Technical field

[0001 ]. The present invention relates to a method for producing a conductive or semiconductive material comprising conducting or semiconducting nanotubes or nanowires, and at least an organic macromolecular material (e.g. short molecule, polymer, ... ) which can be processed in solution. The invention also relates to the resulting material comprising a conductive or semiconductive nanoscale network made of nanotubes or nanowires.

Background art

[0002]. Nanotubes and nanowires, made of carbon, metal or other

semiconducting materials have several interesting properties for electronic devices (transistors, solar cells, electrodes, etc . ). For example, single walled carbon nanotubes possess high aspect ratio and high charge carrier mobilities, making them very attractive for next generation of carbon based electronic devices. In a photovoltaic (OPV) solar cell, single walled carbon nanotubes (SWNTs) can transport charges very efficiently when placed in contact with a semiconducting polymer such as poly-3-hexylthiophene (P3HT). Moreover, the property of ultrafast charge transfer between a semiconducting polymer and a SWNT holds great promise to enhance the performance of SWNT based OPVs.

In order to produce efficient charge transfer and transport to the electrodes, the good dispersion of the nanotubes and the formation of a continuous (percolated) network, with good interconnection of tubes, is primordial. For practical reasons and ease of processing, the formation of these percolated domains should be made by solution processing. Network formation in solution is indeed low cost, can produce large area coverage and room temperature processing. Today, common solution processed methods used for electronic device applications are

spin-coating or drop-casting , which produce random networks and aggregates with inferior electrical properties. This is because In the random network, the tubes are randomly placed (not controlled), and are not optimally interconnected (Fig. 1 a). Ideally, one needs to control the formation of the networks and tube placement with nanoscale precision to increase interconnection and between tubes and connection to interfaces, e.g. electrodes, in a film or device. This nanoscale precision can be obtained by producing nano-patterned networks, which result in very efficient percolation between tubes, and which form an interconnected path from top to bottom of the film for more efficient charge transport (Fig.1 b, 1 c). Moreover, another drawback of random networks is that they require large quantities of nanotubes to form conducting pathways. This reduces electrical conductivity due to the formation of bundles which increase network resistivity (network resistivity increases with bundle diameter), it increases materials costs, and lowers the optical transparency needed for specific

applications.

[0003]. It has previously been demonstrated that the formation of nano- and micro-networks of nanotubes in a polymer matrix by a thermal method resulted in improved conduction compared to a random network, and the conductivity was increased by approx.10 fold in a semi-conducting polymer/SWNT composite at a low concentration of 0.1 1 wt.% SWNTs. However, one drawback is the use of relatively high pressures (15-50 bar) and high temperatures (150 to 200°C) may affect the properties of condition sensitive materials (e.g. organic polymers comprising delocalized (conjugated) ττ- electrons, e.g. polythiophenes which can oxidize at elevated temperatures). Temperature annealing has also been shown to modify the crystallinity, and the optical and electrical properties of polythiophene polymers. Is has been discovered that semiconductive materials formed by a method comprising low temperatures and being solvent based provide properties which are beneficial for opto-electronic applications using organic semiconductor materials. The present invention relates to a method performed at low temperature (ideally room-temperature) and low pressure which provides well defined arrays of nano-engineered nanotube/nanowire networks with much improved charge transport in organic polymers comprising delocalized (conjugated) ττ- electrons, such as , for example but not limited to, a P3HT film matrix (Fig. 2). The presented invention is controllable, scalable and enables the formation of nano- sized networks with exceptional properties not achieved by other methods. [0004]. Charge transport in P3HT was enhanced by approximately two orders of magnitude compared to a random network produced by traditional solution based methods and around 4-5 times more compared to our previous results with a thermal method. Moreover, we show that even very low nanotube loadings (much below the percolation threshold previously reported) still produced highly conductive networks due to the nanoscale interconnectivity of the nano-networks, and the improved dispersion of the nanotubes in the solvent. The low amount of nanoitubes used reduces bundling, increases transparency and provides an economical solution for electronic applications. Finally, the method is simple, fast and prevents any unwanted change in materials' properties which may be caused by the use of high temperatures

[0005]. CN101328276 discloses a method for providing SWNT composite film the method comprising the use of gum arabicum and a step where the

temperature is kept at 70 to 100 degrees C. The polymer emulsions used are exemplified as styrene-acrylic emulsions, pure acrylic emulsions and epoxy emulsions. The document does not disclose semiconductive polymers comprising monomers containing aromatic moieties. P3HT does not to disclosed. The method involves a step with temperatures way above RT.

