A method of change carrier density and an electronic device manufactured according to said method

Abstract

A method of producing an electronic device, the method comprising: modifying a charge carrier density of an electrically conductive material, the electrically conductive material comprising an ionic material, wherein modifying the charge carrier density of the electrically conductive material comprises modifying the charge carrier density while the electrically conductive material is in a state such that ionic motion of the ionic material is frozen. An electronic device is also described.

Description

[0001] A METHOD OF CHANGE CARRIER DENSITY AND AN ELECTRONIC DEVICE MANUFACTURED ACCORDING TO SAID METHOD

[0002] FUNDING INFORMATION

[0003] The project leading to this application has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 101020872).

[0004] TECHNICAL FIELD

[0005] The present disclosure relates to a method for producing an electronic device, and an electronic device. In particular, the electronic device may be a thermoelectric device.

[0006] BACKGROUND

[0007] Conducting polymers are emerging as an important class of tuneable electrically conductive materials (e.g. tuneable metals) for neuromorphic computing, bioelectronics and thermoelectrics. However, the charge carrier transport physics of these materials at high charge carrier densities remain poorly understood.

[0008] SUMMARY

[0009] The present disclosure relates to a method for producing an electronic device (e.g. a thermoelectric device), and an electronic device (e.g. a thermoelectric device), that may exhibit improved conductivity, and / or an enhanced Seebeck coefficient, relative to known examples of such devices. In particular, electronic devices according to the present disclosure may exhibit increased conductivity and an increased Seebeck coefficient relative to known examples.

[0010] An electronic device produced according to the present disclosure may comprise an electrically conductive material, which may be an organic material (e.g. a polymer, or an organic semiconductor), into which an ionic material (e.g. a dopant) has been introduced. The ionic material (which may comprise a first dopant, as referred to herein), the concentration (e.g. doping concentration, doping level, or a doping density) of the ionic material, and the operating temperature of the electronic device may be selected such that, when charge carriers are added into (e.g. injected), or removed from, the material (e.g. by electrical injection, optical injection, and / or by the introduction of a (second) dopant), the additional charge carriers experience a high degree of delocalization. The high degree of delocalization may be caused by the freezing of the ionic motion of the ionic material, which may result in enhanced transport properties of the charge carriers.

[0011] In some examples, enhanced transport properties of the charge carriers may attributable to extreme band filling, and / or “freezing” of a soft Coulomb gap of the material.

[0012] Adding or removing charge carriers (e.g. via a dopant) may result in a shift of the Fermi energy, Ep, out of equilibrium and into the rapidly changing part of the density of states (DOS) in the soft gap.

[0013] Described herein is a method of producing an electronic device, the method comprising: modifying a charge carrier density of an electrically conductive material, the electrically conductive material comprising an ionic material. Modifying the charge carrier density of the electrically conductive material may comprise modifying the charge carrier density while the electrically conductive material is in a state such that ionic motion of the ionic material is frozen; and / or modifying the charge carrier density of the electrically conductive material may comprise modifying the charge carrier density under conditions wherein spatial locations and / or orientations of ions of the ionic material are unchangeable in response to said modifying of the charge carrier density.

[0014] For example, the ionic motion of the ionic material may be considered to be frozen when spatial locations and / or orientations of ions in the ionic material are unchangeable in response to the modifying of the charge carrier density.

[0015] The method may further comprise introducing the ionic material into the electrically conductive material. For example, the ionic material may be introduced into the electrically conductive material by any suitable technique, for example by chemical doping, by ion exchange doping, and / or electrochemically introducing the ionic material into the electrically conductive material. The ionic material may be introduced into the electrically conductive material before modifying the charge carrier density of the electrically conductive material.

[0016] In some examples, the electrically conductive material may be formed by introducing the ionic material into a material that is substantially not electrically conductive.

[0017] Alternatively or in addition to being in a state such that ionic motion of the ionic material is frozen and / or wherein spatial locations and / or orientations of ions of the ionic material are unchangeable in response to said modifying of the charge carrier density, the method may comprise modifying the charge carrier density of the electrically conductive material while under extreme band filling conditions. For example, when the ionic material has been introduced to the electrically conductive material to a concentration of at least one ion per repeat unit.

[0018] A device according to the present disclosure may comprise a light emitting diode (LED), a laser diode, a solar cell, and / or a transistor (e.g. an organic electrochemical transistor (OECT)).

[0019] The electronic device may be a thermoelectric device.

[0020] The electrically conductive material may comprise a semiconductor.

[0021] The electrically conductive material may comprise an organic semiconductor, a molecular semiconductor, a polymer (e.g. a conjugated polymer), a two-dimensional material and / or a disordered metal. Preferably, the electrically conductive material is a material capable of exhibiting a Coulomb gap.

[0022] The ionic material may be introduced into the electrically conductive material while the electrically conductive material is in such a state as to allow for ionic motion.

[0023] To the surprise of the inventors, slightly varying the carrier density of a highly doped film using field effect shows a significant current increase for both positively and negatively increasing gate voltages at very high ionic doping levels. Without wishing to be bound by theory, this may be attributable to a freezing of ionic motion at low temperatures, resulting in a frozen soft Coulomb gap in the density of states (DOS). Under these conditions, these (additional) charge carriers may be surprisingly delocalized.

[0024] A device produced according to the methods described herein may therefore exhibit increased conductivity compared to existing devices, for example devices based on conductive polymers. In the case of thermoelectric devices, a device produced according to the methods described herein may also exhibit an increased Seebeck coefficient. This is a surprising result, as usually (e.g. in the case of polymer-based devices) the increased doping levels needed to achieve a desired conductivity result in a reduced Seebeck coefficient. Devices produced according to the present disclosure may not suffer from this trade-off.

[0025] The ionic material may have a first transition temperature. The first transition temperature may be a temperature below which the ionic motion of the ionic material is frozen, and / or the first transition temperature may be a temperature below which spatial locations and / or orientations of ions of the ionic material are unchangeable in response to modifying the charge carrier density. The method may comprise cooling the electrically conductive material to a temperature below the first transition temperature. For example, the method may comprise cooling the electrically conductive material to a temperature below the first transition temperature prior to modifying the charge carrier density of the electrically conductive material.

[0026] A device produced according to the present disclosure may need to be maintained at a temperature below the first transition temperature, including during operation, in order to maintain the advantageous effects of the methods described herein.

[0027] The electrically conductive material may be brought into a state such that the ionic motion of the ionic material is frozen by bringing the electrically conductive material to a temperature that is below the first transition temperature.

[0028] Modifying the charge carrier density of the electrically conductive material may comprise doping the electrically conductive material with a dopant. The dopant may have a second transition temperature, the second transition temperature being a temperature below which ionic motion of the dopant is frozen. The second transition temperature may be lower than the first transition temperature. In some examples, the “dopant” may be referred to as a “second dopant”, and the ionic material may be referred to as a “first dopant”.

[0029] Providing the (second) dopant may result in a shift of EF out of equilibrium.

[0030] Doping the semiconductor material with the dopant, having a transition temperature lower than the transition temperature of the ionic material, or first dopant, may advantageously provide highly delocalized charge carriers in the bulk.

[0031] Dopants and / or ionic materials according to the present disclosure may comprise ionic liquids and / or ionic salts. Dopants and / or ionic materials may, in some examples, comprise redox active molecular dopants (i.e. neutral molecules that charge transfer with the polymer and become dopant ions, e.g. F4TCNQ) or neutral molecules which undergo a chemical reaction to generate ions (e.g. NOPF6).

[0032] An electrically conductive material according to the present disclosure can include a polymer (e.g. a conjugated polymer), a polymer semiconductor, a molecular semiconductor, a two-dimensional material, an organic semiconductor, an ion conductor (e.g. an inorganic ion conductor such as manganese cobalt oxide), and / or an inorganic semiconductor (e.g. a transition metal dichalcogenide or a perovskite).

[0033] Modifying the charge carrier density of the electrically conductive material may comprise electrically injecting charge carriers.

[0034] In some examples, modifying the charge carrier density of the electrically conductive material can comprise optical injection, e.g. by photoexcitation of carriers (for example using a laser or other light source).

[0035] In some examples, modifying the charge carrier density of the electrically conductive material can comprise modulation doping of the electrically conductive material.

[0036] Modifying the charge carrier density of the electrically conductive material may comprise neutralizing some of the ionic material, for example by a chemical reaction, exposure to light, etc. The electrically conductive material may comprise a polymer (e.g. a conjugated polymer). Advantageously, a polymer may provide a wide range of carrier concentration tuneability. Further advantageously, polymers may be easily integrated into device architectures (e.g. OECTs). In addition, conducting polymers may be particularly useful in applications for neuromorphic computing, bioelectronics, and / or thermoelectrics.

[0037] The method may comprise introducing the ionic material such that a concentration of the ionic material in the electrically conductive material is at least one ion per repeat unit of the polymer. In some examples, the method may comprise introducing the ionic material such that a concentration of the ionic material in the electrically conductive material is more than one ion per repeat unit of the polymer.

