Final Report Summary - H2ESOT (Waste Heat to Electrical Energy via Sustainable Organic Thermoelectric Devices)
Executive Summary:
Entry paragraph for the General Public: Could you turn waste heat directly into electricity? Our program was about achieving that (we can do it, but not efficiently at present...if you want the bottom line already...). We do this using the thermoelectric (TE) effect - Let’s explain that: Everyone knows that if you put one end of a metal bar in a fire the heat spreads down the bar to the cold end. It does this by the atoms in the lattice vibrating, transferring the energy from one to another (we call these vibrations ‘phonons’). What very few people realise though is that some of the energy can also be transferred by the movement of electrons (or other charge carriers) down the conductor temperature gradient (we call these ‘polarons’). The more ‘polaron’ energy transported down the temperature gradient you get (rather than ‘phonon’ energy transfer) the more electrical power can get out from a better thermoelectric effect.
H2esot arose out of curiosity to investigate ad hoc and unpublished suggestions that the molecule tetrathiotetracene [TTT, CAS No. 193-44-2, see www.h2esot.com or http://www.nottingham.ac.uk/~pczsw/SWGroup/] could provide thermoelectric (TE) devices of very high efficiencies if suitably doped single crystals could be fashioned from it. Thermoelectric devices can generate electrical power directly from a temperature gradient and their efficiency at doing so is measured by a dimensionless comparator ZT, where T indicates the temperature of operation (in K). In theoretical models 1D conduction in single crystals of TTT2I3 can be predicted to afford ZT values >2. Employing materials this efficient, powerful new ways to trap and use profitably waste heat become viable. Over three years from 2013, 6 partners in chemistry, experimental and theoretical physics and electrical engineering came together to test the hypothesis that TTT2I3 and related materials could be exceptional thermoelectric materials. None of the experimental groups had worked together in the field of thermoelectrics prior to h2esot.
Traditionally TTT has been a very hard chemical to prepare efficiently and even more challenging to purify, and this has always limited interest in using the compound. H2esot chemists have developed two different robust routes to prepare both TTT and its derivatives in large scales via tetracene (see https://en.wikipedia.org/wiki/Tetracene). These routes are quick, cost efficient and amenable to large scales (5-50 g in academic laboratories with the potential for future industrial scale up). Aside from its ability to form TTT, tetracene is a widely used fragment in organic electronics and h2esot’s methodologies for preparing it quickly constitute a significant contribution. By using gas-streaming sublimation techniques it has been possible to prepare gram quantities of highly pure TTT and lustrous gold coloured crystals of TTT2I3 with average electrical conductivities of >1000 S cm-1 (for comparison: Cu is 6 × 10^5, Hg 1 × 10^4, graphite 20, Si 1.6 × 10^-5, rubber 10^-16 S cm-1). A new apparatus to measure this on small crystals was achieved. Although the h2esot crystals produce significant (for an organic solid) ZT efficiencies at room temperature the absolute values attained are very much smaller than those predicted by theory. The values attained are however, very significant given the short duration of the h2esot program and the fact that none of the team had worked together before. Nonetheless several ‘proof of concept’ TE generators were built using the crystals – the first time this has been achieved with TTT. An average crystal produces only small levels of power which is not commercially competitive with current systems, based on bismuth telluride, even though the latter is itself expensive.
Aside from potential future technological impacts the h2esot program has, thus far, generated 12 published papers (all open access by one method or another), >12 conference contributions, 5 popular science outputs (web news articles, facebook, Twitter) and contributed to the training of six PhD students, some supported independently of h2esot. The h2esot partners have continued to collaborate after the program.
Project Context and Objectives:
Based on its original ‘Description of Work’ Annex the key goals of the h2esot project were defined by six objectives - these are listed below together with the reasons why these ‘bottlenecks’ existed [given in brackets] are shown below:
• O1: Fast preparation of TTT and its parent tetracene (T) and derivatives thereof. [Known TTT preparations require many separate chemical transformations to be carried out to access the needed tetracene (T) precursors efficiently (often >5)].
• O2: Efficient purification of TTT to >99.99% purity. Preparation of crystalline samples of (TTT)aXb (X = I, a =2, b = 3; X = TCNQ, a = 1, b = 2) and properties thereof. [Simple TTT has a very poor solubility (10-4 M) and has been traditionally purified by (ineffective) laborious precipitations and static vacuum sublimations. Therefore, single crystal devices do not perform at their intrinsic limit so far].
• O3: Preparation of more process-able (on bulk scales) TTT-derivatives with improved solubility, n-type or other improved electronic behavior. [Few (<30) derivatives of TTT have ever been prepared – little or no correlation data between substituent effects and properties (electronic, solubility or otherwise) exists/is published].
• O4: Crystalline film deposition and characterisation of TTT derivatives. [The poor solubility of TTT invalidates solution processing. FET required knowledge on TTT-derivative film deposition and growth control does not exist].
• O5: Fabrication and quantification of devices built from new TTT-based materials. Defining industrially usable materials with respect to performance and stability under operation. [The lack of any commercial TTT prevents experimental physics and industrial electronics groups investigating device construction for industrial needs].
