Recently molecular photovoltaics, such as dye sensitized cells (DSCs) and perovskite solar cells (PSCs) have emerged as credible contenders to conventional p-n junction photovoltaics. Their certified power conversion efficiency currently attains 25.5 %, exceeding that of the market leader polycrystalline silicon. This lecture covers the genesis and recent evolution of DSCs and PSCs, describing their operational principles and current performance. DSCs have meanwhile found commercial applications for ambient light harvesting and glazing producing electric power from sunlight. The scale up and pilot production of PSCs are progressing rapidly but there remain challenges that still need to be met to implement PSCs on a large commercial scale. PSCs can produce high photovoltages rendering them attractive for applications in tandem cells, e.g. with silicon and as power source for the generation of fuels from sunlight. Examples to be presented are the solar generation of hydrogen from water and the conversion of CO2 to chemical feedstocks such as ethylene, mimicking natural photosynthesis.

In this contribution, I will first report on our recent studies on solution-based n-type doping of polymer semiconductors, aimed at understanding current limits in molecular doping schemes. I will then focus on the development of printed organic micro thermoelectric generators (µTEG), as interesting energy harvesters candidates for distributed low power electronics and sensors networks. In particular, I will on describe an architecture where 128 thermocouples are directly embedded in a plastic substrate to improve the thermal coupling and simplify the fabrication process. Such proof-of-principle µTEG allows to envisage future efficient devices delivering µW/cm² close to room temperature and at low temperature differences.

X. Rodríguez-Martínez¹, F. Saiz¹, S. Marina², H. Chen³, O. Zapata-Arteaga¹, B. Dorling¹, J. Martin^2,4,5, I. McCulloch^3,6, R. Rurali¹, J. S. Reparaz¹, M. Campoy-Quiles^1,*

¹ Institut de Ciència de Materials de Barcelona, ICMAB-CSIC, Campus UAB, 08193, Bellaterra, Spain

² POLYMAT and Polymer Science and Technology Department, Faculty of Chemistry, University of the Basque Country UPV/EHU, Manuel de Lardizabal 3, 20018, Donostia- San Sebastián, Spain

³ King Abdullah University of Science and Technology (KAUST) Solar Center (KSC), Thuwal, 23955-6900, Saudi Arabia

⁴ Ikerbasque, Basque Foundation for Science, 48011 Bilbao, Spain

⁵ Centro de Investigacións Tecnolóxicas, Universidade da Coruña, Campus de Esteiro, 15403, Ferrol, Spain

⁶ Department of Chemistry, University of Oxford, Oxford, OX1 3TA, United Kingdom

The thermal conductivity is the main parameter defining how heat propagates in a solid. It is of paramount importance for heat dissipation in electronics and transmission lines, but also in thermoelectrics. Despite its importance, little attention has been paid to its actual value, in part due to the general belief that the thermal conductivity in organic semiconductors “should be” small. Literature shows, however, values spanning more than two orders of magnitude.

In this invited talk, we will show a combined experimental/theoretical study, in which we have measured the thermal conductivity for a large polymer library and discovered the coexistence of two fundamentally distinct regimes. Semicrystalline conjugated polymers behave as conventional theory predicts, with increasing order leading to an increase in thermal conductivity, and this correlates also with an increase in charge carrier mobility. In other words, thermal and electrical transport go hand in hand. On the other hand, materials that do not show long range order in GIWAXS behave very differently, not following the same classic theory. As a consequence, for the latter, charge carrier mobility and thermal conductivity appear to be anticorrelated. We rationalize our results using Spearman statistics as well as theoretical calculations, which allow us to provide simple and exploitable design rules for materials that are able to decouple thermal and electronic transport, namely, texture/orientation and monomer/sidechain weight. Our results open a new avenue for highly efficient organic thermoelectrics.

Millimeter thick bulk materials are needed to facilitate the design of thermoelectric devices. While submicrometer-thin films of organic semiconductors are needed for fundamental spectroscopic studies, they are not ideal for the construction of devices, unless folded or wrinkled. In my talk, I will discuss our recent work on the preparation of conducting bulk materials, from foams to fibers, and explore the interplay of electrical and mechanical properties. Further, I will discuss some of the challenges related to doping of bulk materials, as well as the (in)stability of the doped state. Finally, I will present prototype textile and 3D-printed devices that can be manufactured with bulk materials.