[0038] The method may comprise introducing the ionic material such that a concentration of the ionic material in the electrically conductive material is such that, while the electrically conductive material is in a state such that ionic motion of the first dopant is frozen, the device exhibits a non-linear transfer curve. For example, the device may comprise a source terminal, a drain terminal, and a gate terminal (e.g., the device may be a field effect transistor (FET) and / or an OECT), and a transfer curve measured for the device may include a non-linear component. For example, the inventors have found that devices produced according to the present disclosure can surprisingly exhibit a symmetric, non-linear, component below the transition temperature of the first dopant. Without wishing to be bound by theory, the non-linear component may be indicative of delocalized charge carriers, as described in more detail herein. (It will be appreciated that a delocalized charge carrier may, in some examples, be considered as more delocalized than when the ions are at equilibrium, for example the delocalized charge carriers may still be considered localized relative to a more ordered material).

[0039] The electrically conductive material may comprise a highest occupied molecular orbital (HOMO) and a next-to-highest occupied molecular orbital (HOMO-1). An energy difference between the HOMO and the HOMO-1 may be smaller than a width of each of the HOMO and the HOMO-1 . The electrically conductive material may comprise a lowest unoccupied molecular orbital (LIIMO), and a next-to-lowest unoccupied molecular orbital (LLIMO+1). An energy difference between the LIIMO and the LLIMO+1 may be smaller than a width of each of the LIIMO and the LLIMO+1 . For example, the energy difference between the HOMO and the HOMO-1 (and / or the LIIMO and the LLIMO+1) may be a maximum of half of the energy width of either of the HOMO or the HOMO-1 (the LIIMO or the LLIMO+1).

[0040] In some examples, the electrically conductive material may comprise indacenodithiophene-co-benzothiadiazole (IDTBT).

[0041] In some examples, the method further comprises electrically connecting the electrically conductive material to a source terminal and / or a drain terminal. In some examples, the device may also comprise a gate terminal, for example to form a transistor (e.g. a FET and / or an OECT). It will be appreciated that a FET may comprise a dielectric between the electrically conductive (e.g. semiconductor) material and a gate terminal. An OECT may comprise an electrolyte between the electrically conductive (e.g. semiconductor) material and a gate terminal.

[0042] Additionally described is an electronic device produced according to the methods described herein.

[0043] The electronic device may be, for example, a thermoelectric device, a LED, a laser diode, a solar cell, and / or a transistor (e.g. an OECT).

[0044] Also described herein is a method of generating electricity. The method may comprise producing an electronic device according to any of the methods described herein. The electronic device may be a thermoelectric device. The method may further comprise applying a thermal gradient across the electronic device. The thermal gradient may be applied across the electrically conductive material while the electrically conductive material is in a state such that ionic motion of the ionic material is frozen, and / or under conditions wherein spatial locations and / or orientations of ions of the ionic material are unchangeable in response to said modifying of the charge carrier density. The temperature at which the thermal gradient is applied may be below the first transition temperature. In some examples, the temperature at which the thermal gradient is applied may be below the second transition temperature.

[0045] Advantageously, a thermoelectric device produced according to the present disclosure may exhibit enhanced conductivity (e.g. due to the delocalized carriers) as well as an enhanced Seebeck coefficient, which may result in enhanced electricity generation (e.g. an enhanced electromotive force, e.g. generated by the Seebeck effect) in comparison to known thermoelectric devices having the same or similar compositions.

[0046] Further described herein is an electronic device comprising an electrically conductive material. The electrically conductive material may be doped with an ionic material. The ionic material may have a first transition temperature. The first transition temperature may be a temperature below which ionic motion of the ionic material is frozen. The first transition temperature may be a temperature below which spatial locations and / or orientations of ions of the ionic material are unchangeable in response to a modification of the charge carrier density. The semiconductor material may be further doped with a dopant. The dopant may have a second transition temperature. The second transition temperature may be a temperature below which ionic motion of the dopant is frozen. The second transition temperature may be a temperature below which spatial locations and / or orientations of ions of the dopant are unchangeable in response to a modification of the charge carrier density. The second transition temperature may be lower than the first transition temperature.

[0047] For example, the ionic motion of the ionic material may be considered to be frozen when spatial locations and / or orientations of ions in the ionic material are unchangeable in response to the modifying of the charge carrier density.

[0048] A device according to the present disclosure may comprise a light emitting diode (LED), a laser diode, a solar cell, and / or a transistor (e.g. an organic electrochemical transistor (OECT)).

[0049] The electronic device may be a thermoelectric device.

[0050] In some examples, the device may comprise a source terminal, a drain terminal, and / or a gate terminal. The source terminal, the drain terminal, and / or the gate terminal may be electrically connected to the semiconductor material.

[0051] The ionic material dopant may result in a high degree of band filling. The device may be configured to operate at a temperature below the first transition temperature such that the soft Coulomb gap may be frozen. The dopant may therefore result in highly delocalized charge carriers. Furthermore, the dopant may be selected such that EF is moved into the rapidly changing part of the DOS in the soft gap.

[0052] The electrically conductive material may comprise a highest occupied molecular orbital (HOMO) and a next-to-highest occupied molecular orbital (HOMO-1). An energy difference between the HOMO and the HOMO-1 may be smaller than a width of each of the HOMO and the HOMO-1 . The electrically conductive material may comprise a lowest unoccupied molecular orbital (LIIMO), and a next-to-lowest unoccupied molecular orbital (LLIMO+1). An energy difference between the LIIMO and the LLIMO+1 may be smaller than a width of each of the LIIMO and the LLIMO+1 . For example, the energy difference between the HOMO and the HOMO-1 (and or the LIIMO and the LLIMO+1) may be a maximum of half of the energy width of either of the HOMO or the HOMO-1 (the LIIMO or the LUMO+1).

[0053] In another example, a device according to the present disclosure may comprise an electrically conductive material (e.g. an electrically conductive organic material), wherein the electrically conductive material comprises: a highest occupied molecular orbital (HOMO), and a next-to-highest occupied molecular orbital (HOMO-1).

[0054] The HOMO may comprise a first energy distribution distributed around a first binding energy (e.g. a HOMO binding energy). The HOMO-1 may comprise a second energy distribution distributed around a second binding energy (e.g. a HOMO-1 binding energy). The second binding energy may be higher than the first binding energy. A Fermi energy of the electrically conductive material may be closer to the second binding energy than to the first binding energy.

[0055] An energy difference between the first binding energy and the second binding energy may be smaller than a width of each of the first energy distribution and the second energy distribution.

[0056] Alternatively, or in addition, the electrically conductive material may comprise: a lowest unoccupied molecular orbital (LIIMO), and a next-to-lowest unoccupied molecular orbital (LUMO+1). The LUMO may comprise a third energy distribution distributed around a third binding energy. The LUMO+1 may comprise a fourth energy distribution distributed around a fourth binding energy. A Fermi energy of the electrically conductive material may be closer to the fourth binding energy than to the third binding energy.

[0057] An energy difference between the third binding energy and the fourth binding energy may be smaller than a width of each of the third energy distribution and the fourth energy distribution.

[0058] A device comprising such an electrically conductive material may particularly benefit from extreme band filling effects allowing access to conduction states in the HOMO- 1 / LUMO+1.

[0059] In some examples, the Fermi energy may be in an energy region in which energy states at the Fermi level are dominated by states from the HOMO-1 / the LLIMO+1.

[0060] The electrically conductive material may comprise a polymer (e.g. a conjugated polymer). Advantageously, a polymer may provide a wide range of carrier concentration tuneability. Further advantageously, polymers may be easily integrated into device architectures (e.g. OECTs). In addition, conducting polymers may be particularly useful in applications for neuromorphic computing, bioelectronics, and / or thermoelectrics.

[0061] The electrically conductive material may be doped with the ionic material to a concentration (e.g. a doping concentration, doping level, or doping density) of at least one ion per repeat unit (e.g. per repeat unit of the polymer). In some examples, the electrically conductive material may be doped with the ionic material to a concentration of more than one ion per repeat unit (e.g. per repeat unit of the polymer).

[0062] The electrically conductive material may be doped with the ionic material to a doping level such that, while the semiconductor material is at a temperature below the first transition temperature, the device exhibits a non-linear (e.g. field effect) transfer curve. For example, the device may comprise a source terminal, a drain terminal, and a gate terminal (e.g., the device may be an OECT), and a transfer curve measured for the device may include a non-linear component. For example, the inventors have found that devices produced according to the present disclosure can surprisingly exhibit a symmetric, non-linear, component below the transition temperature of the ionic material. Without wishing to be bound by theory, the non-linear component may be indicative of delocalized charge carriers, as described in more detail herein.

[0063] In some examples, the electronic device is a thermoelectric device configured to generate electricity when a thermal gradient is applied across the semiconductor material while the electronic device is at a temperature below the first transition temperature.

