In order for crosslinking to occur effectively, the crosslinker moiety must be present in very high loadings (substantially higher than 5 mol%). High loadings of the crosslinker moiety itself or by-products of the reaction of the crosslinker moiety tend to interfere with efficient charge-carrier transport in the device layers. Furthermore, the crosslinker moieties can indirectly interfere with efficient charge-carrier transport through their dilution and disordering effects. "Dilution effect" refers to the effect caused by the decrease in the relative concentration of conjugated segments in the polymer as the loading of crosslinking moieties in the final product increases. By "disordering effect" is meant that the backbone of the final crosslinked polymer is twisted out of plane due to the high loading of crosslinked moieties. As a result, the driving voltage of the devices is correspondingly increased, and their service life is shortened.

In view of the above, it should be appreciated that there remains a need to provide other methods of polymer device fabrication in which initially cast or deposited polymer layers become insoluble in solvents used in subsequent processing steps.

It is therefore an object of the present invention to provide new methods of polymer device fabrication, preferably combined with high performance. Furthermore, it is an object of the present invention to provide polymeric devices obtainable by this new method.

In this context, the inventors have surprisingly found that hitherto unknown low concentrations of crosslinking moieties (<5 mol %) can be successfully used in semiconducting layers in the manufacture of devices. The crosslinking moiety can be mixed with the semiconducting polymer or be part of the polymer backbone or side chains to form a crosslinked polymer product with a low concentration of the crosslinking moiety. Furthermore, the inventors have found that this low concentration of cross-linking moieties does not substantially degrade the performance of the polymer in polymeric devices.

Accordingly, in a first aspect the present invention provides a method of forming a polymer device comprising the steps of:

(i) depositing a solution comprising a polymer or oligomer and a moiety for crosslinking on a substrate to form a layer;

(ii) curing the layer formed in step (i) under conditions to form an insoluble cross-linked polymer;

It is characterized in that the crosslinking moiety is present in step (i) in an amount ranging from 0.05 to 5 mol%, based on the total number of moles of polymer or oligomer and crosslinking moiety in the solution.

The crosslinking moiety may be mixed with the polymer or oligomer in the solution deposited in step (i). Alternatively, the crosslinking moiety may be part of the polymer/oligomer backbone or side chains of the polymer/oligomer in solution.

The crosslinking moiety is at a level in the range of 0.05 mol% to 5 mol%, preferably at a level in the range of 0.05 mol% to less than 5 mol%, more preferably at a level in the range of 0.05 to 3 mol%, still more preferably at A level in the range of 0.1 mol% to 2 mol% and still more preferably a level in the range of 0.1 mol% to 1 mol% is present in the solution in step (i), based on the polymer or oligomer and crosslinking in solution The total number of moles of moieties is used.

When a crosslinking moiety is mixed with a polymer or oligomer in the solution, the level of crosslinking moiety present can be readily measured according to the following formula:

n _{cross-linking agent} /(n _{cross-linking agent} +n _polymer )×100

where _ncrosslinker is the number of moles of crosslinking moiety, and _npolymer is the number of moles of repeating units of the polymer or oligomer. For random or alternating copolymers _{described according to the general formula AxByC(1-xy); and for block copolymers described according to the general formula AxByCz , the repeating units of} the polymers are important for the present invention Purposes are defined as A, B, and C.

In the three examples (a), (b) and (c) shown below, _npolymer is (moles of A) + (moles of B) + (moles of C). In each of the examples (a), (b) and (c) shown below, _npolymer is (6+6+6)=18. By way of example only, if one mole of a crosslinking moiety is present in solution with any of polymers (a), (b) or (c), the crosslinking moiety will be present in a ratio of 1/(1 +18) x 100 = 5.26 mol% concentration present.

(a) ABCBBACACCBAACBCBA

(b) ABCABCABCABCABCABCABC

(c)AAACCCBBBCCCAAABBB

The relative amounts of crosslinker moieties and repeat units can be measured by microanalysis or by NMR.

Where the crosslinking moiety is part of a polymer or oligomer backbone or side chain, for the purposes of the present invention, the level at which the crosslinking moiety is stated to be present should be measured as shown below.

Consider a polymer or oligomer comprising repeating units having one of the following structures:

Part of the cross-linking group attached to the main chain as a side chain

cross-linking group

Where R is a repeating unit in the backbone of a polymer or oligomer comprising a hydrocarbon and X is a crosslinking moiety, then the level of crosslinking moiety present should be measured as follows:

n _{cross-linking agent} /(n _{cross-linking agent} +n _polymer )×100

where _ncrosslinker is the number of moles of crosslinking moiety X, and _npolymer is the number of moles of repeating units of the polymer or oligomer.

The insoluble crosslinked polymer formed in step (ii) contains 0.05-5 mol % of crosslinking moieties. Preferably, the crosslinking moiety is present in an amount ranging from 0.05 mol % to less than 5 mol %, more preferably in an amount ranging from 0.05 mol % to 3 mol %, more preferably in an amount ranging from 0.1 mol % to 2 mol % amount, still more preferably present in an amount ranging from 0.1 mol% to 1 mol%, based on the total moles of insoluble crosslinked polymer in the layer formed in step (i).

Consider insoluble cross-linked polymers comprising repeat units with one of the following structures:

Where R and R are repeating units in the backbone of the crosslinked polymer and X is a crosslinking moiety, then the level of presence of the crosslinking moiety should be measured as follows:

n _{cross-linking} /(n _{cross-linking} +n _polymer )×100

where _ncrosslinks is the number of moles of crosslinked moieties X', and _npolymer is the number of moles of repeating units of the polymer.

The present method provides a simple route to crosslinking as-deposited polymer films to achieve any desired film thickness, for example from about 1 nm to about 500 nm in one process cycle. This is highly versatile and versatile craft. In the present method, this can be achieved in many cases without introducing a greater concentration of charge-carrier or exciton traps. As a result, it is possible to fabricate a wide variety of practical polymer-polymer hybrid structures and to incorporate them advantageously into polymer devices, especially in light-emitting diodes, photodiodes and field-effect transistors. The present method also opens up new avenues for the patterning of polymer films.

Altogether, practical and essentially unlimited multilayer stacks and patterned polymer films can be fabricated using the general strategy defined by this method.

Unexpectedly, the present inventors have discovered that crosslinking at this low level produces substantially no "killer" defects or by-products that seriously impair the electrical and optoelectronic functions of the semiconducting polymer layer.

Curing in step (ii) can be achieved by electron beam radiation. Preferably, however, the curing conditions include exposure to short wavelength radiation in an inert atmosphere, more preferably exposure to deep ultraviolet radiation in an inert atmosphere. More preferably, the wavelength of the deep ultraviolet radiation is in the range of 200nm to 400nm, more preferably 245nm to 370nm, even more preferably 250nm to 260nm. Particularly preferred wavelengths are about 254-nm, available from low-pressure mercury lamps, and 248 nm, available from eg KrF excimer lasers. Therefore, it is preferred that the moiety for crosslinking is sensitive to short-wavelength radiation of the above-mentioned wavelength.

When the solution deposited in step (i) contains oligomers, curing in step (ii) will polymerize and crosslink the oligomers to form an insoluble crosslinked polymer.

Advantageously, it has been found that semiconducting polymers are generally stable to the actual deep UV dose required for exposure, as long as the exposure is performed in an inert atmosphere.

The present inventors have discovered that many of the main semiconducting polymers (polythiophenes, polyarylvinylidenes, polyfluorenes and their copolymers) unexpectedly have a common transparent window in the deep ultraviolet at about 200-300 nm. This spectral characteristic matches well with the sensitivity spectrum of the preferred crosslinking moieties discussed below, and ideally with commercially available light sources.

Preferred crosslinking moieties include fluorinated aryl azides. It is further preferred that there are no hydrogen atoms located ortho to the azide in the fluorinated arylazides. For this purpose, in some embodiments it is preferred that the fluorine atom be located anywhere ortho to the azide.

More preferably, the crosslinking moiety comprises perfluoroaryl azide, still more preferably perfluorophenyl azide or perfluoronaphthyl azide.