Organic semiconductors are excellent candidates for low temperature thermoelectric generators. However, such thermoelectric applications require that the materials be doped and highly conductive.

Here, we show how doping affects the Seebeck coefficient in organic semiconductors using kinetic Monte-Carlo simulations. Employing a hopping transport approach we demonstrate that at high dopant loading, carrier-carrier interactions can reduce the Seebeck coefficient. This results in systems with intrinsic disorder still following Heike’s formula for thermopower at high dopant density. Reducing these carrier-carrier interactions results in an increased Seebeck coefficient and power factor. Specifically a realistic reduction in carrier-carrier interactions can increase the power factor by more than a factor 15, increasing ZT above 1 for organic thermoelectrics.

The lack of highly conducting n-type materials greatly hampers the realization of performing organic thermoelectric generators (OTEGs); molecular doping (e.g. blending with electron-donating compounds) has stemmed as one of the possible solutions to enhance charge transport. Despite the actual interaction mechanism has not yet been fully disclosed [1, 2], benzimidazoles derivatives have so far emerged has one of the most promising class of organic n-type dopants due to their solution-processability, generality and relatively good shelf-life when stored in air [3].

Here we report on a modification of the widely used and commercially available 4-(2,3-Dihydro-1,3-dimethyl-1H-benzimidazol-2-yl)-N,N-dimethylbenzenamine (N-DMBI) obtained by replacing the two methyl end-groups with an iminostilbene moiety, creating an innovative derivative named IStBI. The introduction of a bulkier and more sterically hindered substituent bears significant changes in the crystallinity (shown through DSC and Thermogravimetric analysis) and packing in the dopant-host matrix, with only a marginal effect on the compound electronic structure, therefore preserving its electron-donating character. The study is conducted using the widely spread n-type polymer P(NDI2OD-T2), ensuring a large platform of comparable results available in the literature [4, 5, 6]. All the processing was carried out in solution and under N2-atmosphere due to the lack of air and moisture stability of the obtained semiconducting blends. Thin films, casted through spin-coating technique, have been extensively characterized in their electrical conductivity and Seebeck coefficient. After optimizing the processing conditions (i.e. solvent, dopant concentration and mixing temperature) and post-deposition thermal treatment, we obtained an electrical conductivity of 1.14 x 10^-2 S cm^-1, which is to the best of our knowledge among the highest values ever reported for doped P(NDI2OD-T2) [7]. This resulted in also an improved power factor of 8 x 10^-3 mW/mK². To support these findings structural and morphological analysis have been performed through GIWAXS analysis and AFM imaging techniques respectively, suggesting reduced disruption of the pristine polymer crystalline structure compared to other reported dopant molecules leading to improved charge transport properties.

Molecular doping is a powerful tool that is widely used for controlling the electrical properties of conjugated polymers. To design and develop polymer:dopant systems with improved ionization and/or dissociation efficiency, it is essential to measure the number of charges that are created on the backbone of the conjugated polymer. Here, it will be shown how a combination of UV-vis-NIR spectroscopy, spectro-electrochemistry and electron paramagnetic resonance (EPR) spectroscopy can be used to determine the number of polarons. P-doping of polythiophenes with alkyl or oligoether side chains will be compared in terms of the oxidation level and electrical properties. The strongest polaron density that can be achieved with oxidants such as F4TCNQ and Magic Blue is on the order of 5x10²⁶ m^-3, corresponding to an oxidation level of about 15%. Dopants with a high electron affinity, in combination with polymers that feature a low ionization energy of less than 5 eV give rise to double doping. The resulting exchange of two electrons between polymer and dopant allows to double the ionization efficiency and hence half the number of dopant molecules that are needed to achieve a certain oxidation level.