[0064] Advantageously, a thermoelectric device according to the present disclosure may exhibit enhanced conductivity (e.g. due to the delocalized carriers) and / or an enhanced Seebeck coefficient, which may result in enhanced electricity generation (e.g. an enhanced electromotive force, e.g. generated by the Seebeck effect) in comparison to known thermoelectric devices having the same or similar compositions.

[0065] In some examples according to the present disclosure, the electrically conductive material described herein comprises a side-chain-free polymer (e.g. a side-chain-free conjugated polymer), i.e. a polymer without side chains. A side-chain-free polymer material may be particularly suited to devices for operation at room temperature. Examples of side-chain-free polymers include poly(3,4-ethylenedioxythiophene) (PEDOT) materials such as poly(3,4-ethylenedioxythiophene):tosylate (PEDOT:Tos) and poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT: PSS), as well as materials such as poly(benzimidazobenzophenanthroline) (BBL) and poly(benzodifurandione) (PBFDO).

[0066] As described herein, the inventors have observed non-linear transfer curves at room temperature for devices wherein the electrically conductive material comprises a side- chain-free polymer, indicating that carriers are delocalised at room temperature.

[0067] BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Some embodiments of the disclosure will now be described by way of example only and with reference to the accompanying figures, in which:

[0069] Figure 1a illustrates an example structure of an ion-gated OECT device as described herein; Figure 1b illustrates a transfer curve for an IDTBT OECT;

[0070] Figure 1c illustrates the chemical structures of IDTBT, BMP TSFSI, and PVDF-HFP;

[0071] Figure 1d shows a plot of Seebeck coefficient vs. temperature for IDTBT at various doping levels;

[0072] Figure 1e illustrates a Seebeck-conductivity plot at 200 K for the data shown in Figure 1d;

[0073] Figure 1f illustrates the DOS of IDTBT calculated by density functional theory;

[0074] Figure 2a schematically illustrates a dual-gated OECT experiment as described herein;

[0075] Figure 2b shows the electrical conductivity of the polymer Figure 2b shows six selected back-gate transfer curves (insets) measured at different doping states along the IDTBT OECT transfer characteristics (main panel) at 160 K;

[0076] Figure 2c shows back-gate transfer curves for an ex-situ doped IDTBT device between 300 K and 220 K exhibiting non-linear behaviour at low T;

[0077] Figure 3 illustrates spectroscopic evidence for enhanced delocalization of nonequilibrium states;

[0078] Figure 4a illustrates a back-gate transfer curve of IDTBT at the measurement temperature of 190 K;

[0079] Figure 4b shows thermovoltage measurements at various back-gate voltages;

[0080] Figure 5a shows the molecular structure of DPP-BTz;

[0081] Figure 5b shows an OECT transfer curve of a typical BMP-TFSI gated device;

[0082] Figure 5c shows the temperature dependence of the Seebeck coefficient for the BMP- TFSI gated device;

[0083] Figure 5d illustrates a DFT band structure of DPP-BTz;

[0084] Figure 6a shows the molecular structure of PBTTT;

[0085] Figure 6b shows an OECT transfer curve of a typical BMP-TFSI gated device;

[0086] Figure 6c shows the temperature dependence of the Seebeck coefficient for the BMP- TFSI gated device;

[0087] Figure 6d illustrates a DFT band structure of PBTTT;

[0088] Figure 7 shows ultraviolet photoemission spectra of PBTTT, DPPBTz, and IDTBT;

[0089] Figure 8 shows X-ray photoemission spectra of PBTTT, DPPBTz, and IDTBT;

[0090] Figure 9 illustrates cyclic voltammogram and the corresponding conductivity at different doping stages for PBTTT, DPP-BTz, and IDTBT;

[0091] Figure 10 illustrates equilibrium and non-equilibrium carrier mobilities in IDTBT over 14 doping states; Figure 11 illustrates non-equilibrium transport signatures in double-gated DPPBTz and PBTTT measurements;

[0092] Figure 12 schematically illustrates an example method of producing an electronic device according to the present disclosure;

[0093] Figure 13 schematically illustrates an example of a thermoelectric device according to the present disclosure; and

[0094] Figure 14 illustrates a field-effect transfer curve obtained for a field-effect transistor comprising a poly(3,4-ethylenedioxythiophene):tosylate (PEDOT:Tos) at room temperature.

[0095] DETAILED DESCRIPTION

[0096] Conducting polymers are emerging as an important class of tuneable electrically conductive materials (e.g. tuneable metals) for neuromorphic computing, bioelectronics and thermoelectrics. At carrier concentrations approaching one charge per repeat unit, polymer charge transport is governed by electron-electron and electron-ion interactions with dopant counter-ions. Fundamental aspects of such correlated electron physics, such as the potential formation of an interaction-driven, soft Coulomb gap in the density of states (DOS) tied to the Fermi level, have until recently been poorly understood.

[0097] In the present disclosure, the inventors demonstrate that the correlated transport physics can be studied over a very wide range of carrier concentrations in p-type, organic electrochemical transistors (OECTs) by inducing extreme band filling, i.e., removing all electrons from the highest occupied molecular orbital and even accessing deeper energy levels without degradation. By adding a second, field-effect gate electrode to the OECT, additional electrons / holes can be injected at set charge states. Under conditions where ionic motions are frozen, these additional carriers are surprisingly delocalised, and their non-equilibrium transport signatures provide unique insight into the formation of a frozen, soft Coulomb gap in the DOS. In the present disclosure, the inventors therefore demonstrate that the transport properties of these materials can be enhanced so that significantly higher conductivities are achievable.

[0098] The charge transport physics of conjugated polymers has been studied intensively since the Nobel Prize winning discovery of electrical conduction in doped polyacetylene. At low charge densities (e.g. < 1018- 1019cm-3), that are relevant for light-emitting diodes, solar cells or transistors, it is governed by static energetic disorder caused by structural heterogeneity and dynamic disorder due to electron-phonon coupling. The recent surge of interest in OECTs and polymer thermoelectrics for waste heat harvesting has highlighted the need to better understand also the charge transport at much higher carrier concentrations (e.g. > 102° - 1021cm3), on the order of one charge per molecular repeat unit. A similar transport regime is encountered in charge-transfer or radical ionbased molecular conductors, but is easier to understand due to the availability of singlecrystals. The (quasi) one-dimensional TT-TT stacking of these organic metals leads to strongly correlated, Luttinger liquid behaviour in the metallic phase at high temperatures. The strong electron-electron, electron-phonon and electron-ion interactions cause phase transitions into charge / spin-density waves or superconducting ground states at low temperatures, which can open up energy gaps at the Fermi level, EF, and determine the transport properties. In conducting polymers such interaction-induced gap formation has also been proposed, but is much harder to study as it tends to be masked or suppressed by structural heterogeneity and disorder.

[0099] However, polymers also offer two key advantages for achieving fundamental understanding of this high carrier density regime: the ability to tune carrier concentration over a wide range and the ease of integration into device architectures. In an electrolytegated OECT, a particular doping level can be dialled-in by applying an ion gate voltage, VIG, and charge transport at this doping level can then be studied by different techniques. There have been very recent reports of n-type and p-type OECTs, in which the lowest unoccupied molecular orbital (LIIMO) or HOMO can be completely filled with electrons or holes, respectively. However, in most materials such extreme band filling has been limited by the onset of electrochemical degradation.

[0100] In the present disclosure, the inventors demonstrate in particular that in a class of p-type, donor-acceptor polymer-based OECTs band filling can stably be driven even further allowing access to conduction states in the HOMO-1. By combining this with a new experimental method, that adds a second field-effect gate electrode to the OECT ion gate, the inventors demonstrate a powerful tool for studying the correlated electron physics of conducting polymers over a very wide range of doping levels. It will be appreciated that the experimental results described in the present disclosure demonstrate new and general strategies for enhancing the transport properties of electrically conductive materials, in particular polymers (e.g. conjugated polymers).

[0101] Figure 1a illustrates an example structure of an ion-gated OECT device employed in the experiments described herein. These OECTs comprise large side gates to induce ion injection from a solid ion gel based on BMP TFSI I PVDF-HFP (Figure 1c) coated on top of the polymer channel. The devices are equipped with multi-functional electrodes for accurate measurement of four-point-probe conductivity and thermoelectric Seebeck coeffcient. The present disclosure focuses on three representative polymers: indacenodithiophene-co-benzothiadiazole (IDTBT), a donor-acceptor copolymer that exhibits high carrier mobilities in organic field-effect transistors (FET) due a low degree of energetic disorder despite exhibiting only local chain alignment and a high paracrystallinity g » 25%; poly(2,5-bis(3-alkylthiophen-2-yl)thieno(3,2-b)thiophene) (PBTTT), a polythiophene-based, semicrystalline polymer that can be doped to conductivities > 1000 S / cm; and poly[[2,5-bis(2-octadecyl)-2,3,5,6-tetrahydro-3,6- diketopyrrolo[3,4-c]pyrrole-1 , 4-d iy l]-alt-(2-octy I nony l)-2 , 1 ,3-benzotriazole] (DPP-BTz), a semi-crystalline, low band gap copolymer also with high FET mobilities.