In a first embodiment, particularly preferred moieties for crosslinking are mixed in step (i) with polymers or oligomers in solution, and advantageously crosslinking takes place via the following mechanism, according to which The crosslinking reaction involves a bonding reaction between the crosslinking moiety and the polymer or oligomer unit, as opposed to a bonding reaction between the crosslinking moiety itself. In other words, there is substantially no self-coupling or self-polymerization of the crosslinking moiety in step (ii). The inventors have found that crosslinking moieties having the above-mentioned capabilities are advantageously used in rather low concentrations.

It should be noted that the preferred cross-linking moieties as described above do not substantially self-couple or self-polymerize in step (ii). However, such crosslinking moieties are capable of self-coupling or self-polymerization. It is simply required in one embodiment of the invention that they substantially do not do so in step (ii) of the method of the invention.

Further preferred crosslinking moieties in this first embodiment have the following general formula I:

N ₃ -Ar _F -N ₃ I

wherein _ArF includes substituted or unsubstituted fluorinated aryl groups.

More preferred crosslinkers have the general formula II:

N ₃ -Ar _F -L-Ar _F '-N ₃ II

wherein _ArF and _ArF ' independently each comprise a substituted or unsubstituted fluorinated aryl group and L comprises an optional divalent or multivalent linker group. Preferred substituents include substituents that are bulkier than F, such as _CF3 . In one embodiment it is preferred that one or both of _ArF and _ArF ' have at least one (preferably one) substituent which is more bulky than F.

Preferred linker groups include electron withdrawing moieties bonded to Ar and/or Ar _F '. Such electron withdrawing moieties include -CO-, -C(O)O-, S(O) _2O- , C(O)NR-, or S(O) _2NR- , where R is H or a substituent , together with flexible spacers such as (CH ₂ ) _x (x=1-5), (CH ₂ ) _x -O-(CH ₂ ) _x (x=1-3), or cyclohexanediyl segments . The linker group may comprise two electron-withdrawing moieties (one bonded to _ArF and the other bonded to _ArF ') connected by a flexible spacer group. More preferred linker groups include -CO-, -C(O)O-( _CH2 ) _n -O(O)C-(n=1-5) and, -C(O)O-cyclohexane Diradicals -O(O)C-, and -S(O) ₂ -NR-( _CH2 ) _x -NR-S(O) _2- .

In the general formula II, it is more preferred that Ar _F and Ar _F ' each comprise a substituted or unsubstituted fluorinated phenyl or naphthyl, still more preferably a substituted perfluorophenyl or perfluoronaphthyl. Most preferably, Ar _F and Ar _F ' each comprise a perfluorophenyl group. For the reasons discussed below for compounds XVI to XVIII, in one embodiment one or more (preferably one) of the fluorine groups on the perfluorophenyl or perfluoronaphthyl can advantageously be replaced by a larger bulky group Groups such as fluorinated alkyl groups (eg trifluoromethyl) are substituted. This substitution advantageously takes place on 2 bits.

Thus, in one embodiment Ar _F and/or Ar _F ' preferably comprises:

wherein B is a bulky substituent such as a fluorinated alkyl group (eg trifluoromethyl).

The linker groups between the perfluorophenyl or perfluoronaphthyl groups are preferably short linker groups of the type described above.

Each azido group (-N ₃ ) is photochemically decomposed to electron-deficient nitrene, which readily enters into CH bonds, especially alkyl CH bonds: -N+-CH=-CN(H)-. In this manner, each end of the bis-azide is attached to the CH segment. Two CH segments nominally belong to adjacent polymer chains or oligomers and thus form a crosslink between them. For this reason, the polymer or oligomer preferably includes many saturated hydrocarbon segments. However, to improve solubility, saturated hydrocarbons, in particular alkyl CH's, are often present in polymers or oligomers.

The present inventors have found that crosslinking proceeds through high-efficiency insertion into C-H bonds, in particular into alkyl C-H bonds, without generally disrupting the π-conjugation present on the main chain. Furthermore, the ability of the crosslinking moiety to form bonds with the polymer or oligomer has been discovered by the present inventors to allow the use of lower concentrations of the crosslinking moiety.

The present inventors have found through experiments that this moiety for crosslinking is generally free from the defect of leaving residues that may hinder polymer properties such as charge transport and light emitting properties.

Furthermore, the crosslinking mechanism of this crosslinking moiety has been found to be generally compatible with the presence of π-conjugation in semiconducting polymers.

Furthermore, it has been found that the solid state crosslinking efficiency of conjugated polymers is unexpectedly high (>80%) for this crosslinking moiety. Without wishing to be bound by theory, the high efficiency of this type of crosslinker is believed to be due to the presence of fluorine atoms (especially ortho fluorine) in the ring, which suppress competing parasitic ring expansion side reactions.

The following novel crosslinkers have been synthesized by the present inventors. These groups of crosslinkers are suitable for mixing in solution with polymers or oligomers:

- alkylene glycol bis(4-azido-2,3,5,6-tetrafluorobenzoate),

-Alkylenediamine bis(4-azido-2,3,5,6-tetrafluorobenzamide),

-Alkylenediamine bis(4-azido-2,3,5,6-tetrafluorobenzenesulfonamide),

- cycloalkylene glycol bis(4-azido-2,3,5,6-tetrafluorobenzoate),

-cycloalkylenediamine bis(4-azido-2,3,5,6-tetrafluorobenzamide),

-cycloalkylenediamine bis(4-azido-2,3,5,6-tetrafluorobenzenesulfonamide),

Specific examples of these novel crosslinkers include:

- Ethylenediaminebis(4-azido-2,3,5,6-tetrafluorobenzamide);

-Ethylenediaminebis(4-azido-2,3,5,6-tetrafluorobenzenesulfonamide)

-1,3-cyclohexanediol bis(4-azido-2,3,5,6-tetrafluorobenzoate),

- 1,4-Cyclohexanediol bis(4-azido-2,3,5,6-tetrafluorobenzoate).

The above-mentioned novel compounds wherein at least one (preferably one) of the fluorine atoms in tetrafluorobenzamide, tetrafluorobenzenesulfonamide or tetrafluorobenzoic acid ester is replaced by a bulkier group (such as a fluorinated alkyl group, for example CF ₃ ) The equivalent of the crosslinker is ideal. Substitution is preferably at position 2.

Crosslinking agents having the general formula II:

N ₃ -Ar _F -L-Ar _F '-N ₃ II

wherein _ArF and _ArF ' are each as defined above, can be synthesized by a method comprising the step of reacting F- _ArF -L- _ArF' -F with an azide.

Preferably, the azide is a metal azide, more preferably an alkali metal azide, still more preferably sodium azide.

F-Ar _F -L-Ar _F' -F can be prepared by reaction of a suitable fluorinated aryl (eg pentafluorophenyl) acid halide. The co-reactants depend on the nature of L. However, the reaction of fluorinated aryl (eg, pentafluorophenyl) acid chlorides with diols or diamines is often useful.

For example, F-Ar _F -L-Ar _F '-F can be combined with HO-R-OH or NHY-R-NHY by F-Ar _F -COX or F-Ar _F -SO ₂ X (X=Cl, Br) (Y = H, alkyl, or aryl, preferably alkyl or aryl for improved solubility in the final product) (R = alkylene, cycloalkylene) react to form F-Ar _F -L -Ar _F '-F to make.

For ethylene glycol bis(4-azido-2,3,5,6-tetrafluorobenzoate), this can be achieved by 1 molar equivalent of ethylenediamine with a slight excess of 2 molar equivalents of pentafluorobenzyl It is prepared by the reaction of base chloride to produce ethylenediamine bis(pentafluorobenzoate). Ethylene glycol bis(pentafluorobenzoate) can then be reacted with a slight excess of 1 molar equivalent of metal azide to produce ethylene glycol bis(4-azido-2,3,5,6-tetrafluoro parabens).

For ethylenediaminebis(4-azido-2,3,5,6-tetrafluorobenzenesulfonamide), this can be achieved by 1 molar equivalent of ethylenediamine with a slight excess of 2 molar equivalents of pentafluorobenzenesulfonyl chloride Reaction to produce ethylenediamine bis (pentafluorobenzenesulfonamide) to prepare. Ethylenediaminebis(pentafluorobenzenesulfonamide) can then be reacted with a slight excess of 1 molar equivalent of metal azide to produce ethylenediaminebis(4-azido-2,3,5,6-tetrafluorobenzene sulfonamide).