[0006]. CN103739903A discloses high conductivity carbon nanotube rubber composite which is produced by using a latex. The document does not disclose the use of disclose semiconductive polymers comprising monomers containing aromatic moieties. P3HT apperas not to be disclosed.

[0007]. CN103073891 discloses a method for the preparation of high- conductivity flexible conductive composite materials. The method encompasses a stage where graphene and carbon nano-tube are uniformly dispersed (CNTs) in an aqueous solution and where further resorcinol, formaldehyde and catalyst (sodium carbonate) are added, whereby the reaction temperature is 85°C and the reaction time three days. A graphene -CNT-resorcinol-formaldehyde organic gel is formed. In a further step the CNTs are carbonized in a furnace at a temperature of 900-1000 degrees centigrade. The material does not contain CNT. Furthermore, the method encompasses high temperature stages. Additionally, P3 does not disclose semiconductive polymers comprising monomers containing aromatic moieties.

Invention Invention

[0008]. The present invention relates to a method for producing a conductive or semiconductive material comprising nanotubes and at least a first polymer, the method comprising the steps of:

a) dispersing nanotubes in a solution comprising at least a first solvent thereby forming a dispersion,

b) providing a composition comprising the first polymer and at least a second solvent,

c) depositing the dispersion on a substrate forming a first film,

d) depositing the composition on the first film thereby forming a second film, and e) imprinting the films with a mold providing 3D structures in the nano range, where the imprinting is conducted at a temperature below about 50°C.

[0009]. According to an embodiment, the invention relates to a method for producing a semiconductive material. Said semiconductive material is preferable applied in photovoltaics and Organic Light-Emitting Diodes (OLEDS).

[00010]. The invention is also directed to a conductive or semiconductive material obtainable by the method, and a photovoltaic component comprising the semiconductive material with SWNTs and a semiconductive polymer.

[0001 1 ]. The dispersion comprises nanotubes and at least a first solvent. The first solvent can be a mixture of several solvents. The first solvent should facilitate the formation of a nanotube dispersion which preferably also exhibits properties facilitating the formation of a film on a substrate. It is preferable if the solvent has fast to moderate evaporating properties. Hence, according to an embodiment the first solvent has a vapour pressure above about 10 mmHg (at 20°C), above about 50 mmHg, above about 100 mmHg, above about 150 mmHg, above about 200 mmHg. Typically, the properties of the first solvent is adjusted with regard to type of nanotube. Suitably, the first solvent is water, or a carbon based organic solvent, such as dichlorobenzene, chlorobenzene, trichlorobenzene, chloroform, toluene, xylene, dimethylformamide, or a fluorinated solvent or any mixtures of the exemplified solvents. The first solvent may also be selected from non-chlorinated hydrocarbons or functionalised non-chlorinated hydrocarbons, such as non- chlorinated hydrocarbons comprising from 1 to 15 carbon atoms, possibly also comprising e.g. one or more hydroxyl groups.

[00012]. According to a further embodiment, the dispersion may also comprise a second polymer. This second polymer may be the same as the first polymer comprised in the composition. However, the second polymer may also be a different type of polymer with respect to the first polymer. Suitably, first and second polymers are of the same type.

[00013]. The presence of a second polymer in the dispersion may facilitate the nanotubes to stay on the substrate during the film forming process.

[00014]. The dispersion can have a concentration of nanotubes from about 0.00001 mg/ml up to about 10 mg/ml, suitably from about 0.0001 mg/ml up to about 10 mg/ml, suitably from about 0.0001 mg/ml up to about 5 mg/ml.

[00015]. The first polymer of the present method can be selected from inorganic polymers, organic polymers or even mixtures thereof. According to an embodiment the first polymer is selected from organic polymers.