[0102] Figure 1 b illustrates a transfer curve for an IDTBT OECT. Figure 1c shows the chemical structures of IDTBT, BMP TFSI, and PVDF-HFP. Figure 1d shows the Seebeck coefficient vs. temperature for IDTBT at various doping levels. Figure 1e is Seebeck- conductivity plot (200K) for the data shown in Figure 1d. Lines are a guide to the eye. Figure 1f shows the DOS for IDTBT calculated by DFT. Doping level n indicates number of dopant ions per IDTBT monomer.

[0103] As discussed in more detail below, IDTBT shows unusual OECT behaviour not previously reported: the inventors observed a peak in the drain current around IG = -0.6V followed by a drop. However, rather than reaching a fully insulating state as seen previously in other polymers, the inventors instead observed a second current rise when VIG < -1V (Figure 1 b). In the present disclosure, three different transport regimes are referred to: Regime I at low doping levels up to the peak, regime II between the peak and the valley / insulating state, and regime III beyond the valley. Despite IDTBT’s high FET mobilities, the conductivity at the peak is only on the order of 1 S / cm, but in regime III, significantly higher conductivities up to 10 S / cm are reached. To better understand the three regimes, the inventors measured the temperature dependence of conductivity o and Seebeck coefficient S as a function of VIG. O has a non-metallic temperature dependence in all three regimes and decreases strongly by several orders of magnitude between room temperature and 50 K. However, S is distinctly different in the three regimes, and provides more direct insight into the electronic structure. In regime I, S is positive; when moving towards the peak, it reduces in magnitude and approaches a linear temperature dependence characteristic of metallic transport (Figure 1d). In regime II, S switches from p-type to n-type, increases in magnitude with increasing |VIG|, and becomes near temperature independent between 100 - 250K. When sweeping VIG across the valley, S switches sign again, moving abruptly from a large negative to a large positive value. In regime III, the temperature dependence of S is again approximately linear with T, but S is approximately a factor of 3 larger at a comparable p-type conductivity than in regime I, as shown in the S-o plot at 200K in Figure 1e. This characteristic behaviour was observed reproducibly in multiple IDTBT samples and in several other copolymers with related chemical structures (SI section S3.2). In PBTTT, only regime I is accessible and S remains positive at all gate voltages, while in DPP-BTz, a fully insulating state is reached after regime II, and a sign reversal of S from positive to negative is observed at the conductivity peak (Figures 5 and 6). The observed behaviour is consistent with density function theory electronic structure calculations, which show that the HOMO-1 level is expected to be more accessible in IDTBT than in DPPBTz and PBTTT (Figure 1f).

[0104] These observations are consistent with a simple band filling interpretation: In regimes I and III, S is determined by the population of holes at the top of the HOMO and HOMO-1 bands, respectively, while in regime II, it is the electrons remaining at the bottom of the HOMO band that determine the sign of S. At the transition between regime II and III, the Seebeck coefficient is governed by the competition between two bands — electrons in the HOMO and holes in the HOMO-1 — causing S to be near zero in the valley, but take on large values on either side. The origin of the peculiar temperature independence of S in regime II could be a manifestation of electron-electron interactions at high charge densities causing electrons at the bottom of the HOMO to be more susceptible to localisation than holes at the top of the HOMO. Complementary measurements are also fully consistent with the band filling interpretation: From ex-situ X-ray and ultraviolet photoemission (XPS / UPS) measurements of films electrochemically doped to different states within each transfer curve (Figures 7 and 8) and gate current charging analysis, a 1 :1 ratio of TFSI ions to IDTBT repeat units near the conductivity peak is estimated, consistent with half band filling. The doping level in the valley is approximately 1.5 ions per monomer, indicative of some overlap between HOMO and HOMO-1 bands. Regime III should correspond to carrier densities of up to 3 ions per monomer, but direct measurements of the highest doped samples were found challenging due to sample dedoping during measurements under ultrahigh vacuum. Density functional theory calculations (Figure 1f) show a relatively small HOMO bandwidth reflecting the donor-acceptor character of IDTBT with a close-lying HOMO-1 , consistent with the electrochemical accessibility of this band. Cyclic voltammetry (CV) (Figure 9) reveals two overlapping redox features at 0.6-0.7V V, followed by a third feature at higher potential, corresponding to the three transport regimes. The conductivity peak occurs between the first two redox waves, which each appear to be single-electron transfer events, consistent with our picture of maximum conduction near half band filling. In-operando optical and infrared spectroscopy and electron spin resonance (ESR) as a function VIG are also fully consistent with band filling. In regime I in particular, the bleaching of the TT-TT* band at 2 eV due to removal of charge from the HOMO reaches completion exactly at the conductivity peak, consistent with removal of one electron per repeat unit. The inventors considered alternative mechanisms for the non-monotonic OECT and Seebeck behaviour, including changes in structural ordering upon doping and a previously reported mechanism involving the coexistence of crystalline and amorphous domains; however, these did not provide a consistent explanation for the observations described herein.

[0105] While electrochemical doping in OECTs evidently provides an effective way to shift Ep deep into the HOMO / HOMO-1 band, changes in the shape of the DOS due to Coulomb interactions do not directly manifest themselves in OECT electrical characteristics. To probe DOS changes directly, the inventors developed a new double-gating technique, which adds a (bottom) FET back-gate to the (top) OECT ion gate (Figure 2a). First, IG is applied at a temperature above the ionic liquid freezing point (e.g., 270K) to dope the bulk of the polymer to a defined level, and then the device is cooled down with IG held on. Once the ions are frozen (e.g., 160K), VIG can be removed without changing the concentration of counter-ions in the film. The inventors then sweep the back-gate voltage (VBG), and record corresponding back-gate transfer curves. This experiment essentially probes the transport properties of additional electronic carriers injected into the DOS at a given bulk doping state set by VIG, i.e., without being charge compensated by extra counter-ions. By choosing either a positive or a negative VBG, the inventors were able to inject either additional electrons or holes. An important practical requirement for such dual-gating experiments is minimizing the thickness of the polymer layer to maximize the back-gate field-effect, as only the ~ 1 nm thick region near the polymer / SiO2 interface is affected by BG. The inventors used as thin as practically possible (10 nm) polymer films; however, the back-gate field-effect response is still not expected to exceed a few percent of current modulation. The number of interfacial charge carriers that can be modulated between VBG = ±60V is equivalent to a bulk charge concentration < 1019cm-2, this needs to be compared to a bulk charge concentration induced by VIG of » 1021cm-2near the peak.

[0106] Figure 2b shows six selected back-gate transfer curves (insets) measured at different doping states along the IDTBT OECT transfer characteristics (main panel) at 160 K. The full data set at 14 doping states is shown in Figure 10. Since film conductivity changes by orders of magnitude with increasing doping level, the drain current is normalized as (ID(VBG) - lo) / lo, where Io = ID(VBG = 0) is the current of the OECT at the corresponding doping state in the main panel. In each inset the maximum gate voltage modulation is indicated as a percentage value. The experimental data reveal several very surprising observations: Firstly, the observed back-gate modulation is easy to measure; it is much larger than expected, and varies from 12% near the peak to 297% near the valley. Secondly, the inventors had expected the back-gate response to follow the OECT iongate transfer characteristics — to be unipolar p-type in regime I and III, and n-type in regime II. At the peak, the inventors had expected a current reduction for both positive and negative VBG. Instead, the inventors observed a significant current increase in both directions. At most other doping states, the bottom-gate transfer characteristics also exhibit a similar “graphene-like” ambipolar behaviour, though the current rise for negative VBG is stronger than for positive VBG in regime I and III and weaker in regime II. Only at the beginning of regime I and III (doping state #1 and #11 in Figure 10) is the expected, unipolar response, approached, although even there a significant non-linearity is retained. Intriguingly, at the highest doping state (#14) the characteristics resembles that at the first peak (#4) suggesting that the system might be approaching half-filling of the HOMO-1 band. Unfortunately, further increase of doping level was prevented by the electrochemical stability of the devices. Similar back gate response was also observed in DPPBTz and PBTTT (Figure 11).

[0107] Figure 2c shows back-gate transfer curves for an ex-situ doped IDTBT device between 300 K and 220 K exhibiting non-linear behaviour at low T. Transfer curves are vertically shifted for clear presentation. The magnitude of the arrow represents 0.4 pS. Inset is a schematic of the decomposition of the 220 K transfer curve into a linear (equilibrium) component and a symmetric (non-equilibrium) component.