For step (i), ideally, the solution comprises a blend/mixture of polymers and crosslinking moieties. However, in another embodiment, the solution may additionally contain a polymer or oligomer in which a crosslinking moiety is bonded to the polymer or oligomer, typically as a pendant group.

In this embodiment, particularly preferred moieties for crosslinking are part of the backbone of the polymer or oligomer or attached as a side chain to the polymer or oligomer, and advantageously the crosslinking utilizes the following mechanism According to which mechanism the crosslinking reaction involves a bond forming reaction between the crosslinking moiety and the polymer or oligomer unit, as opposed to a bond forming reaction between the crosslinking moiety itself. In other words, there is substantially no self-coupling or self-polymerization of the crosslinking moiety in step (ii).

In this embodiment, preferred structural units in polymers or oligomers comprising a crosslinking moiety have the general formula III or IV:

where Ar _F comprises a substituted or unsubstituted fluorinated aryl group and R is a structural unit (typically a repeating unit) in the backbone of the polymer or oligomer.

These preferred crosslinking moieties comprise azido groups, and thus these crosslinking moieties have the same advantages as described above with respect to the first embodiment of the preferred crosslinking agent according to the invention.

In formulas III and IV, Ar _F can be as defined anywhere above for formula II.

In formulas III and IV, preferably Ar _F comprises a substituted or unsubstituted fluorinated phenyl or naphthyl group, more preferably a perfluorophenyl or perfluoronaphthyl group. Most preferably, Ar _F includes perfluorophenyl groups.

In general for the process of the invention, the deposition in step (i) is achieved, for example, by any suitable solution-processing method. In this connection, mention may be made of ink-jet printing, centrifugal casting, screen printing, dip coating, and offset printing.

Suitable materials for the substrate will depend on the polymer device being formed. For LEDs and photodiodes/photodetectors, preferred substrates include ITO layers on glass, ITO layers on PET, ITO layers on Si, and the like. For FETs, preferred substrates include layers of glass, PET, polycarbonate, and the like. The substrate may itself comprise a laminate structure. In other words, the substrate may itself comprise a variety of different layers.

As mentioned above, in step (ii) of the method of the invention, the layer formed in step (i) becomes insoluble. In order to achieve this insolubility, a sufficient degree of crosslinking must occur in step (ii) when the layer formed in step (i) is subjected to crosslinking conditions. The precise amount (within specified ranges) of the crosslinking moieties required in the solution deposited in step (i) in order to achieve the desired degree of crosslinking in step (ii) will depend on the molecular weight distribution characteristics of the polymer . In general, the higher the molecular weight of the polymer, the lower the amount of crosslinker required. The minimum amount required is suitably determined by gel-fraction experiments. This experiment can be performed as follows:

The polymer or oligomer is tested at a concentration in a suitable solvent (for example 0.5-2.5 wt% in an aromatic hydrocarbon solvent) and at a concentration in the range of 0.05-5mol% (based on crosslinking groups in solution and total moles of polymer/oligomer) of crosslinking groups; or use this fraction of crosslinking moieties to synthesize the polymer.

Cast films by spin coating or inkjet printing and then curing;

Measurement of thickness by profilometry, ellipsometry or interferometry;

• Expose the membrane to a dose of 100 mJ/ ^cm2 of crosslinking radiation.

Soak (or develop) the film in a solvent that normally dissolves the polymer, then blow dry or spin to remove the solvent;

Re-measure the film thickness;

• Repeat the above sequence at different molar ratios of crosslinking moieties until insolubility and desired thickness are achieved.

The preferred concentration range of the polymer or oligomer in the solution is 0.5-2.5% by weight before the addition of the crosslinking moiety.

The crosslinked polymer formed in step (ii) is advantageously a conducting, semiconducting or insulating polymer. Preferably, the crosslinked polymer is a semiconducting polymer.

When the polymer is present in the solution used in step (i) and the crosslinking moiety is mixed with the polymer, ideally the polymer in solution may be a conducting, semiconducting or insulating polymer. Preferably, the polymer in the solution in step (i) is a semiconducting polymer. When the polymer in solution is a semiconducting polymer, curing the layer in step (ii) advantageously does not substantially affect the semiconducting properties of the polymer.

In contrast to semiconducting polymers, conducting polymers are typically heavily doped (>5 mol%, by repeating unit) into a conducting state. As a result, conducting polymers typically have a charge carrier concentration >10 ¹⁸ cm ⁻³ . "Conductive polymer" typically refers to a polymer having a conductivity > 1 x 10 ^-5 S/cm. As such, their electrical properties are substantially insensitive to additional impurities. Such conductive polymers are mainly used as transmission lines or electrode connectors. Importantly, they often have transmission windows that extend greatly into various parts of the visible, ultraviolet and deep ultraviolet spectral regions, with increased laxity for photopatterning processes.

Semiconducting polymers are typically undoped or inherently doped at low concentrations (typically 0.001 mol % or less). In contrast to conducting polymers, semiconducting polymers typically have a charge carrier concentration of <10 ¹⁵ cm ⁻³ . A "semiconducting polymer" typically refers to a polymer having a conductivity &lt; 1 x 10&lt;^&quot;8&gt; S/cm. These polymers primarily form the core of various polymer device technologies including LEDs, FETs and PVs. They typically have a rather narrow transmission window in the visible-ultraviolet region explained above. They also possess important and unique transport and photophysical properties that are more sensitive to the amount of impurities.

The inventors have found that semiconducting polymers can be crosslinked according to the present invention with substantially no loss of functionality.

In particular, the inventors have found that the crosslinking moiety has practically no effect on the photoluminescent and electroluminescent properties of various conjugated polymer films, especially when used in an amount of 0.1-0.5 mol% and when polymerized When the molecular weight of the compound is sufficiently high (>300,000).

The inventors have found that the crosslinking moieties and crosslinking methods described in the present invention are in fact even comparable to wide bandgap materials, especially blue EL polymers, as described for example in WO03/095586. The process does not introduce excitons or charge traps, which impair device performance in these particularly sensitive materials.

For low molecular weight materials, higher concentrations of crosslinking moieties (>0.5%) need to be used, and in some cases this will start to have a detrimental effect on the photoluminescent efficiency of the polymer, as in Figure 11 for crosslinking. The combined structure is represented by ethylene glycol bis(p-azido-2,3,5,6-tetrafluorobenzoate). In many cases, photoluminescent efficiency can be partially restored by annealing.

Polymer XI:

Polymer XII:

Moieties for cross-linking:

Exposure conditions: 254-nm Hg line; 1mW/cm ² , 2min

The undesired effect of the crosslinking moiety at high concentrations can further be minimized by increasing the steric volume on the crosslinking moiety. For the crosslinking moieties of the general formula II this can be achieved, for example, by a suitable choice of L, or of substituents on Ar _F , Ar _F ' and/or L. For example, the following series are observed relative to photoluminescence extinction:

XVI>XVII>XVIII

So, if the crosslinking moiety needs to be present in high concentration due to the low molecular weight of the polymer that needs to be crosslinked, and therefore photoluminescence quenching becomes an issue, then there is a higher steric volume near the perfluorinated ring The crosslinking moiety of , for example 2-trifluoromethyl substituted compound XVIII, can be chosen to alleviate or completely suppress this problem.

This illustrates the versatility of fluorinated phenylazides in providing efficient crosslinking to various conjugated polymers without compromising the desired photophysical and electronic properties of the polymers.

Preferably the polymer or oligomer backbone is at least partially conjugated. Furthermore, in one embodiment, the polymer or oligomer backbone is preferably substantially or even completely conjugated.

As for the structure of the polymer or oligomer, it is preferable that it includes a plurality of saturated hydrocarbon segments (-CH ₂ - and -CH-) in the side chain or the main chain. It is preferred that the polymer or oligomer include a plurality of aliphatic hydrogens.

The preferred weight fraction of these segments in the polymer or oligomer is 10-100%, for semiconducting polymers the preferred weight fraction is 10-70%.