[00016]. According to a further embodiment, the first polymer is selected from conducting or semi-conducting polymers (with delocalized π electrons). Semiconducting polymers can be used with this method to produce electronic devices such as photovoltaic components. According to an embodiment the first polymer is selected from semi-conducting polymers. The semi-conducting polymers may be organic polymers comprising delocalized (conjugated) ττ- electrons. Delocalized ττ- electrons may also be referred to as conjugated π electrons. Conjugated ττ- electrons or a conjugated system is a system of overlapping p-orbitals (bridging intervening sigma bonds) with delocalized electrons in compounds with alternating single and multiple, often double, bonds which normally decreases the overall energy of the compound and stability. The polymer comprising delocalized

(conjugated) ττ- electrons may comprise heterocyclic aromatic monomers substituted with a moiety providing some steric properties to the overall polymeric network. Typically, the moiety may be a straight or branched alkyl group, suitably straight alkyl group. The polymer comprising delocalized (conjugated) ττ- electrons is suitably a polymer comprising heterocyclic aromatic monomers such as alkyl substituted thiophene monomers. An example is a polythiophene polymer such as poly(3-hexylthiophene-2,5-diyl) also referred to as P3HT. The polymer has suitably an average molecular weight form about 3-300 kg/mol.

[00017]. According to the process a composition is deposited on the first film. The composition comprises the first polymer and at least a second solvent. This second solvent may be a mixtures of several solvents. It may also be similar or identical to the first solvent. Preferably, the second solvent will not evaporate too fast. Hence, according to an embodiment the second solvent may have a vapour pressure below about 200 mmHg (at 20°C), below about 150 mmHg, below about 100 mmHg, below about 50 mmHg. The properties of the second solvent are chosen such to admitting dispersion of the first polymer and good film formation. The second solvent may be selected form any of the exemplified first solvents, such as water, or a carbon based organic solvent, such as dichlorobenzene, chlorobenzene, trichlorobenzene, chloroform, toluene, xylene, dimethylformamide, or a fluorinated solvent or any mixtures of the exemplified solvents. According to an embodiment, the second solvent is a mixture of chloroform and 1 ,2- dichlorobenzene. Suitably, the ratio of chloroform and 1 ,2-dichlorobenzene is in the range of from 95:5 to 60:40 based on weight. Further, it is preferred that the composition has a concentration of polymer below 8 wt %, preferably between 1 and 5 wt%.

[00018]. The method comprises providing the dispersion and the composition, depositing the dispersion on a substrate thereby forming a first film, depositing the composition on the first film thereby forming a second film and thereafter imprinting the films, while the second film (composition) is still partially wet, with a mold providing patterns (three dimensional structure) in the nano-range, wherein the imprinting is conducted at a temperature below about 50°C. Suitably, the substrate and films are compressed during the imprint. One of the benefits is that the method is conducted a low temperatures, such as below about 45°C, such as below about 40°C, below about 35°C, below about 30°C, such as up to about 10°C. As the method is carried out at low temperatures semiconducting polymers can been used which are temperature sensitive.

[0001 ] After deposition of the dispersion comprising the nanotubes and the composition comprising the first polymer forming the first and second film, the films are imprinted. By the imprinting process the films obtain a specific three

dimensional geometric structure said structure increasing the charge mobility normal (vertical) to the substrate plane provided that first polymer is a semi- conductive organic polymer. The three dimensional geometric structure of the film is given by the three dimensional geometric structure of the mold. The three dimensional geometric structure of the film has an implication on the properties of the conductive or semiconductive material. The term nano range implies a value of any dimensional property such as one dimensional properties e.g. length, diameter, radius from around 1 nm up to around 100 pm. The three dimensional geometric structure of the film should exhibit patterns with a height of from about 10 nm up to about 10 pm, suitably from about 30 nm up to about 10 pm. Suitably, the shortest length of the pattern is from 30 nm up to about 10 pm. According to an embodiment, the three dimensional geometric structure of the film has the shape of pillars. The pillars preferably have an average cross-section area of from about 10 ^"18 up to 10 ^"6 , suitably from about 10 ^"18 to about 10 ^"7 , preferably from about 10 ^"16 to about 10 ^" 8. The pillars have a height of from about 5 nm up to about 10 pm, suitably from about 50 nm up to about 10 pm. Suitably, the pillars have a circular cross-section with a diameter of from 5 ^* 10 ^"9 to about 100 ^* 10 ^"6 .