[0108] To analyse the non-linear transfer curves, the inventors decomposed them into a linear component that is anti-symmetric with respect to ±VBG), and a non-linear, symmetric component centered at VBG = 0 (Figure 2c). These can be consistently interpreted as probing different features of the DOS in the vicinity of the Fermi level set by VIG. The linear component, which follows the behaviour expected from the OECT characteristics as a function of VIG, is believed to be determined by the slope of the DOS at EF. In regime I and III, the DOS rises as EF is pushed towards the centre of the HOMO and HOMO-1 band, respectively, resulting in a p-type linear component. Likewise in regime II, the DOS falls as EF is pushed beyond the center of the HOMO band, resulting in an n-type linear component. When EF is in the centre of the band, the slope of the DOS at EF vanishes, and the linear component disappears. This is clearly correlated with the sign of the Seebeck coeffcient (Figure 1d) — a p-type (n-type) linear component in regimes I and III (II) is associated with positive (negative) S and the vanishing linear component at the peak and the valley manifests itself in near zero S.

[0109] Without wishing to be bound by theory, the inventors suggest that the symmetric component can be understood as probing a local minimum or gap in the DOS that is superimposed on the overall shape of the DOS as just discussed, and remains tied to EF as IG is varied, but does not move with EF as VBG is swept. We can understand not only the ambipolar nature, but also the unexpectedly strong magnitude of the back-gate response if we assume that the additional charge carriers induced by VBG are able to access states on the respective side of the DOS minimum that have either higher DOS and / or higher degree of charge delocalisation than states at the DOS minimum.

[0110] The observed ambipolar signatures are reminiscent of field-effect gating experiments on other disordered metals, such as conducting metal oxides or granular metals, at low temperatures (< 4K). One explanation put forward for these is a slow, non-ergodic carrier relaxation in the presence of a soft Coulomb gap caused by electron-electron interactions. However, there are important differences — here, the inventors observed these signatures at much higher temperatures and stably, even when measuring very slowly; the inventors did not observe, for example, so-called double-dip signatures under similar measurement conditions. This makes it unlikely that glassy electron relaxation is responsible in the present system, as this would be expected to be fast at such high temperatures, i.e. , a soft Coulomb gap would remain tied to the shifting Fermi level also when VBG is changed. To better understand this, the inventors investigated the temperature dependence of the back-gate transfer curves. The ambipolar characteristics were observed over a wide temperature range below 190 K. The back-gate field-effect approaching room temperature can’t be studied in OECTs (as ions would move in and out of the film in response to the back-gate), so the inventors used IDTBT films that were doped ex-situ into defined states by ion exchange doping, or by washing off the ionic liquid after electrochemical doping. The change of conductance AG = G-Go (Go = G(VBG = 0)) is plotted versus VBG at different temperatures in Figure 2d. Strikingly, at 300 K, the transfer curve of this mildly doped IDTBT film in Regime I is unipolar p-type and can be fitted perfectly with a straight line. The non-linearity reproducibly appears only below a transition temperature of » 240 K.

[0111] To investigate whether the appearance of the non-linear component is linked to the relaxation dynamics of the counter-ions rather than of the electrons the inventors used nuclear magnetic resonance (NMR) to probe the nuclear spin relaxation time of the fluorine atoms in the FSI dopant ions. These provide evidence for a significant slowing of the relaxation of the FSI ion dynamics in the same temperature range in which the non-linear transfer characteristics appear (Figure 2c). This suggests that the origin of the unusual back-gate field-effect response may be slow, non-ergodic ion relaxation: At room temperature, ionic motion is fast enough that ions can always find their lowest energy equilibrium configuration adjusting to the change in carrier concentration during the sweep of BG; at 260 K, the slowing relaxation time of the ions becomes comparable to the VBG sweep rate, which manifests itself as hysteresis in the transfer curves (Figure 2c). At even lower temperatures the ions can no longer respond to VBG-induced changes in carrier concentration and remain frozen in configurations they adopted when VBG = 0 V was applied before device cooling. Indeed, when a non-zero VBG is applied during cooling, the system “memorizes” its high temperature equilibrium configuration at this specific voltage, and the minimum of the ambipolar transfer characteristics at low temperature shifts to the value of VBG applied during cooling.

[0112] Figure 3 illustrates spectroscopic evidence for enhanced delocalization of nonequilibrium states. Figures 3a and 3b show CMS spectra of an IDTBT:TFSI film doped to regime I collected at 270K (Figure 3a) and 190 K (Figure 3b). Insets bottom left show transfer curves at each temperature. Symbols are experimental data, solid lines are fits from the decomposed even and odd component spectra as described herein. Figure 3c shows odd (top) and even (bottom) component spectra used to fit the data. The odd component (270K data shown) matches closely with OECT moving difference spectra. The even component (190K - 270K data shown) is nearly independent of doping level and reflects an increase in carrier delocalization upon FET gating below the ionic glass transition temperature.

[0113] To understand why such non-equilibrium states induced by sweeping VBG at low temperatures lead to increased conductance, the inventors employed charge modulation spectroscopy (CMS), which measures the device differential optical absorption as BG is modulated and provides direct spectroscopic characterisation of the additional carriers. Redshifts in the position of the polaron-induced absorption band in the infrared (0.1 -0.5 eV) indicate increasing polaron delocalization. At 270K the CMS signal is dominated by an anti-symmetric response with respect to VBG (Figure 3a): for this device in regime I a negative VBG leads to an increased polaron induced absorption, while a positive VBG leads to a corresponding bleaching signal without spectral changes (apart from an absorption tail below 0.1 eV from the CMS response of the Si substrate). At 190 K (Figure 3b) the CMS spectrum becomes more complex, and the induced absorption and bleaching spectra at negative and positive VBG, respectively, do no longer have the same shape.

[0114] To interpret these spectra, the inventors mirror the analysis of the electrical characteristics and decompose the measured absorption A(w) into component spectra that are symmetric and anti-symmetric with respect to VBG, i.e. A(w) = ca(w)VBG + CS(W)|VBG| . A least-squares fit at each frequency provides a good match with experiment. At 270K, the anti-symmetric component dominates (Figure 3c) and closely matches with the corresponding moving difference spectrum obtained from the OECT data (Figure 3c). This indicates that above the transition temperature the polarons induced by ion gating or field-effect are indistinguishable spectroscopically and have the same degree of polaron delocalisation. This is consistent with the previous work by the inventors, in which it was found that at such high doping levels the ion density acts as a uniform background charge density.

[0115] Below the transition temperature, while the anti-symmetric component remains qualitatively unchanged, a significant symmetric component is now also present. The response of the Si appears in the symmetric component as well but can be largely cancelled by subtracting off the 270K symmetric component. The resulting 190K- 270K symmetric component (Figure 3c) is remarkably consistent in both spectral shape and intensity across a range of doping levels and shows a broad negative, bleaching feature at high energy and positive, induced absorption at low energy, that integrate to near zero. It therefore does not reflect a change in carrier density, but instead indicates that the polaron bands of both additional electron and hole carriers are shifted to lower energy, i.e. the carriers become more delocalized. This is consistent with the ease of measuring the back-gate response and with an analysis of the transfer characteristics in terms of field-effect mobilities, which reveals that the mobilities extracted from the ambipolar, symmetric component tend to be up to one order of magnitude larger than those extracted from the linear component, or those estimated from the film conductivity. The CMS results provide clear spectroscopic evidence that below » 240K gating allows Ep to climb up the edges of the gap allowing the carriers to access more delocalized states — the soft gap at Ep becomes frozen.

[0116] Figure 4 demonstrates the enhancement of the Seebeck coefficient in the vicinity of the Coulomb gap. Figure 4a Illustrates a back-gate transfer curve of IDTBT at the measurement temperature of 190 K. Changes in the magnitude of the Seebeck coefficient across the gap are indicated on the transfer curve, as inferred from Figure 4b, which shows thermovoltage measurements at various back-gate voltages, VBG.

[0117] Figure 5 illustrates regimes I and II band filling in DPP-BTz. Figure 5a shows the molecular structure of DPP-BTz. Figure 5b shows an OECT transfer curve of a typical BMP-TFSI gated device. A decrease of conductivity at high gate voltages is seen, suggesting a crossover to Regime II. Figure 5c shows the temperature dependence of the Seebeck coefficient showing a transition to n-type transport beyond the conductivity peak, confirming the accessibility of Regime II in DPPBTz. Figure 5d illustrates a DFT band structure of DPP-BTz showing a relatively narrow HOMO reminiscent of IDTBT.

[0118] Figure 6 illustrates regime I band filling in PBTTT. Figure 6a shows the molecular structure of PBTTT. Figure 6b shows an OECT transfer curve of a typical BMP-TFSI gated device. A monotonic increase of conductivity with increasingly negative gate voltages is seen, suggesting that the device remains in Regime I throughout. Figure 6c shows the temperature dependence of the Seebeck coefficient, where the absence of n- type transition provides confirmation of PBTTT remaining in Regime I even at the highest doping levels attainable in studies leading to the present disclosure. Figure 6d illustrates a DFT band structure of PBTTT, showing a significantly broader HOMO compared to DPPBTz and IDTBT.