Preferred crosslinked semiconducting polymers comprise repeating units Ar ₁ comprising:

(1) Substituted or unsubstituted phenylene or aryl vinylene, such as p-phenylene vinylene;

(2) 9,9-disubstituted fluorenes, such as 9,9-dialkylfluorenes (which may be further substituted);

(3) substituted or unsubstituted triarylamines;

(4) Unsubstituted or substituted heteroaromatic units such as thiophene, benzothiadiazole, quinoline, pyridine, preferably substituted analogues; or

(5) Unsubstituted or substituted oxadiazoles.

Preferred substituents include alkyl, cycloalkyl, alkoxy, aryl, and aryloxy.

The polymer may further contain one or more co-repeat units in combination with Ar ₁ . Preferred co-repeat units include para-linked unsubstituted or substituted phenylene; unsubstituted or substituted phenylene vinylene; 2,5-linked substituted or unsubstituted phenylene thiadiazoles; 2,5-linked substituted or unsubstituted thiophenes; and thiophenes or terthiophenes, substituted or unsubstituted triarylamines or bis(triarylamines). It is especially preferred that the crosslinked polymer contains one or more co-repeat units when Ar ₁ comprises a 9,9-disubstituted fluorene.

All these polymers can be crosslinked according to the invention.

Of course, the polymer or oligomer must be soluble in order for it to be in solution in step (i). For this purpose, the polymer or oligomer may comprise solubilizing groups. Preferred solubilizing groups include alkyl, alkoxyaryl, cycloalkyl, aryloxy, and cycloalkyloxy.

It is preferred that the crosslinking moiety has absorption in a narrow transmission window in the deep UV. Typically this is in the range of 200nm to 300nm, more preferably 245 to 270nm, even more preferably 250nm to 260nm. The absorption of the crosslinking moiety can be measured by UV visible absorption spectroscopy.

The crosslinking moiety advantageously has a suitable absorption in this range, since this corresponds to a transition window common to many semiconducting polymers. Therefore, cross-linking and imaging of polymers can be accomplished at low exposure doses.

A preferred thickness of the cured layer formed in step (ii) is in the range of 500 nm or less. When the curing in step (ii) is performed by exposure to UV radiation, the insoluble layer formed can be 500 nm thick to a few nm thick after a suitable UV exposure. Layers with a final thickness greater than 500 nm can be produced by repeated coating and curing, if desired. The desired final film thickness depends on the end use. In principle there is no limit to the number of layers produced by this method.

The desired thickness of the insoluble layer formed in step (ii) depends to some extent on the function of the layer. When the layer is an implanted interlayer in a polymer LED, the preferred thickness is in the range of 5 to 20 nm. When the layer is a charge transport layer, eg in a photodiode, the preferred thickness is in the range of 10 to 50 nm. When the layer is a cladding layer in a waveguide device, the preferred thickness is in the range of 100 to 400 nm. When the layer is in the channel layer of a FET, the preferred thickness is in the range of 20 to 300 nm.

In one embodiment the layer deposited in step (i) may be a polymer blend or composite. Crosslinking may be advantageously performed in order to increase the thermal stability of the cured polymer blend or composite or to optimize the resistance of the final cured layer to solvent dissolution.

For the curing conditions in step (ii), when curing involves exposing the layer to ultraviolet radiation, this is preferably at a power of 1-100 mW/cm ² , and the preferred exposure time is in the range of about 0.1-100 s . The energy dose on the cured layer is preferably 1-100 mJ/cm ² , more preferably 5-20 mJ/cm ² .

This layer is cured in step (ii) so that it becomes insoluble. This means that the layer and thus the polymer is not completely soluble in any solvent in which the layer is soluble prior to crosslinking. Achieving this result depends on the desired level of crosslinking for the particular layer, as described above. Typically, the cured layer becomes insoluble in common organic solvents. Furthermore, generally the layer will become insoluble in aromatic hydrocarbon solvents including, toluene, xylene, mesitylene, durene, hydrogenated naphthalene, etc., and halogenated solvents such as chloroform, Chlorobenzene and so on. These solvents are thus available for subsequent processing in device fabrication.

A specific test for determining insolubility can be described as follows:

Films are cast by spin coating or inkjet printing and then cured;

Accurately measure the thickness by profilometry, ellipsometry or interferometry, call it _d1 ;

Soak (or develop) the film in a solvent that normally dissolves the polymer, then blow dry or spin dry;

• Measure the film thickness again, call it _d2 .

When the layer is completely "insoluble", there is no decrease in film thickness after soaking the film/layer (ie d ₂ /d ₁ =1.0). In many cases, however, this layer only has to be partially insoluble. Provided the retained fraction (d ₂ /d ₁ ) is known, any reduction can be allowed in the design of the device. Generally, however, d ₂ /d ₁ needs to be greater than 0.4, preferably greater than 0.5 to be useful.

Typically, after curing in step (ii), the layer can be contacted with a solvent. The fact that the crosslinked polymer is insoluble in solvents (in which equivalent uncrosslinked polymers are soluble) means that the solvent with which the layer comes into contact can be selected from a broad class of solvents including common organic solvents. This contact does not dissolve the crosslinked polymer formed in step (ii).

Optionally, after curing in step (ii), the layer may be washed with a suitable solvent. This washing step may be included when the layer is patterned crosslinked in step (ii). This involves exposing only selected areas of the layer of step (i) to the curing conditions in step (ii). This can be achieved, for example, by exposing a mask to UV radiation. Material in the exposed areas becomes insoluble, while material in the unexposed areas will remain soluble. This allows material in the unexposed areas to be removed during the washing step.

Optionally, after curing in step (ii), the layer may be chemically modified by wet chemical methods, by suitable chemical reactions. Such chemical reactions may include aromatic sulfonation, aminomethylation, or other derivatization reactions.

Sulfonation introduces _SO3H groups into a portion of the polymer repeat units. This can be used to produce, for example, self-doped conducting polymer layers. This particular reaction can be performed under a wide range of conditions, such as allowing the layer to react with chlorosulfonic acid in dilute chloroform at -60°C.

Another reaction that may be useful is the methylation of NH groups introduced by crosslinking reactions. This reaction will replace the hydrogen atom with a potentially more stable methyl group. This particular reaction can be performed by allowing the layer to react with methyl iodide at room temperature, followed by washing with triethylamine in a chloroform-ethanol mixture. So after the polymer layer becomes insoluble, various chemical reactions can be performed to change or adjust the bulk and surface properties of the layer.

Additional (second) layers may be deposited on the layer formed by the method of the invention. In this regard, the layer formed in the process of the invention is insoluble in any solution used to deposit the additional (second) layer, taking into account the curing in step (ii).

Optionally, the method of the invention may additionally comprise the step of annealing the insoluble polymer formed in step (ii) of the method according to the first aspect of the invention. Annealing can be performed according to the method described below according to the sixth aspect of the present invention.

Typically the polymeric device in the method according to the first aspect of the invention is an optical device. Preferably, the device is a polymer light emitting device, a polymer transistor such as a field effect transistor, a photodetector, a photovoltaic device, a waveguide device, or a distributed Bragg reflector.

Polymer LED devices can be fabricated with crosslinked hole-transporting polymer layers deposited by the method of the present invention. The device may further comprise a light emitting polymer layer and/or an electron-transporting polymer layer and/or an exciton-blocking polymer layer. Optionally, these additional layers can be deposited according to the method of the invention. For polymer LEDs, for example, a structure can be designed that includes a hole-injecting and electron-blocking polymer layer formed on the anode, followed by a light-emitting polymer layer, followed by an electron-injecting, hole-blocking polymer layer. blocking and exciton-blocking polymer layer, and then there is the cathode. Advantageously, suitable polymers can also be patterned as emissive layers, resulting in full color displays as discussed above. Alternatively, the light emitting layer can be a film of a crosslinked blend of polymers.

Polymer waveguide LED devices can be fabricated with one or more crosslinked polymer cladding layers deposited by the method of the present invention. The device further comprises a light emitting core layer, which optionally may be deposited according to the method of the invention. Waveguide devices are characterized by a core layer (or strip) having a higher refractive index than the adjacent cladding layer (or peripheral layer). The core and cladding layers may each comprise one or more monolayers. Because of its higher refractive index, light of the appropriate wavelength that satisfies the phase-matching condition is intercepted by total internal reflection and guided in the core (or strip). This light can be emitted on the edge of the device or directed into another region where it can be out-coupled. Light emitted in this manner can be highly directional and well coupled to optical fibers.