[00019]. Suitable methods of depositing the dispersion and composition for forming films include spin-coating, drop-casting, spray-coating or blade-coating. [00020]. The method also comprises imprinting the films with a mold. The master molds can be formed by optical lithographical, electron-beam lithography, copolymer self-assembly, colloidal assembly, patterning, anodization techniques, etching, molding or nanoimprinting methods. A flexible polymer mold is then replicated from the master mold by casting as described below. A 10: 1 mixture of PDMS base and curing agent was degassed in a dessicator under rough vacuum and gently poured onto the silicon master mold, followed by curing in an oven for 5 hours at 150°C. After cooling back to room temperature, the PDMS replica mold was carefully detached and used for further imprinting steps. A film of the macromolecular material was spin-coated or drop-casted on the substrate from solution, and was imprinted with the PDMS mold until solid nano- or micro- structures, replicating the mold, have formed and created a solid patterned film. It is preferred that the mold is flexible and is fabricated from a material having an elastic Young's modulus below about 1 .8 GPa, suitably below about 1 .5 GPa, more preferably below about 1 .0 GPa. To have a flexible mold with an elastic Young's modulus as indicated seems to have a beneficial impact on the alignment of the nano networks (alignment of nanotubes and first polymer) i.e. providing good conformability. Poly(dimethylsiloxane) PDMS is a preferred material of the mold.

[00021 ]. The nanotubes or nanowires can be made of carbon, metal or a semiconducting material, and with a diameter as small as about 0.5 nm, and a length up to 100 microns. According to an embodiment, the nanotubes are Single Wall Carbon Nanotubes (SWNTs). Single Wall Carbon Nanotubes (SWNTs) can preferably have a diameter of from about 0.5 up to about 2 nm, and a length of from about 50 nm to about 1500nm.

[00022]. The substrate can comprise several layers. Usually the substrate comprises at least one layer which can be silicon, conducting oxide, plastic substrate, another organic or inorganic film coating.

[00023]. Brief description of drawings [00024]. The invention is now described, by way of example, with reference to the accompanying drawings, in which:

[00025]. Fig. 1 shows different network configurations : nanoscale network vs. random

[00026]. Fig. 2 shows schematically the method for creating the nano-engineered networks.

[00027]. Fig 3 shows mobility data for a conductive polymer as function of nanotube concentration for 3 methods: (the current solvent low temperature method, thermal method, random network).

[00028]. Fig. 4 shows an example of nano-network arrays with approximately 300 nm diameter imaged by Electron Microscopy (A), and by Atomic Force Microscopy (B,C)

[00029]. Fig 5 shows difference in bundle diameter between the room temprature solvent method and the thermal method. The solvent method results in smaller bundles of nanotubes compared to thermal method (18nm vs 23nm) and much less larger bundles, resulting in reduced electrical network resistivity.

[00030]. Fig. 6 shows the custom-made solvent imprint chamber Experimental data

[00031 ]. Example of sample preparation for a semiconducting (conjugated) polymer.

[00032]. The substrates used were highly P-doped silicon wafers. They were cleaned by dipping them in an ultrasonic bath in acetone and in IPA(what is this?). They were then dried with nitrogen and cleaned again in a flow of carbon dioxide ice crystals (snow-jet) in order to remove remaining dust particles. The single wall carbon nanotubes (SWNT) were produced by the CoMoCAT method and supplied by Sigma Aldrich. The SWNT had a diameter of approx. 0.7-0.9 nm and a length of approx.700 nm. More than 50% of them were (6,5) tubes. The tubes were used as received, without any chemical modification or doping. The P3HT (American Dye Source) had a molecular weight of 32 kg/mol and a regio-regularity of 98%. First, the SWNTs were dispersed in orthodichlorobenzene (ODCB) with a concentration of 0.13mg/ml_ by ultrasonication. The dispersion was then further diluted in order to obtain the required nanotube concentrations in a 1 : 1

chloroform: ODCB solvent mixture. A more concentrated solution of 1 .04 mg/ml_ was also made in order to be able to prepare the samples with the highest SWNT contents. 0.5 wt.% P3HT were added to the obtained solution in order to help the SWNTs stay on the substrate after spin coating. A solution of 1 .5 wt.% P3HT in a 80:20 chloroformODCB mixture was prepared for the second layer. The SWNT solution was spun on the substrate at 5000 rpm for 60 s, and the second layer was spun at 3000 rpm for 2 s. The still partially wet sample was then imprinted at room temperature (20-22°C) using a home-made solvent imprint chamber. A low pressure (1 -2 bar) was maintained during network formation and the sample was dried at room temperature using a gentle flow of nitrogen.

[00033]. Example of sample preparation for an insulating (non-conducting) polymer.

The same method as described above is employed for an insulating polymer.