[0119] Figure 7 shows the doping induced changes in the frontier molecular orbitals of conjugated polymers as inferred from ultraviolet photoemission spectroscopy (UPS). UPS spectra of PBTTT (Figure 7a) DPPBTz (Figure 7b), and IDTBT (Figure 7c) at various doping levels are shown. The top right insets show a detailed view of the Fermi level, highlighting the presence of Fermi edge indicative of metallic states in the highly doped samples. Notice the complete emptying of the HOMO in the insulating state of DPPBTz’s Regime II.

[0120] Figure 8 shows the sulfur 2p peak in the X-ray photoemission spectra of PBTTT (Figure 8a), DPPBTz (Figure 8b), and IDTBT (Figure 8c) at various doping levels. The dopant ion peak appears at higher binding energies, denoted as STFSI. The number of ions per monomer n is indicated on each doped spectrum.

[0121] Figure 9 illustrates cyclic voltammogram (bottom panels, sweep rate 10 mV s-1) and the corresponding conductivity at different doping stages (top panels) for PBTTT (Figure 9a), DPP-BTz (Figure 9b), and IDTBT (Figure 9c). In DPPBTz, we see two redox features - one at the onset of Regime I, and another following the complete doping of Regime II. In IDTBT, three redox features are observed - two close together coinciding with the conductivity peak, and a third one at the onset of Regime III. In PBTTT, in contrast only one redox feature is observed at the onset of its Regime I. The inventors’ findings establish new and general strategies for enhancing the transport properties of conjugated polymers: the inventors have demonstrated that in systems that can be operated stably in regime III significantly higher conductivities are achievable. This may be because the polarons injected into the HOMO-1 levels experience a higher dielectric constant and more efficient screening of Coulomb interactions, as the film already contains a high number of charges / counterions. The inventors have also shown that through freezing of the soft Coulomb gap highly delocalized charge carriers can be induced which have at least an order of magnitude higher mobility than carriers induced under equilibrium conditions. Furthermore, the ability to move Ep into the rapidly changing part of the DOS in the soft gap should also provide a potential route for enhancing Seebeck coefficients for thermoelectric applications: Within the Boltzmann transport framework the magnitude of S is proportional to the slope of the energydependent conductivity function, which is smallest at the bottom of the gap, but rises as more delocalized states with larger DOS become accessible at the edges of the gap. For a proof of principle the inventors attempted to detect changes in the thermopower of an IDTBT double-gated OECT upon field-effect gating. These measurements were performed on the most conductive state of Regime III (similar to state #14). Indeed, when the soft gap is frozen at 190 K, a 10% larger Seebeck coefficient of 66 pV K'1is recorded when a -40 V back-gate potential is applied compared to the un-gated value of 60.5 pV K'1. This is remarkable, given that the applied back-gate only affects carriers near the dielectric interface. From a parallel conduction model for the Seebeck coefficient the inventors estimate that up to an order of magnitude larger thermopower could be achieved, if also the Fermi level in the bulk could be moved away from the DOS minimum. For +40 V back-gate S is reduced compared to the ungated state, which can be understood as a balance of the p-type thermopower response of the bulk and the expected n-type response from field-effect modulated carriers.

[0122] Materials systems that allow freezing the soft gap above room temperature should be possible using polymers and ionic liquids / salts with glass transitions and melting temperatures above room temperature. Furthermore, it should be possible to induce such highly delocalized carriers not just at interfaces, but also in the bulk, by first doping the polymer at elevated temperature with a high glass transition counterion, and then shifting Ep out of equilibrium by lightly doping in a second doping step under conditions where the counterions introduced during the first doping step are unable to move, for example by using an ion with a second, lower glass transition temperature. Other potential methods include generation of additional carriers by optical injection or modulation doping in polymer blends.

[0123] In general, improvements in S and conductivity could be achieved by addition or removal of charge carriers (removing carriers will often lead to a smaller improvement but still will increase both S and conductivity).

[0124] In some examples, the ions that are freezing out do not necessarily need to be dopant ions. The ions could for instance be extra ions added to the material for the sole purpose of increasing the background charge density to generate (or enhance) a coulomb gap.

[0125] The electrically conductive material described herein may comprise a semiconductor, but it will be understood that (in particular for thermoelectric applications) the electrically conductive material may be doped into a metallic (or nearly metallic) state. The electrically conductive material could also comprise any disordered electrical conductor that supports a Coulomb gap.

[0126] Following from the points above, it may be possible to achieve the effects described herein with no dopants at all. Starting with a metallic conductor and introduce an ionic material (e.g. a zwitterion or an ionic liquid, i.e. overall charge neutral) then freeze this ionic component out, the interaction of the charge carriers with these ions may lead to a frozen coulomb gap.

[0127] Without wishing to be bound by theory, the transition temperature of the ionic material (and / or dopant) may be connected to a transition temperature of the electrically conductive material (e.g. polymer). Suitable polymers according to the present disclosure may include IDTBT, PCDTBT, or BBL.

[0128] Suitable ionic materials may include TFSI. It may be the case that increasing the sidechain lengths in cations and anions increases the glass transition temperature, which may prove useful in developing a device that can operate at high temperatures (e.g. room temperature or above). For example, Electrochi mica Acta 56 (2011) 1300-1307 (the contents of which are hereby incorporated by reference) describes a glass transition temperature for TFSI of around 180 K, and also reports on the same cation with IM 14- (trifluoromethanesulfonyl)(nonafluorobutylsulfonyl)imide), similar to TFSI but with a long fluorinated sidechain, exhibiting a glass transition about 40 K higher. A higher glass transition temperature of the ionic material may correspond to a higher transition temperature according to the present disclosure.

[0129] Increasing the anion size may also lead to an increased transition temperature. For example, bis(oxalate)borate (BOB) may be a suitable ionic material. Generally the entire class of weakly coordinating anions (Angewandte Chemie International Edition 57 (43), 13982-14024, the contents of which are hereby incorporated by reference) which are large, non-reactive anions, might be of interest.

[0130] It may be possible to introduce large anions directly with molecular dopants, which themselves become the dopant counterion. A suitable material for this purpose may be phosphomolybdic acid (PMA), which is a very large ion that displays very low diffusion rates in polymers (Nature Mater 16, 474-480 (2017), the contents of which are hereby incorporated by reference).

[0131] Figure 14 illustrates a field-effect transfer curve obtained for a field-effect transistor comprising a poly(3,4-ethylenedioxythiophene):tosylate (PEDOT:Tos) at room temperature. The electrical conductivity of the field-effect transistor is -100 S / cm. The solid lines (left axis) in Figure 14 show the variation in drain current (units of 10'5A), while the dotted lines (right axis) show the gate leakage (units of 10'10A). The field-effect transfer curve shows a significant non-linearity, indicating that non-equilibrium states may be generated in PEDOT:Tos at room temperature, similarly to the low temperature observations described above.

[0132] Similar results to those illustrated in Figure 14 were observed by the inventors in other PEDOT materials, such as poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS).

[0133] Without wishing to be bound by theory, the observation of the non-linear field-effect transfer curve at room temperature in these materials may be due to the absence of polymer side-chains. For example, in polymer materials comprising side-chains (e.g. alkyl side-chains), the side-chains may be sufficiently mobile at room temperature, and their local configurations may change sufficiently at room temperature, that ions can move through the network of side-chains. At low temperatures (i.e. below the transition temperature of the side-chain), the motion of the side-chains may be frozen out so that ions are unable to move and respond to changes in the density of electronic carriers. Conjugated polymer backbone glass transitions typically occur at much higher temperatures, well above room temperature. A side-chain-free polymer material as the electrically conductive material may, therefore, be sufficiently rigid at room temperature so as to prevent ion movement at room temperature, resulting in a device according to the present disclosure that is operable at room temperature (e.g. because the spatial locations and / or orientations of ions are unchangeable in response to changes in charge density at room temperature).

[0134] In addition to the PEDOT materials described above, other examples of side-chain-free polymer materials that may be suitable as the electrically conductive material include poly(benzimidazobenzophenanthroline) (BBL) and poly(benzodifurandione) (PBFDO).

[0135] Figure 12 schematically illustrates an example method 1200 of producing an electronic device according to the present disclosure.

[0136] In an optional step S1202, the method 1200 may comprise introducing an ionic material into an electrically conductive material. As described herein, the electrically conductive material may comprise a semiconductor, a polymer, an organic semiconductor, a molecular semiconductor, a polymer, a two-dimensional material, and / or a disordered metal. Preferably, the electrically conductive material is a material capable of exhibiting a Coulomb gap. As further described herein, the ionic material may comprise an ionic liquid, an ionic salt, and / or a redox active molecular dopant, or a natural molecule that undergoes a chemical reaction to generate ions.

[0137] The ionic material may be introduced into the electrically conductive material by any suitable technique, for example by chemical doping, by ion exchange doping, and / or electrochemically introducing the ionic material into the electrically conductive material.

[0138] In a step S1204, the method 1200 comprises, while the electrically conductive material is in a state such that ionic motion of the ionic material is frozen, modifying a charge carrier density of the electrically conductive material. As described herein, the ionic motion of the ionic material may be considered to be frozen if spatial locations and / or orientations of ions of the ionic material are unchangeable in response to modifying the charge carrier density.