Polymer distributed Bragg reflectors can be fabricated using multiple layers of crosslinked alternating high and low refractive index polymers deposited by the method of the present invention. The Bragg reflector comprises multiple quarter-wavelength thickness ( _dH , _dL ) layers of high ( _nH ) and low ( _nL ) refractive index materials. Light having a wavelength satisfying the Bragg condition (λ/2=n _H d _H +n _L d _L ) is strongly reflected in the stack. The Bragg reflector can be coupled with another Bragg reflector or mirror to form an optical resonator. Such resonators have important uses as wavelength selectors.

Polymer microporous LED devices can be fabricated with one or more crosslinked polymer distributed Bragg reflectors deposited by the method of the present invention. The device may further include a light emitting core layer. Optionally, a luminescent core layer may be deposited according to the method of the invention.

Polymer FET devices can be fabricated with crosslinked semiconducting polymer layers deposited by the method of the present invention. The device may further comprise a cross-linked insulating polymer layer. Optionally, an insulating polymer layer may be deposited according to the method of the invention. This is especially the case if this layer is deposited before the semiconductor layer. Devices can be top-gate, side-gate or bottom-gate configurations. For polymer FETs, for example, a structure is designed that includes a charge-transporting semiconducting polymer formed on a substrate between the source and drain, followed by an insulating polymer that acts as a gate insulator. This insulating polymer layer can be deposited from the same solvent used to deposit the semiconducting polymer (after it has been crosslinked).

Polymer photovoltaic devices can be fabricated with photoresponsive layers comprising crosslinked polymer blends or polymer composites.

A second aspect of the invention provides a polymer device obtained or obtainable by a method as defined anywhere for the first aspect of the invention. The device may be as defined anywhere for the first aspect of the invention. In any of the devices described above for the first aspect of the invention and in devices according to the second aspect of the invention, preferred materials for the cathode include alkaline earth metals such as barium and calcium.

A third aspect of the invention provides the use of a polymer device according to the second aspect of the invention.

A fourth aspect of the present invention provides a solution containing a polymer or oligomer and a crosslinking moiety, characterized in that the crosslinking moiety is present in an amount ranging from 0.05 to 5 mol %, based on polymerization in solution The total weight of polymers or oligomers and crosslinking moieties.

In the fourth aspect of the invention, the polymer or oligomer, the moiety for crosslinking and the solvent may be the same as described anywhere above for the first aspect of the invention. Furthermore, their concentration in solution may be the same as described anywhere above for the first aspect of the invention.

Preferably the crosslinking moiety is mixed with a polymer or oligomer. A preferred solution according to the fourth aspect of the invention contains a semiconducting polymer. In a particularly preferred embodiment, the crosslinking moiety has the general formula II, wherein _ArF and _ArF ' are each a perfluorophenyl group.

A fifth aspect of the invention provides the use of a solution according to the fourth aspect of the invention for the manufacture of a polymer device as defined in the first and second aspects of the invention.

A sixth aspect of the present invention provides a method of forming a polymer device comprising the steps of:

(i) depositing a solution comprising a polymer or oligomer and a crosslinking moiety on a substrate to form a coating;

(ii) curing the layer formed in step (i) under certain conditions to form an insoluble cross-linked polymer;

(iii) annealing the insoluble polymer formed in step (ii); and

(iv) optionally chemically modifying the surface or bulk of the insoluble polymer formed in step (ii).

In the method of the sixth aspect of the present invention, polymers, oligomers, crosslinking moieties and their concentrations and crosslinking conditions may preferably be the same as defined everywhere for the first aspect of the present invention. In the sixth aspect of the present invention, the moiety for crosslinking is not necessarily present in step (i) in an amount ranging from 0.1 to 5 mol%.

Preferably, the annealing in step (iii) is performed at a temperature in the range of 120 to 200°C.

With regard to the sixth aspect of the invention, the inventors have surprisingly found that an annealing step after film deposition and crosslinking advantageously at least partially restores any observed performance degradation that occurred in step (ii).

The chemical modification in step (iv) of the method according to the sixth aspect of the invention may be performed as defined for the first aspect of the invention.

The invention will now be described in more detail with reference to the accompanying drawings, in which:

Figure 1 shows typical membrane retention (or gel point) profiles at different crosslinker ratios.

Figure 2 shows the relationship between film thickness and film retention.

Figure 3 shows the device characteristics of the device with the TFB interlayer.

Figure 4 shows the device characteristics of a similar device to that reported in Figure 3 but without the TFB interlayer.

FIG. 5 shows the current density versus diode voltage measured in Example 3. FIG.

FIG. 6 shows the luminance measured in Example 3 versus the diode voltage.

FIG. 7 shows the external quantum efficiency measured in Example 3 as a function of diode voltage.

Figure 8 shows a comparison of the device properties of a device according to Example 5 and a similar device without crosslinker.

FIG. 9 shows the transfer characteristics of the FET according to Example 6. FIG.

FIG. 10 shows the transfer characteristics of the device according to Example 7. FIG.

Figure 11 shows the photoluminescence efficiency of the crosslinking moiety ethylene glycol bis(p-azido-2,3,5,6-tetrafluorobenzoate) in two different polymers.

Exposure conditions were: 254-nmHg line; 1 mW/cm ² , 2 min.

FIG. 12 shows UV-visible absorption spectra collected during the multilayer fabrication process described in Example 8. FIG.

FIG. 13 shows the photocurrent curves of the device described in Example 8. FIG.

Example

Synthesis of cross-linking agent:

The following synthetic reaction scheme is used: a diol or diamine is reacted quantitatively with the appropriate pentafluorophenyl halide in the presence of an acid scavenger, and the product is then reacted quantitatively with sodium azide to give the bis (perfluorophenyl azide) crosslinker.

Example 1: Ethylene glycol bis(4-azido-2,3,5,6-tetrafluorobenzoate)

Ethylene glycol (135 mg, 2.1 mmol) and triethylamine (510 mg, 5.0 mmol) dissolved together in 10 mL of dry ether were added to pentafluorobenzoyl chloride (1.2 g, 5.0 mmol) in 10 mL of dry ether middle. A white precipitate of triethylammonium chloride was obtained, which was then filtered. The filtrate was washed with 3 x 20 mL of water, dried over MgSO ₄ , and evaporated to recover ethylene glycol bis(pentafluorobenzoate) (I) as a colorless liquid (yield, 75%). Sodium azide (150 mg, 2.3 mmol) in 2.2 ml water and 3.7 mL acetone was then reacted with I (500 mg, 1.1 mmol) in 2 mL acetone and stirred overnight on a 60°C hot plate. A white precipitate was obtained. The solution was reduced to half volume by evaporation, and the precipitate was filtered off (yield, 80%) and recrystallized twice from 1:3 chloroform-hexane to give ethylene glycol bis(4-azido-2,3, 5,6-Tetrafluorobenzoate) (II), as white crystals. FTIR: 2134 (N ₃ asymmetric stretching), 1731 (C=O stretching), 1645, 1483, 1252 (N ₃ symmetric stretching), no acid OH band at 3000-3500.

Example 2: Ethylenediamine bis(4-azido-2,3,5,6-tetrafluorobenzoate)

Ethylenediamine (130 mg, 2.1 mmol) and triethylamine (510 mg, 5.0 mmol) dissolved together in 10 mL of dry ether were added to pentafluorobenzoyl chloride (1.2 g, 5.0 mmol) in 10 mL of dry ether middle. A white precipitate was obtained which was filtered. The residue was washed with chloroform, then water, and then filtered to recover ethylenediamine bis(pentafluorobenzoate) (III) as white needle crystals (yield, 100%). Sodium azide (270 mg, 4.2 mmol) in 1.7 mL water and 4.0 mL DMF was reacted with III (870 mg, 2.0 mmol) in 20 mL DMF and stirred overnight on a 60° C. hot plate. A white precipitate was obtained. The solution was reduced to half volume by evaporation and the precipitate was filtered off (yield, 35%), dried and recrystallized from DMF to give ethylenediamine bis(4-azido-2,3,5, 6-tetrafluorobenzoate) (IV). FTIR: 2129 (N ₃ asymmetric stretching), 1666 (C=O stretching), 1554, 1483, 1242 (N ₃ symmetric stretching), no acid OH band at 3000-3500.