Choice of appropriate solvents is required to make sure the polymer is soluble in the selected solvent. We give here specifications for a polystyrene polymer.

A first thin layer of SWNTs (mixture of semiconducting and metal- lie nanotubes) dispersed ortho-dichlorobenzene (oDCB) and polystyrene (PS) was formed on a conducting substrate (doped silicon, indium tin oxide (ITO)), and a second layer of PS only was spun on top of the first layer. Alternatively, insulating substrates (glass, plastic substrate) can also used. The SWNTs were

ultrasonicated in oDCB for better dispersion. The dispersion was then filtered and diluted into 50 vol. % chloroform in order to prepare a solution with a concentration of nanotubes between 0.00001 and 0.5 wt. % in a PS solution. The PS/SWNTs solution was spun on a substrate (highly doped silicon, ITO, polyethylene terephtalate (PET), or glass) at 2000-6000 rpm for 15 s in order to make a smooth, 15-30 nm thick layer. Next, a 7 wt. % PS solution was spun at 5000 rpm for 60 s to create an 860 nm thick film to produce a micropatterned composite film with 4 microns diameter patterns. For smaller nanopatterned networks, a thinner layer of PS was deposited from a 3wt.% PS solution spun at 3000rpm for 60s to create a 1 10 nm thick film layer with high conductivity and transparency. Patterning was performed with a mold at room temperature (20-22°C) and low pressure (1 -2 bar) using a home-made solvent imprint chamber and a gentle flow of nitrogen to dry the sample. Fig. 4 refers to results of conductivity measured with nanopatterned networks made of PS and SWNTs at different nanotube concentrations between 0.00001 and 0.5 wt. %, and compared to conductivity of the random network. The data shows much higher conductivity of the nano-network at the same

concentration, and also importantly much lower percolation threshold (conductive at much lower concentration).

[00034]. Patterning of Samples

[00035]. The samples were patterned using a exible polydimethylsiloxane (PDMS) mold. The liquid PDMS solution (Sylgard 182) was mixed in a 10: 1 ratio with its precursor and degasified in a vacuum desiccator. The still liquid mix was then poured on a patterned silicon master mold and placed in an oven at 150°C for several hours. The cured PDMS was then peeled of the silicon master and used as a flexible mold. The imprint chamber is shown in Fig. 2. The flexible mold was attached to a moving 2 piston while the sample to be patterned was placed at the bottom of the chamber. A weight was placed on the piston before it is slowly lowered down on the sample at room temperature, adding a low pressure of 1 to 2 bar. Nitrogen then flowed through the chamber in order to dry the polymer film, allowing subsequent demolding of the sample. The resulting features were characterized using optical microscopy, atomic force microscopy and scanning electron microscopy.

[00036]. Electrical characterization The conductivity of the samples was characterized with home-made smooth and planar flexible electrodes, prepared from a smooth PDMS stamp covered with 150 nm silver or gold layer (Kurt Lesker PVD 75 evaporation deposition system). The electrode was made flexible so that it was possible to obtain good conformal contact between the electrode and the top of the patterns, which was not possible using traditional rigid electrodes.

A voltage was applied between the top and the bottom of the films with a small step size, and the current was measured using a sourcemeter (Keithley 2450). The current was measured vertically from the bottom of the conductive substrate to the top of the patterns. This was repeated 10 times in each location, and in 3 different locations on each sample for consistency. The data was reproducible within 10-15%. The conductivity, and/or charge carrier mobility were extracted from the l/V characteristics. The resulting conductivity data was deduced using voltage applied, current measured and the geometry of the sample measured by electron microscopy and atomic force microscopy characterization, with less than 5% error of measurement. The accuracy of measurement on the current measured and voltage applied were within 1 % each. The conductivity of the samples depends strongly on the amount of nanotubes added to the dispersion. A conductivity of approximately 0.01 S/m was obtained at low concentration of SWNTs (0.001 wt% of polymer) in the nano-network in polystyrene, whereas the random network gave no current at the same concentration. A similar conductivity of 0.01 S/m was obtained in the random network by increasing the concentration of nanotubes by 5000 times. This shows the advantage of the method described in reducing the amount of nanotubes to still obtain good conductivity at very low concentration thanks to the formation of a percolated nanoscale network.

The same property was observed and measured in nano-networks made with semi-conducting, conjugated, polymers, and the semiconductive mobility was always much higher in the nano-network compared to the random network at any concentration of nanotubes (Fig 3).