Three lactone-based rigid semiconducting polymers were deliberately designed to tackle one the major limitations in the development of n-type organic thermoelectrics, such as electrical conductivity and air stability. Here, we show that phenyl core maximization along the backbone can play a major role in optimizing the thermoelectric performance. Especially when combined with the rigid locked conformation imposed by aldol condensation. Experimental and theoretical investigations demonstrated that increasing the phenyl acene content from 0% phenyl (P-0), to 50% (P-50), and 75% (P-75) resulted in progressively i) larger electron affinities up to -4.37 eV due to the increased density of electron withdrawing groups, thereby suggesting a more favorable doping process when employing (N-DMBI) as the dopant. ii) Larger polaron delocalization due to the more planarized conformation, which ultimately led to lower hopping energy barrier. As a consequence, the electrical conductivity increased by three orders of magnitude to achieve values of up to 12 S/cm, which is one the scarcely reported n-type polymers with electrical conductivities over 10 S/cm. Thereby, enabling power factors of 13.2 μWm−1 K−2 , and is among the highest reported in literature for n-type polymers. Importantly, the electrical conductivity of the doped P-75 maintained a high value of 1.2 S/cm after 18 days of exposure to air.  These findings further highlight the benefits of phenyl core maximization to the electronic performance of the materials and suggest this approach as an effective design strategy to optimize the thermoelectric performance, thus also presenting new insights into material design guidelines for the future development of air stable n-type organic thermoelectrics.

Organic semiconductors are exciting candidates for organic thermoelectrics due to their low cost, abundant constituting elements, and, intriguingly, an assumed and overlooked low thermal conductivity. Therefore, research focuses primarily on improving the electronic contribution of the thermoelectric figure of merit ZT through processes like electrical doping and modification of the structural order. However, it is not clear how these processes affect thermal conductivity, and a mere handful of publications have explored this subject. Here we present a high-throughput methodology based on annealing- and doping-gradients to analyze and correlate the electrical and thermal conductivity. We can obtain data equivalent of > 100 samples using a single polymer film through this approach. As a testbed for our experiments, we employ poly(2,5- bis(3-alkylthiophen-2-yl)thieno[3,2-b]thiophene) (PBTTT) and the dopant 2,3,5,6-tetrafluoro-7,7,8,8-tetra-cyanoquinodimethane (F4TCNQ). We show that doping in PBTTT does not deteriorate the crystalline quality of the film but reduces thermal conductivity by a factor of two, even at relatively low doping levels. Our results indicate that the lattice contribution of the system dominates thermal transport and that impurities within the polymer network can have a drastic effect on the total thermal conductivity.

With the increased attention to alternative energy sources in the last decades, there has been a renewed interest in developing thermoelectric materials and devices. In this regard, organic semiconductors (OS) have emerged as a suitable alternative for low-power/low-cost thermoelectric devices. OS can be processed at room temperature using large-area printing techniques. They are potentially biocompatible and have intrinsically low thermal conductivity. Combined with the high abundance of their atomic elements, these features make them the ideal candidates for low-power electronic applications, such as wearable electronics, smart sensors, and the Internet of Things (IoT). Most of the printing technologies developed to date are suitable for thin-film patterning, which is not ideal for thermoelectric generators (TEGs), where the active material thickness should exceed at least 100 µm to sustain a temperature gradient properly. New geometries have been employed to solve this issue, exploiting flexible substrates to switch from a 2D to a 3D configuration through the folding of the flexible support. However, such devices still suffer from intrinsic limitations in the thermocouples density, and the addition of folding steps increases the chance of device failures due to the risks of open/short circuit in the generator. Here, we show the possibility of using 3D printing, a digital direct writing technique, to fabricate TEG with a vertical structure easily. The device, printed on top of a 2-µm-thick free-standing parylene substrate, is fully organic. The p-type and n-type legs are developed by printing PEDOT:PSS and BBL:PEI formulations, respectively. Finally, both the TEG scaffold and capping are realized using a 3D printable PDMS ink. Here we present the first 3D printed all-organic TEG, consisting of 8 thermocouples and capable of delivering 50 nW with a temperature gradient between its edges below 15 K.