Embodiment 3: Ethylenediamine bis(4-azido-2,3,5,6-tetrafluorobenzenesulfonamide)

Ethylenediamine (70 mg, 1.0 mmol) and triethylamine (240 mg, 2.0 mmol) dissolved together in 8 mL of anhydrous chloroform were added to pentafluorobenzenesulfonyl chloride (710 mg, 2.2 mmol) in 2 mL of anhydrous chloroform . A white precipitate was obtained which was filtered. The filtrate was washed with 3 x 4 ml semi-saturated KCl aqueous solution, then dried over MgSO ₄ and evaporated to recover ethylenediamine bis(pentafluorobenzenesulfonamide) (V) in the form of white crystals (yield, 80%). Sodium azide (105 mg, 1.6 mmol) in 1.0 mL water and 4.0 mL acetone was then reacted with V (390 mg, 0.76 mmol) and stirred overnight on a 60°C hot plate. A white precipitate was obtained. The solution was reduced to half volume by evaporation and 2 mL of water were added. The precipitate was then filtered off, dried and recrystallized from chloroform to give ethylenediaminebis(4-azido-2,3,5,6-tetrafluorobenzenesulfonamide) (VI) in the form of white crystals. Soluble in acetone, isopropanol and xylene, but insoluble in water or hexane. FTIR: 3315 (NH stretching), 2130 (N ₃ asymmetric stretching), 1642, 1493, 1358 (SO ₂ asymmetric stretching), 1230 (N ₃ asymmetric stretching), 1169 (SO ₂ symmetric stretching), 989, There is no hydrated acid band at 1650-2800.

Example 4: 1,3-Cyclohexanediol bis(4-azido-2,3,5,6-tetrafluorobenzoate) As in Example 1, but using 1,3- Cyclohexanediol.

Example 5: 1,4-Cyclohexanediol bis(4-azido-2,3,5,6-tetrafluorobenzoate) As in Example 1, but using 1,4- Cyclohexanediol.

Example 6: Ethylene glycol bis(4-azido-2-trifluoromethyl-3,5,6-trifluorobenzoate)

In a similar manner, ethylene glycol (2.1 mmol) and triethylamine (5.0 mmol) dissolved together in 10 mL of dry ether were added to 2-trifluoromethyl-3,4 in 10 mL of dry ether, 5,6-Tetrafluorobenzoyl chloride (5.0 mmol). A white precipitate of triethylammonium chloride was obtained, which was filtered off. The filtrate was washed with 3 x 20 mL of water, dried over _MgSO4 , and evaporated to recover crude ethylene glycol bis(2-trifluoromethyl-3,4,5,6-tetrafluorobenzoate). Sodium azide (150 mg, 2.3 mmol) in 2.2 water and 3.7 mL acetone was then reacted with I (1.1 mmol) in 2 mL acetone and stirred overnight on a 60°C hot plate. The solution was reduced to half volume by evaporation, and the precipitate was filtered off and recrystallized twice from 1:5 chloroform-hexane.

(A) Crosslinking of hole-transporting-and-electron-blocking interlayers or interlayer stacks to obtain precise control over interlayer thickness and injection characteristics:

Embodiment 1 (F8BT-illuminator LED):

a. Removal of photoresist from pre-patterned ITO glass substrates by purging with acetone, isopropanol and nitrogen. The ITO surface was then exposed to oxygen plasma for 10 min in a barrel etcher (Tegal Barrel Etcher 421; typical conditions: pressure, 450 mbar; power 150 W). Poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDT:PSS) formulated with a PEDT:PSS ratio of 1:15 was then spin-coated from aqueous solution to give 60-70 nm thick films. The films were baked under nitrogen for 15 min on a hot plate set at 150°C.

b. Poly(9,9-dioctyl) formulated with 1.8% ethylene glycol bis(4-azido-2,3,5,6-tetrafluorobenzoate) (I) corresponding to the weight of TFB Fluorene-alternating-[phenylene-(N-p-2-butylphenyl)imine-phenylene]) (TFB) was then spin-coated from a 0.9 w/v% mixed-xylene solution in nitrogen, A 15-nm thick film was obtained. The film was then exposed through a photomask to 254-nm deep ultraviolet radiation for 2 min in nitrogen with an illumination of 1 mW/ ^cm2 on the film surface.

(1. The preferred interlayer polymer on the anode interface has hole-transport and electron-blocking properties. The polymer also needs to be stable and not participate in unwanted electrochemical reactions with the light-emitting polymer , this reaction degrades the electrical properties of the interface in operation. Exemplary interlayer polymers include TFB and poly(triarylamine).

(2. The preferred polymer concentration is determined by the desired thickness of the interlayer film, which can be easily found by spin coating experiments. Typically, for a target interlayer thickness of 10-20 nm, the required polymer concentration is 0.2-1w /v %, depending on polymer viscosity, and required spin speed is 2000-8000 rpm Alternatively, metered inkjet printing can be used.

(3. The preferred crosslinker-to-polymer molar ratio is determined by the number-average molecular weight of the polymer used and its distribution. The crosslinker works by increasing the effective molecular weight of the polymer in the film to when it is no longer However, if present, low molecular weight species cannot reach the gel limit and therefore still leach out in good solvents. Therefore, as the crosslinker ratio or polymer molecular weight is reduced, the retained film The fraction of thickness is small. A typical membrane retention (or gel point) profile is shown in Figure 1. For a number average molecular weight of ¹⁰⁵ , the crosslinker ratio required for reasonable membrane retention >50% is 1.5-2 w/w %. For a number average molecular weight of 106 the required ratio is 0.3-0.6 w/w %. The ratio chosen should be the lowest value to ensure the desired level of crosslinking.

(4. The preferred irradiation dose is 10-100 mJ/cm ² . The minimum dose required depends on the intrinsic absorption of the polymer at 254-nm by the filter effect. It can easily be tested by dose series to make sure.

(5. The preferred radiation wavelength is 254-nm or 248-nm. This is conveniently obtained from a low pressure mercury lamp source (254-nm) or a KrF excimer laser source (248-nm).

(6. The preferred hole-injecting interlayer thickness is controlled by optimizing electron-hole capture in the desired region of the device and can be determined experimentally. Typically, for polyfluorene-derived emitters, the optimal thickness is In the range of 5-20 nm, it is easily achieved with the method disclosed here. The film thickness can be precisely controlled between 5-300 nm. Typical results are shown in Figure 2. See Example 3 below.

c. The substrate is then optionally baked under nitrogen for 60 min on a hot plate set at 180°C.

d. The membrane was then washed with mixed xylenes on the spinner chuck (10 sec soak followed by spin removal at 8000 rpm).

e. Poly(9,9-dioctylfluorene-alternating-[benzo-2-thia-1,3-oxadiazole-4,7-diyl]) (F8BT) was then heated from 1.0w/v in nitrogen % toluene solution was spin-coated to obtain a 65-70nm thick film.

f. The substrate was then baked for 3 minutes in nitrogen on a hot plate set at 120°C.

g. Evaporate 3 nm-thick Ca followed by 120 nm-thick Al onto the film via a shadow mask at a base pressure of 2-4 x 10 ^-6 .

When the device was driven under forward bias (Ca/Al as cathode and ITO as anode), a large forward biased diode current with light emission was obtained. The device characteristics are compared in Figures 3 and 4 with similar devices without the TFB interlayer. The quantum efficiency is improved by a factor of 15, reducing from 4.25V to 3.70V with a TFB interlayer for a driving voltage of eg 1000cd/ ^m2 . The improvement in drive voltage is significant because the lifetime of the device is greatly extended. In this embodiment, the hole-transporting interlayer serves not only as a hole-transporting/electron-blocking layer but also as a buffer layer for interfacial electrochemical reactions. Therefore, the service life is increased by more than 10 times.