[0139] Modifying the charge carrier density may comprise introducing a dopant, electrically injecting carriers (e.g. by an electrical field effect), and / or optically injecting carriers (e.g. by directing a light source, such as a laser, onto the electrically conductive material).

[0140] It may be possible to determine that the ionic motion of a given ionic material in a given electrically conductive material is frozen (and / or that the spatial locations and / or orientations of ions in the ionic material are unchangeable in response to modifying the charge carrier density) using the configuration described herein and illustrated in Figure 2a. For example, when the electrically conductive material is configured as a field effect transistor and / or an OECT, if a field effect transfer curve obtained for the transistor exhibits a non-linearity, the electrically conductive material may be considered to be in a state in which ionic motion of the ionic material is frozen (and / or that the spatial locations and / or orientations of ions in the ionic material are unchangeable in response to modifying the charge carrier density).

[0141] It will be understood that, in some examples, the step S1202 may be omitted. For example, the electrically conductive material may already comprise the ionic material. E.g., the electrically conductive material may behave as a metal. In some examples, the electrically conductive material may be substantially not electrically conductive prior to the introduction of the ionic material.

[0142] Figure 13 schematically illustrates an example of a thermoelectric device 1300 according to the present disclosure. The example thermoelectric device 1300 comprises an electrically conductive material 1304. The electrically conductive material 1304 may be prepared according to any of the methods described herein. For example, the electrically conductive material 1304 may comprise a semiconductor, a polymer, an organic semiconductor, a molecular semiconductor, a polymer, a two-dimensional material, and / or a disordered metal.

[0143] The electrically conductive material 1304 may be doped with: an ionic material, the ionic material having a first transition temperature. The first transition temperature may be a temperature below which ionic motion of the ionic material is frozen. The electrically conductive material 1304 may also be doped with a dopant, the dopant having a second transition temperature. The second transition temperature may be a temperature below which ionic motion of the dopant is frozen. The second transition temperature may be lower than the first transition temperature. The first transition temperature may be a temperature below which spatial locations and / or orientations of ions of the ionic material are unchangeable in response to a modification of the charge carrier density (e.g. induced by the dopant). For example, the ionic motion of the ionic material may be considered to be frozen when spatial locations and / or orientations of ions in the ionic material are unchangeable in response to the modifying of the charge carrier density. The second transition temperature may be a temperature below which spatial locations and / or orientations of ions of the dopant are unchangeable in response to a modification of the charge carrier density.

[0144] Alternatively, or in addition, the electrically conductive material 1304 may comprise: a highest occupied molecular orbital, HOMO, and a next-to-highest occupied molecular orbital, HOMO-1 , wherein an energy difference between the HOMO and the HOMO-1 is smaller than a width of each of the HOMO and the HOMO-1 ; and / or a lowest unoccupied molecular orbital, LIIMO, and a next-to- lowest unoccupied molecular orbital, LLIMO+1 , wherein an energy difference between the LIIMO and the LLIMO+1 is smaller than a width of each of the LIIMO and the LLIMO+1.

[0145] The example thermoelectric device 1300 further comprises a first (e.g. hot) electrode 1302A, and a second (e.g. cold) electrode 1302B. When a thermal gradient is applied across the electrically conductive material 1304, the thermoelectric device can be said to have hot side A and a cold side B, such that the first electrode 1302A is electrically connected to the hot side A and the second electrode 1302B is electrically connected to the cold side B. Applying the thermal gradient across the electrically conductive material 1304 causes (delocalized) charge carriers to diffuse from the hot side A to the cold side B, thereby generating a voltage V by the thermoelectric effect (e.g. by the Seebeck effect).

[0146] In some examples, a plurality of thermoelectric devices 1300 may be connected in series. Although the disclosure has been described in terms of preferred embodiments as set forth above, it should be understood that these embodiments are illustrative only and that the claims are not limited to those embodiments. Those skilled in the art will be able to make modifications and alternatives in view of the disclosure, which are contemplated as falling within the scope of the appended claims. Each feature disclosed or illustrated in the present specification may be incorporated in any embodiments, whether alone or in any appropriate combination with any other feature disclosed or illustrated herein.

[0147] BIBLIOGRAPHY

[0148] The following publications are hereby incorporated by reference in their entirety:

[0149] Chiang, C. K. et al. Electrical conductivity in doped polyacetylene. Physical Review Letters 39, 1098-1101 (1977).

[0150] Fratini, S., Nikolka, M., Salleo, A., Schweicher, G. & Sirringhaus, H. Charge transport in high-mobility conjugated polymers and molecular semiconductors. Nature materials 19, 491-502 (2020).

[0151] Torricelli, F. et al. Electrolyte-gated transistors for enhanced performance bioelectronics. Nature reviews. Methods primers 1 (2021).

[0152] Russ, B., Glaudell, A., Urban, J. J., Chabinyc, M. L. & Segalman, R. A. Organic thermoelectric materials for energy harvesting and temperature control. Nature Reviews Materials 2016 1 :10 1 , 1-14 (2016).

[0153] Giamarchi, T. Theoretical framework for quasi-one dimensional systems. Chemical Reviews 104, 5037-5055 (2004).

[0154] Jerome, D. & Schulz, H. J. Organic conductors and superconductors. https: / / doi.org / 10.1080 / 00018730110116362 51 , 293-479 (2010).

[0155] Koopmans, M. & Koster, L. J. A. Carrier-carrier Coulomb interactions reduce power factor in organic thermoelectrics. Applied Physics Letters 119, 143301 (2021).

[0156] Xu, K. et al. On the Origin of Seebeck Coeffcient Inversion in Highly Doped Conducting Polymers. Advanced Functional Materials 32, 2112276 (2022).

[0157] Kaiser, A. B. Systematic Conductivity Behavior in Conducting Polymers: Effects of Heterogeneous Disorder**. Advanced Materials 13, 927-941 (2001). Lee, J., Panzer, M. J., He, Y., Lodge, T. P. & Frisbie, C. D. Ion gel gated polymer thin- film transistors. Journal of the American Chemical Society 129, 4532-4533 (2007).

[0158] Tanaka, H. et al. Thermoelectric properties of a semicrystalline polymer doped beyond the insulator-to-metal transition by electrolyte gating. Science Advances 6, 8065- 8079 (2020).

[0159] Ito, M. et al. Band mobility exceeding 10 cm2 V-1 s-1 assessed by field-effect and chemical double doping in semicrystalline polymeric semiconductors. Applied Physics Letters 119, 13302 (2021).

[0160] Ito, H., Mada, H., Watanabe, K., Tanaka, H. & Takenobu, T. Charge transport and thermoelectric conversion in solution-processed semicrystalline polymer films under electrochemical doping. Communications Physics 2021 4:1 4, 1-9 (2021).

[0161] Cho, K. G., Adrahtas, D. Z., Lee, K. H. & Frisbie, C. D. Sub-Band Filling and Hole Transport in Polythiophene-Based Electrolyte-Gated Transistors: Effect of Side- Chain Length and Density. Advanced Functional Materials (2023).

[0162] Panzer, M. J. & Frisbie, C. D. Polymer electrolyte-gated organic field-effect transistors: Low-voltage, high-current switches for organic electronics and testbeds for probing electrical transport at high charge carrier density. Journal of the American Chemical Society 129, 6599-6607 (2007).

[0163] Venkateshvaran, D. et al. Approaching disorder-free transport in high-mobility conjugated polymers. Nature 2014 515:7527 515, 384-388 (2014).

[0164] Cendra, C. et al. Unraveling the Unconventional Order of a High-Mobility Indacenodithiophene-Benzothiadiazole Copolymer. ACS Macro Letters 10, 1306-141314 (2021).

[0165] Schott, S. et al. Charge-Transport Anisotropy in a Uniaxially Aligned Diketopyrrolopyrrole-Based Copolymer. Advanced Materials 27, 7356-7364 (2015).

[0166] Liang, Z. et al. n-type charge transport in heavily p-doped polymers. Nature Materials 1- 7 (2021).

[0167] Vaknin, A., Ovadyahu, Z. & Pollak, M. Evidence for interactions in nonergodic electronic transport. Physical Review Letters 81 , 669-672 (1998).

[0168] Amir, A., Oreg, Y. & Imry, Y. Electron Glass Dynamics, https: / / doi.org / 10.1146 / annurev- conmatphys-062910-140455 2, 235-262 (2011). Pollak, M. THE EUROPEAN PHYSICAL JOURNAL SPECIAL TOPICS Review Electrons in Anderson-Mott insulators. Eur. Phys. J. Special Topics 227, 2221-2240 (2019).

[0169] Yu, C. C. Time-dependent development of the coulomb gap. Physical Review Letters 82, 4074-4077 (1999).

[0170] Tsigankov, N., Pazy, E., Laikhtman, D. & Efros, L. Long-time relaxation of interacting electrons in the regime of hopping conduction. Physical Review B - Condensed Matter and Materials Physics 68 (2003). 0307552.