Chemical doping is crucial for the operation of organic thermoelectric generators. This is typically achieved by adding dopant molecules to the polymer bulk, enabling high electrical conductivities. However, this process can result in poor stability and performance due to nonoptimal charge transfer and thin-film morphologies. Besides, once the dopant molecules are incorporated, they tend to diffuse through the free volume between polymer chains or escape during the heating steps, degrading both the electrical and mechanical properties of the semiconductor. In addition, the diffusion of dopant molecules could pose a risk when these materials are placed in contact with the human body like, e.g., in the case of wearable devices. Thus, molecular doping of organic semiconductors significantly limits their implementation into novel thermoelectric applications. In this presentation, we will report on our recent efforts to develop stable and highly conductive polymer blends based on mutual electrical doping. First, we will discuss the case of non-conjugated polymeric dopants and then move toward a new generation of conjugated polymer-donor/polymer-acceptor blends based on the effect of ground-state electron transfer in all-polymer heterojunctions. These molecular dopant-free systems hold promise for developing next-generation thermoelectric generators, particularly targeting high stability, efficiency, and power performance.

Chemically doped semiconducting polymers exhibit thermoelectric characteristics that range from localized (or hopping-like) transport to delocalized (or metal-like) transport.  While a multitude of electronic transport models have been proposed, none of them capture the full spectrum from localized to delocalized transport.[1]  Additionally, existing models do not quantitatively capture the dependency on charge carrier density (or carrier concentration or carrier ratio) that manifests through the measured temperature-dependent electrical conductivity and Seebeck coefficient.  Recently, we developed a semi-localized transport (SLoT) model, building upon past insight, that can describe the full spectrum from localized to delocalized transport.  This new model provides quantitative insight into charge carrier localization that is capable of more accurately describing electronic transport in a broad spectrum of organic thermoelectric semiconducting polymers. 

This invited talk will discuss our recent publication[2] where I will first present motivation showing the previous short comings of our collective understanding with existing models.  Next, I will briefly discuss the development of the SLoT model in the context of the organic thermoelectric field.  I will then discuss the utility and prospects of deeper insight that the SLoT model affords.  Then I will validate the model using the prototypical P3HT polymer doped with FeCl₃ and show its broad applicability in accurately describing other polymers/organic materials (namely, PBTTT, PA, PEDOT, SWCNT, and N2200) that were not previously well described by other models.  I will then extend this process to new polymers  and describe the deeper insight gained from this model.  I will then conclude my talk describing the relevant experimental measurements that research groups should undertake in characterization of their polymers to be able to use the SLoT model, in hopes of encouraging uniform material characterization internationally. 

The future implications of the SLoT model in developing thermoelectric polymers could be profound.  When coupled with chemical and structural characterization, the SLoT model connects the chemistry and structure to the macroscopic transport properties.  Once the SLoT model parameters are calculated, we can quantify fundamental limits to a polymer’s potential (e.g., ability to achieve high electrical conductivity or high Seebeck coefficient).  Ultimately, this allows us to accelerate the rational development of chemically doped organic thermoelectrics affording new functionality (e.g., thermal or electronic switching, thermoelectric cooling or power generation, etc.).

The ability to precisely control the equilibrium carrier concentration in organic semiconducting devices is of great interest. Solution processed doped layers are of extreme importance for high throughput production of organic electronic devices via roll-to-roll or ink-jet printing. The conventional picture of doping organic semiconductors involves full electron transfer from the semiconductor to the dopant (p-doping) or from the dopant to the semiconductor (n-doping) and the formation of species (polarons or bipolarons) that are the itinerant charge carriers. This process is known as “molecular doping”, which requires matching of the energy levels between the semiconductor and the dopant. In this talk, I will discuss unconventional doping methods that do not require energy level matching such as neutral polymers in non-aqueous solvents doped by Lewis acid-water complexes (Brønsted acids) and “self-doped” charged polymers in aqueous solution.

High performance flexible thermoelectric polymer-based material requires a high Seebeck coefficient, high electrical conductivity and low thermal conductivity.  It is still challenging to meet all these requirements. Here we report the design and synthesis pi-conjugated polyelectrolytes with different degree of naphthalenediimide (NDI) core with covalently tethered quaternized ammonium cations.  They are then incorporated with multi-valent oxalate anion which is a powerful latent electron donor when dispersed as small ion clusters to give in-situ n-dopable NDI polymers. We further optimised their morphologies using good-to-borderline solvent, and achieved both high seebeck coefficient and conductivity.  Their doping level and electron mobility are also characterised.