Example 2 (F8BT-TFB-blend-emitter LED):

As in Example 1, the different light-emitting polymer layer in step (e) was a 1:1 blend of F8BT and TFB, resulting in a 1.4 w/v% mixed-xylene solution.

The lifetime, quantum efficiency and voltage for this desired brightness are enhanced by the presence of well-defined hole-transporting- and electron-blocking interlayers at the anode interface of 10-15 nm thickness. Quantum efficiency was improved by 20-100% depending on the blend formulation but no associated voltage penalty was found. This shows that even for hole-transporting polymers formulated as blends, the fabrication of a well-defined and continuous hole-transporting layer with a thickness of 5-20 nm on the anode junction can further improve performance and stability.

Embodiment 3 (F8BT-illuminant LED):

As in Example 1, the thickness of the different TFB interlayers in step (b) was controlled between 10 nm and 30 nm with 5-nm resolution. The thickness of the interlayer was confirmed by atomic force profilometry. The current density, luminance and external quantum efficiency of the devices are shown in Figures 5, 6 and 7, respectively. This example serves to demonstrate the benefits of precise control of interlayer thickness. As a result, systematic optimization of device performance is possible.

Embodiment 4 (stacked hole injection interlayer):

As in Example 1, except that an additional layer of a second hole-transport polymer is used as the second layer in the interlayer stack. This polymer was chosen here so that 1.6% of I, based on the weight of poly(vinylcarbazole) (PVK), was compounded with PVK. The mixture was spin-coated from a 0.25 w/v% chloroform solution in nitrogen to obtain a 10-nm thick film. The film was then exposed through a photomask to 254-nm deep ultraviolet radiation for 2 min in nitrogen with an illumination of 1 mW/cm2 on the film surface.

(7. A preferred property of the second interlayer polymer at the anode interface (if used) is to provide a hole-transport level intermediate between the film of the first interlayer polymer and the film of the light-emitting polymer. bulk, while not providing any accessible electron transport levels for leakage of electrons. In this way, a trapezoidal hole-transport energy level is set to facilitate hole injection into the light emitting polymer itself.

(8. Additionally, this polymer is used as a capping layer to suppress undesired electrochemical reactions with the light-emitting polymer, which would alter the electrical properties of the interface during operation.

PVK partially meets these considerations and is therefore not ideal. Since PVK is a better electron sealer than TFB, an improvement in low pressure efficiency is still obtained. This example thus demonstrates that stacked intermediate layers can advantageously be constructed from suitable polymers.

(B) Crosslinking and photopatterning of LEP layers in LEDs:

Embodiment 5 (photopatterned OC1C10-PPV LED):

a. Removal of photoresist from pre-patterned ITO glass substrates by purging with acetone, isopropanol and nitrogen. The ITO surface was then exposed to oxygen plasma for 10 min in a barrel etcher (Tegal Barrel Etcher 421; typical conditions: pressure, 450 mbar; power 150 W). Poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDT:PSS) formulated with a PEDT:PSS ratio of 1:15 was then spin-coated from aqueous solution to give 60-70 nm thick films. The films were baked under nitrogen for 15 min on a hot plate set at 150°C.

b. with 2% ethylene glycol bis(4-azido-2,3,5,6-tetrafluorobenzoate) (I) and poly(2-methoxy-5-( 3,5-Dimethyl)octyl-1,4-phenylenevinylene) (OC10-PPV) compounded and then spin-coated from a 0.5 w/v% mixed xylene solution under nitrogen to give 47- nm thick film. The film was then exposed to 254-nm deep ultraviolet radiation through a photomask for 2 min in nitrogen with an illumination of 1 mW/ ^cm2 on the film surface and then developed with toluene.

c. The substrate is then optionally baked under nitrogen for 5 min on a hot plate set at 200°C.

d. Evaporate 3 nm-thick Ca followed by 120 nm-thick Al onto the film via a shadow mask at a base pressure of 2-4 x 10 ^-6 mbar.

When the device was driven under forward bias (Ca/Al as cathode and ITO as anode), a large forward biased diode current with light emission was obtained. The device properties are compared in Figure 8 to similar devices without crosslinker and photopatterning. Similar current densities were obtained in both cases. The ability to drive current through the device is therefore not degraded. Quantum efficiency was worse for photopatterned devices, but this was brought back to the original level by reducing the crosslinker fraction to <0.5 wt%. This example serves to show that the crosslinking process described herein can be used for LEP layers. The requisite bipolar injection and electroluminescence from LEPs do not drop strongly even at higher crosslinker ratios and small film thicknesses (which make the current-voltage characteristics particularly sensitive to injection junctions). These devices were therefore not optimized: the OC1C10-PPV thickness of 47nm was significantly less than the optimum of 60-65nm, and the crosslinker fraction was used in significant excess of what needed to be used (approximately 0.5%).

(B) Crosslinking and photopatterning of the channel layer in FETs:

Example 6 (Crosslinked/Photopatterned P3HT-Channel Bottom-Gate FET):

a. A pre-patterned p ⁺ -doped Si substrate with a 200 nm top dielectric layer of silicon dioxide and gold (Au) source-drain contact pads was passed through a barrel etcher (TegalBarrel Etcher 421, typical conditions : pressure, 450mbar; power 200W) exposed to oxygen plasma for 10min to clean, then rinsed with CMOS grade water, isopropanol, and then dried in a nitrogen jet. The pre-patterned channel length is 10 μm and channel width is 2 mm. Hexamethyldisilazane was spin-coated onto the substrate at 900 rpm for 30 s, and then the substrate was baked in air on a heating plate at 120° C. for 2 minutes.

b. With 5% ethylene glycol bis(4-azido-2,3,5,6-tetrafluorobenzoate) (I) based on the weight of P3HT with regioregular poly(3-hexylthiophene) (P3HT) was compounded and spin-coated from a 1.8 w/v% chloroform solution to give a 50-nm thick film. The film was then exposed to 254-nm deep-UV radiation through a photomask in nitrogen for 2 min with an illumination of 1 mW/ ^cm2 on the film surface, then developed with chloroform, and then dried with nitrogen.

c. The device was then annealed on a hot plate at 100° C. for 2 min in a nitrogen atmosphere.

The transfer characteristics of this FET are shown in Figure 9. The threshold gate voltage (V _th ) is about -0V. When the gate voltage (V _g ) was increased above this threshold, a robust turn-on of channel conductivity was found. An "on" source-drain channel current of 100 μA was obtained for a drain voltage (V _ds ) of -30V and a gate voltage (V _gs ) of -50V. The on-off ratio is better than 100 for a gate voltage of -50V to +50V. The FET fluidity obtained from the I _sd -V _gs slope in the linear range by the conventional equation is close to 3×10 ⁻² cm ² /Vs. Typical flowability of spin-coated initial P3HT films without crosslinker and light exposure is 10 ⁻¹ to 10 ⁻² cm ² /Vs. Forward and reverse scan features are overlappable. This example serves to demonstrate that the ideal field effect mobility can be substantially preserved even at higher crosslinker fractions after the photocrosslinking process described here. Field-effect mobility is particularly sensitive to the presence of defects and impurities that tend to separate interfaces. This finding confirms the present method as a promising means to pattern polymer transistors with high field-effect mobility. In practical devices, FET transistors will be constructed on other substrates, such as glass, polyethylene, poly(ethylene terephthalate), or other polymer materials, and have other materials as source, drain and gate contacts, such as conductive polymer type materials. The molecular weight of the field effect polymer used is about an order of magnitude larger than that used here (approximately 30,000). Therefore, the fraction of crosslinker used can be further reduced to below 1%, further enhancing the attractive features of the present process.

Example 7 (Crosslinked/Photopatterned TFB-Channel Top-Gate FET):

As in Example 6, except that the field effect polymer was used with 1.3% ethylene glycol bis(4-azido-2,3,5,6-tetrafluorobenzoate) based on the weight of TFB (I) Blended TFB used at a concentration of 1.8 w/v% (in mesitylene) instead. It was spin-coated onto the substrate at 1600 rpm for 60 s, resulting in a 30 nm film. The substrate had pre-patterned 20-nm gold (Au) source and drain contacts (staggered array) with 5 μm channel length and 10 mm channel width. The substrate was then exposed to 254 nm radiation through a photomask for 2 min in nitrogen to crosslink the TFB film, developed by a 10-s mesitylene soak, followed by spin removal at 6000 rpm for 30 s.