[0171] Ovadyahu, Z. Relaxation dynamics in quantum electron glasses. Physical Review Letters 99 (2007).

[0172] Qarai, M. B., Ghosh, R. & Spano, F. C. Understanding Bipolarons in Conjugated Polymers Using a Multiparticle Holstein Approach. The Journal of Physical Chemistry C acs.jpcc.1c06767 (2021).

[0173] Jacobs, I. E. et al. Structural and Dynamic Disorder, Not Ionic Trapping, Controls Charge Transport in Highly Doped Conducting Polymers. Journal of the American

[0174] Chemical Society 144, 3005-3019 (2022).

[0175] Comin, M., Fratini, S., Blase, X. & D’Avino, G. Doping-Induced Dielectric Catastrophe Prompts Free-Carrier Release in Organic Semiconductors. Advanced Materials 34 (2022).

Claims

CLAIMS:1 . A method of producing an electronic device, the method comprising: modifying a charge carrier density of an electrically conductive material, the electrically conductive material comprising an ionic material, wherein modifying the charge carrier density of the electrically conductive material comprises modifying the charge carrier density while the electrically conductive material is in a state such that ionic motion of the ionic material is frozen.

2. A method according to claim 1 , wherein modifying the charge carrier density of the electrically conductive material comprises modifying the charge carrier density under conditions wherein spatial locations and / or orientations of ions of the ionic material are unchangeable in response to said modifying of the charge carrier density.

3. A method according to claim 1 or 2, comprising introducing the ionic material into the electrically conductive material.

4. A method according to any one of the preceding claims, wherein the ionic material has a first transition temperature, the first transition temperature being a temperature below which the ionic motion of the ionic material is frozen; and wherein the method comprises cooling the electrically conductive material to a temperature below the first transition temperature.

5. A method according to any one of the preceding claims, wherein modifying the charge carrier density of the electrically conductive material comprises doping the electrically conductive material with a dopant, the dopant having a second transition temperature, the second transition temperature being a temperature below which ionic motion of the dopant is frozen, wherein the second transition temperature is lower than the first transition temperature.

6. A method according to any one of the preceding claims, wherein modifying the charge carrier density of the electrically conductive material comprises electrically injecting charge carriers.

7. A method according to any one of the preceding claims, wherein the electrically conductive material comprises a semiconductor.

8. A method according to any one of the preceding claims, wherein the electrically conductive material comprises a polymer.

9. A method according to claim 8, comprising introducing the ionic material such that a concentration of the ionic material in the electrically conductive material is at least one ion per repeat unit of the polymer.

10. A method according to any one of the preceding claims, comprising introducing the ionic material such that a concentration of the ionic material in the electrically conductive material is such that, while the electrically conductive material is in a state such that ionic motion of the first dopant is frozen, the device exhibits a nonlinear transfer curve.

11. A method according to any one of the preceding claims, wherein the electrically conductive material comprises: a highest occupied molecular orbital, HOMO, and a next-to-highest occupied molecular orbital, HOMO-1 , wherein an energy difference between the HOMO and the HOMO-1 is smaller than a width of each of the HOMO and the HOMO-1 ; and / or a lowest unoccupied molecular orbital, LIIMO, and a next-to-lowest unoccupied molecular orbital, LLIMO+1 , wherein an energy difference between the LIIMO and the LLIMO+1 is smaller than a width of each of the LIIMO and the LUMO+1.

12. A method according to any one of the preceding claims, wherein the electrically conductive material comprises indacenodithiophene-co-benzothiadiazole, IDTBT.

13. A method according to any one of the preceding claims, wherein the electrically conductive material comprises a side-chain-free polymer.

14. A method according to any one of the preceding claims, wherein the electrically conductive material comprises poly(3,4-ethylenedioxythiophene), PEDOT.

15. A method according to any one of the preceding claims, wherein the electrically conductive material comprises poly(3,4-ethylenedioxythiophene):tosylate, PEDOT:Tos.

16. A method according to any one of the preceding claims, wherein the electrically conductive material comprises poly(3,4-ethylenedioxythiophene) polystyrene sulfonate, PEDOT: PSS.

17. A method according to any one of the preceding claims, wherein the electrically conductive material comprises poly(benzimidazobenzophenanthroline), BBL.

18. A method according to any one of the preceding claims, wherein the electrically conductive material comprises poly(benzodifurandione), PBFDO.

19. A method according to any one of the preceding claims, further comprising electrically connecting the electrically conductive material to a source terminal and a drain terminal.

20. An electronic device produced according to the method of any one of the preceding claims, wherein the electronic device is a transistor.

21. A method of generating electricity, the method comprising: producing an electronic device according to the method of any one of claims 1 to 19, wherein the electronic device as a thermoelectric device; and applying a thermal gradient across the electrically conductive material.

22. An electronic device comprising: an electrically conductive material, the electrically conductive material being doped with: an ionic material, the ionic material having a first transition temperature, the first transition temperature being a temperature below which ionic motion of the ionic material is frozen; anda dopant, the dopant having a second transition temperature, the second transition temperature being a temperature below which ionic motion of the dopant is frozen, wherein the second transition temperature is lower than the first transition temperature.

23. An electronic device according to claim 22, wherein the electrically conductive material comprises: a highest occupied molecular orbital, HOMO, and a next-to-highest occupied molecular orbital, HOMO-1 , wherein an energy difference between the HOMO and the HOMO-1 is smaller than a width of each of the HOMO and the HOMO-1 ; and / or a lowest unoccupied molecular orbital, LIIMO, and a next-to-lowest unoccupied molecular orbital, LLIMO+1 , wherein an energy difference between the LIIMO and the LLIMO+1 is smaller than a width of each of the LIIMO and the LUMO+1.

24. An electronic device according to claim 22 or 23, wherein the electrically conductive material comprises a semiconductor.

25. An electronic device according to any one of claims 22 to 24, wherein the electrically conductive material comprises a polymer.

26. An electronic device according to claim 25, wherein the electrically conductive material is doped with the ionic material to a concentration of more than one ion per repeat unit of the polymer.

27. An electronic device according to any one of claims 22 to 26, wherein the electrically conductive material is doped with the ionic material to a concentration such that, while the semiconductor material is at a temperature below the first transition temperature, the device exhibits a non-linear transfer curve.

28. An electronic device according to any one of claims 22 to 27, further comprising a source terminal, a drain terminal, and a gate terminal electrically connected to the electrically conductive material.

29. An electronic device according to any one of claims 22 to 28, wherein the electronic device is an organic electrochemical transistor.

30. An electronic device according to any one of claims 22 to 29, wherein the electronic device is a thermoelectric device configured to generate electricity when a thermal gradient is applied across the electrically conductive material while the electronic device is at a temperature below the first transition temperature.

31. An electronic device comprising an electrically conductive material, wherein the electrically conductive material comprises: a highest occupied molecular orbital, HOMO, comprising a first energy distribution distributed around a first binding energy, and a next-to-highest occupied molecular orbital, HOMO-1 , comprising a second energy distribution distributed around a second binding energy, the second binding energy being higher than the first binding energy, wherein a Fermi energy of the electrically conductive material is closer to the second binding energy than to the first binding energy; or a lowest unoccupied molecular orbital, LIIMO, comprising a third energy distribution distributed around a third binding energy, and a next- to-lowest unoccupied molecular orbital, LLIMO+1 , comprising a fourth energy distribution distributed around a fourth binding energy, the fourth binding energy being lower than the third binding energy, wherein a Fermi energy of the electrically conductive material is closer to the fourth binding energy than to the third binding energy.

32. An electronic device according to claim 31 , wherein a difference between the first binding energy and the second binding energy is smaller than a width of each of the first energy distribution and the second energy distribution; or wherein a difference between the third binding energy and the fourth binding energy is smaller than a width of each of the third energy distribution and the fourth energy distribution.

33. An electronic device according to claim 31 or 32, wherein the electrically conductive material comprises a polymer.

34. An electronic device according to any one of claims 31 to 33, wherein the electronic device is a thermoelectric device.

35. An electronic device according to any one of claims 22 to 34, wherein the electrically conductive material comprises a side-chain-free polymer.

36. An electronic device according to any one of claims 22 to 35, wherein the electrically conductive material comprises poly(3,4-ethylenedioxythiophene), PEDOT.

37. An electronic device according to any one of claims 22 to 36, wherein the electrically conductive material comprises poly(3,4- ethylenedioxythiophene):tosylate, PEDOT:Tos.

38. An electronic device according to any one of claims 22 to 37, wherein the electrically conductive material comprises poly(3,4-ethylenedioxythiophene) polystyrene sulfonate, PEDOT: PSS.

39. An electronic device according to any one of claims 22 to 38, wherein the electrically conductive material comprises poly(benzimidazobenzophenanthroline), BBL.

40. An electronic device according to any one of claims 22 to 39, wherein the electrically conductive material comprises poly(benzodifurandione), PBFDO.

Images