Tang, C.G;  Ang, M.C.Y.; Choo, K.K.; Keerthi, V.; Tan, J.K. ; Nur Syafiqah, M. ; Kugler, T. ; Burroughes, J.H.; Png, R.Q.; Chua, L.L.; Ho, P.K.H., Nature, 2016 ,  539, 536  DOI:  10.1038/nature20133

Tang, C.G;  Nur Syafiqah, M.; Koh, Q.M.; Zhao, C.; Zaini, J.; Seah, Q.J.; Humphries, MJ. Burroughes, J.H.; Png, R.Q.; Chua, L.L.; Ho, P.K.H.  Nature, 2019, 573, 519; DOI: 10.1038/s41586-019-1575-7

Developing highly conducting and air-stable n-type semiconductors for organic thermoelectrics have proven challenging, despite intensive research efforts in recent years. Thermoelectric generators convert heat into usable electricity, due to the Seebeck effect. Constructing an efficient thermoelectric generator based on organic semiconductors is difficult, however, primarily due to a significant shortage of suitable organic materials. While organic p-type semiconductors are abundant and relatively easy to dope, there is a shortage of suitable n-type materials and appropriate dopants. One popular approach is to extrinsically dope conjugated polymers. By mixing molecular dopants into a polymer matrix, it is possible to enhance the overall conductivity of the semiconductor to a certain degree, before morphological instabilities take over and cause microscopic phase separation of the dopant and polymer matrix, significantly limiting the doping efficiency.

In this paper, we will present the synthesis of intrinsic organic conductors, based on organometallic coordination polymers. We will discuss in detail how the choice of the ligand is crucial to control the electrical and thermoelectric properties. We will then take a more detailed look at the nature of the metal cation linking the organic ligands, and investigate their influence on charge carrier polarity and magnetic properties. Contrary to extrinsically doped semiconducting polymers, the organometallic coordination polymers exhibit excellent ambient and morphological stability and could pave the way to a new class of robust organic conductors for plastic electronic applications.

Halide perovskites are well known as promising candidates for photovoltaics and light-emitting diodes. Additionally, promising thermoelectric performance has been reported for a small number a halide perovskites, with this class of materials offering ultralow thermal conductivity, good Seebeck coefficients and potential advantages in processing and sustainability. However, there is not yet a good understanding of how thermoelectric performance of halide perovskites can be optimised. This presentation will report the thermoelectric figure of merit (ZT) for hybrid and inorganic perovskites [1,2]. It will discuss the origins of ultralow thermal conductivity and quanitfy both the Lorenz number and the thermal boundary resistance [3] in polycrystalline films. Extrinsic doping and self-doping will be discussed as methods to optimise the thermoelectric figure of merit zT, with values of zT reaching 0.14 in CsSnI₃ [2]. The case of self-doping by Sn-oxidation in CsSnI₃ will be examined in detail and strategies to improve performance and control the rate of oxidation by modification of deposition procedures, or by using mixed halide and mixed metal stoichiometries will be presented.

Electrochemical doping of organic conjugated polymers can be investigated in OECTs (Organic Electrochemical Transistors), which are sensitive sensors used for example in biological applications. They can be described using two circuits: an ionic circuit and an electronic circuit. The former arises from ions penetrating an organic channel due to switching of the gate, while the latter arises form source-drain electron flow across the organic channel. In the last years, advancing our knowledge of OECTs has allowed a drastic increase of their performance, however a complete picture of the two circuits and how they interact is still lacking. Here, we will introduce two new spectroscopic techniques to study the ionic and electronic transport processes in OECTs: 1) Time-resolved spectro-electrochemistry measures the kinetics of the ionic circuit. This measurement monitors the ion penetration into the organic channel and the doping processes with millisecond temporal resolution. 2) In-situ THz absorption spectroscopy, which investigates the electronic circuit and transport at different doping levels. It probes the nature of the electronic charges and the nanoscale conductivity of the organic channel. Results on PEDOT:PSS based OECTs and sandwish electrochemical devices will be presented. A deep understanding of the internal mechanisms occurring in OECTs is an essential step towards further interfacing with elaborate biological systems.