Bis(dimethylvinylbenzocyclobutene)disiloxane monomer (Cyclotene) in mesitylene diluted to a concentration of 12.7w/v% , Dow Chemical Company, MI, USA) at 6000rpm, 60s, was spin-coated onto the TFB film to obtain a 200nm film. It should be noted that application of this BCB type monomer/mesitylene solution will immediately redissolve the formed TFB layer if the TFB film is not crosslinked. The substrate was then baked under nitrogen for 10 s on a hot plate set at 290°C.

A top-gate electrode was obtained by applying PEDT:PSS (commercially available as "Baytron P" (RTM) from HC Starck of Leverkusen, Germany) as described in Example 1.

The transfer characteristics of this device are shown in Figure 10. The threshold gate voltage (V _th ) is about -30V. This is largely limited by trace amounts of ionic impurities in TFB field-effect polymers. When the gate voltage (V _gs ) was increased above this threshold, a robust turn-on of channel conductivity was found. An "on" source-drain channel current (I _sd ) of a few microamps was obtained for a drain voltage (V _ds ) of -20V, and (V _g -V _th ) of -30V. The on-off ratio is better than 1000. The FET fluidity obtained from the I _sd -V _gs slope in the linear range by the conventional equation is close to 4×10 ⁻⁴ cm ² /Vs. This example serves to show that photocrosslinked field-effect polymers can be advantageously combined with a wide range of subsequent solvent processing steps that would otherwise destroy the integrity of the first formed polymer film (e.g., for deposition of gate dielectric, level (interlevel) planarization layer, interconnect, etc.). The field effect fluidity of the initial spin-coated TFB film is 10 ^-3 -10 ^-4 cm ² /Vs. Flowability is thus not reduced, as in Example 6, demonstrating that the methods and processes described here are compatible with many semiconducting polymers with very different properties.

Example 8 (donor-acceptor multilayer photoconductive device):

Multiple alternating layers of hole-transport donor polymer ("PFB") and electron-transport acceptor polymer ("F8BT") were prepared on a glass substrate patterned in a laterally staggered array. The donor-acceptor terminology is introduced here to indicate that photoexcitation dissociates at the interface between these two polymers, leaving holes on the PFB side of the interface and electrons on the F8BT side of the interface. Each of these layers is approximately 40nm thick. The UV-visible absorption spectra collected at each stage of multilayer fabrication are shown in FIG. 12 . The quality of the film was judged to be excellent by the absence of light scattering in the clear window and the regular formation of the interference fringe pattern as the overall film thickness approaches the optical length scale. In fact the spectrum is structured in an orderly manner with the respective addition of PFB or F8BT films to the multilayer stack. This means that numerous subsequent polymer layers can be deposited at controlled thicknesses without compromising the integrity of the underlying layer. This would not otherwise be possible without the moiety technology for crosslinking described in this invention.

Since the exciton diffusion length is limited in organic materials, it is advantageous to have respective donor and acceptor film thicknesses smaller than the exciton diffusion length (typically 20 nm) so that the maximum charge carrier concentration can be obtained, to provide a large photoconductive response. To test this possibility, two multilayer stacks were fabricated, where the F8BT thickness was 40 nm in one stack and 80 nm in the second stack (device structure shown in Figure 13). The PFB layer thickness was kept equal (about 40 nm) in both stacks. The stack was irradiated with 457-nm (Ar-ion laser), at which wavelength only the F8BT component absorbs light. The cumulative F8BT thickness in the stack should be such that virtually all of the light traversing the stack is absorbed. The resulting photocurrent curves are shown in Figure 13, illustrating a two-fold increase in the photoconductivity of the donor-acceptor stack with the thinner F8BT layer, as expected.

This example is used to further demonstrate the application of this approach (method) in seeking the possibility of manufacturing multilayer organic electronic devices by utilizing complementary donor-acceptor and electron-hole transport properties.

Claims (22)

1. A method for forming a polymer device, comprising the steps of: (i) depositing a solution comprising (1) a semiconducting polymer or oligomer, and (2) a moiety for crosslinking on a substrate to form a layer; (ii) curing the layer formed in step (i) under conditions to form an insoluble cross-linked polymer; It is characterized in that the crosslinking moiety is present in step (i) in an amount ranging from 0.05 to 3 mol%, based on the total number of moles of repeating units of the polymer or oligomer and the crosslinking moiety in the solution. 2. A method according to claim 1, wherein the method comprises the additional step of depositing a solution comprising a second polymer on the crosslinked polymer formed in step (ii). 3. A method according to claim 1 or claim 2, wherein the method comprises the additional step of chemically modifying the insoluble crosslinked polymer. 4. The method of claim 1, wherein the polymer formed in step (ii) is patterned by curing only a portion of the layer formed in step (i) and removing uncured portions. 5. The method according to claim 4, wherein removing the uncured part is performed by washing with a solvent. 6. The method according to claim 1, wherein the crosslinking moiety is sensitive to ultraviolet radiation having a wavelength in the range of 200nm to 400nm, the conditions in step (ii) comprising exposing the layer to Ultraviolet radiation having a wavelength in the range of 200nm to 400nm. 7. The method according to claim 1, wherein the crosslinked polymer formed in step (ii) is partially conjugated, substantially conjugated or fully conjugated. 8. The method according to claim 1, wherein the crosslinked polymer formed in step (ii) is a semiconducting polymer, and wherein the semiconducting polymer comprises fluorene and/or indenofluorene and/or triarylamine and/or thiophene and/or phenylene and/or phenylene vinylene and/or substituted pyrrole repeat units. 9. The method according to claim 1, wherein the thickness of the layer deposited in step (i) is 1-500 nm. 10. The method according to claim 1, wherein said polymer device is a device selected from the group consisting of polymer LED devices, polymer waveguide LED devices, polymer distributed Bragg reflectors, polymer microporous LED devices, polymer FET devices, polymer photodetectors and polymer photovoltaic devices. 11. The method according to claim 1, wherein the crosslinking moiety is part of the main chain of said polymer or oligomer or is attached to the polymer or oligomer as a side chain. 12. The method according to claim 11, wherein said polymer or oligomer comprises a structural unit comprising the crosslinking moiety and having the general formula III or IV: where Ar _F includes substituted or unsubstituted fluorinated aryl groups and R is a structural unit in the main chain of the polymer or oligomer. 13. The method according to claim 1, wherein (2) the moiety for crosslinking is mixed with (1) the polymer or oligomer in solution. 14. The method according to claim 13, wherein the crosslinking moiety has the general formula I: N ₃ -Ar _F -N ₃ (I) wherein _ArF includes substituted or unsubstituted fluorinated aryl groups. 15. The method according to claim 13, wherein the crosslinking moiety has the general formula II: N ₃ -Ar _F -L-Ar _F '-N ₃ (II) wherein Ar _F and Ar _F ' independently each include a fluorinated aryl group and L includes an optional divalent or multivalent linking group. 16. A method according to any one of claims 13 to 15, wherein the polymer or oligomer is present in the solution in an amount of 0.5-2.5 wt% prior to the addition of the crosslinking moiety. 17. The method of claim 1, further comprising annealing the insoluble polymer formed in step (ii). 18. A polymer device obtainable by a method as defined in any one of claims 1 to 16. 19. The polymer device of claim 18, which is a waveguide LED device comprising a core layer and at least one cladding layer, wherein the core layer and the at least one cladding layer each comprise a cross-linked polymer. 20. A solution containing a semiconducting polymer and a crosslinking moiety, characterized in that the crosslinking moiety is present in an amount of 0.05-3 mol %, based on the total of repeating units of the polymer and the crosslinking moiety in the solution number of moles. 21. The solution according to claim 20, wherein the crosslinking moiety is mixed with the polymer in solution. 22. The solution according to claim 21, wherein the crosslinking moiety has the general formula II: N ₃ -Ar _F -L-Ar _F '-N ₃ where _ArF and _ArF ' are each a fluorinated aryl group and L includes an optional divalent or multivalent linking group.

Images