• O6: Underlying understanding of mechanisms resulting in the exceptional behavior of TTT-based systems. Defining theoretical aspects of new n-type TTT systems. [There is no present understanding in how the changing of TTT (placing a substituent at positions 1-4/6-10) affects the device behavior. Understanding of optimized TTT n-type thermoelectrics for devices need to be developed ‘from scratch’].
• Against O1: This has been fully achieved. Two different routes, but equally efficient (3 step), routes to the precursor tetracene (T) [developed by the Sofia and Nottingham groups]. These work on 5-10 g scales in an academic lab but the potential for scale-up to 100g and multi-kg levels appears practical. The Sofia group has optimized a procedure to prepare TTT in a chemically pure form (ca. 98-99% pure).
• Against O2: Gas-phase sublimation [developed at Wuerzburg] has provided the required purity of TTT (>99.99%, based on its residual electrical conductivity). Similarly a process for TTT2I3 could be developed, but not for TCNQ salts.
• Against O3: Some initial functionalized TTTs were prepared but these had properties (solubility in particular) that was not superior to the parent. As the latter performed well in gas-phase processing it was used as the major material investigated. Nevertheless at least one of the routes identified in O1 would allow access to such molecules if they became of interest in the future.
• Against O4: New apparatus allowed the preparation of thin films of TTT from small amounts of material. By appropriate doping a viable film TE device could be fabricated. While the TTT materials used do form crystals there are still issues with isolated crystallites/grain boundaries and this affects device performance.
• Against O5: Entirely new methods for the fabrication of TTT-based single crystal TEG devices have been developed for the first time. Most encouragingly the devices appear thermally stable to many heating/cooling cycles. However, the techniques developed are only appropriate for the research lab, not for production, and thus the TRL is only 3. The behavior of the final device is very significantly less than that predicted by theory at present.
• Against O6: Existing 1D mathematical modeling of the electrical, thermal and Seebeck performance of TTT-based materials has been extended to 2 and 3D systems and to n-type materials. The behavior of these models is in line with previous systems but, presently these do not converge fully with the experimental results. Similarly, due to the complexity of the model it does not allow predictions of optimal TTT derivatives preventing a strong link between theory, organic synthesis and material properties.
Overall the project has achieved fully the base-level objectives it set itself and demonstrated new potential FET at a technology readiness level (TRL) of 3. The absence of n-type materials and poor crystal efficiency has prevented us from attaining a TRL of 6 (the upper level of what could be expected from the project) presently. However, the risk of that was recognized at the inception of this FET project and such a goal is extremely ambitious for a 3 year program for an entirely new consortium with no previous links.
Project Results:
Report overview. This section of the final review of h2esot describes the position of the project at its final 36 month position. It is based on an analysis of the key outputs of the work packages 1-8 as measured against the objectives defined in the original Description of Work (DoW).
Review of Final Work package Positions
Work package 1: Efficient chemical synthesis of T and TTT and their derivatives (WP leader: Woodward, Uni. Nottingham; 1 x postdoc, 36 person-months)
The goals of WP1 was the development of a new routes to tetracene (T), and through that tetrathiotetracene, (TTT) that are of low cost and amenable to large scale production as defined in the original project objectives:
1. Identification of new synthetic organic chemistry able to allow rapid access to tetracene (T) and tetrathiotetracene (TTT) derivatives.
2. Comparison of the efficiency of these processes against traditional approaches to T and TTT (with WP3).
3. Supply of suitable samples to the consortium for purification and device fabrication (to WP2 and WP4).
4. Input into defining synthetic chemistry element to our Roadmap to Commercialisation (WP7).
Within the time-frame of the project WP1 has attained all of these goals; although the routes to the derivatives are less developed than the parent and the route to TTT (developed in WP3) could not be bettered. None of the original proposed routes proved effective, although one route was published and un-expected ‘spin-off’ discovery of the formation of 8-ring compounds was reported. The work package was thus redefined at the mid-term review, with the agreement of the Project Officer, to a new route (the subject of commercial discussions.
Full (currently confidential) experimental details are available presented in WP7 (Technology Roadmap), all of the steps have been trialled/optimised, many at large scales. The cost of tetracene (T) from this approach is highly competitive against current commercial suppliers. The final output from WP1 is summarised below.
Immediate impact: (i) Publications (2 at journal impact factors of 2.7571 and 11.3362). (ii) Presentations at three UK national and one international meeting (in addition to the three review meetings of the h2esot itself).
Near Future FET Impact: (i) Finalisation of the chemistry is continuing under a PhD studentship (J. Richie) that has run in parallel to h2esot on independent funding. He is expected to graduate in early 2017. (ii) Additional publications (one in production, one in preparation). (iii) One application for national UK funding on development of the work has been submitted.
Longer term FET impact: The Nottingham group will continue to interact with the h2esot partners and provide tetracene materials (from stock) where possible on route to developing more efficient tetracene-based TE devices. We have also begun trialling our tetracenes as components in other devices due to their potential low cost and ready availability.
Workpackage 2: Processing of TTT and derivatives, purification, growth of single crystals and characterisation (WP leader: Pflaum, Julius-Maximilian Uni; 1 x postdoc, 18 person-months; or PhD equivalents).
The role of WP2 in producing highly purified (>99.9%) tetrathiotetracene (TTT) and highly conducting single crystals of (TTT)2I3 was central to the project as without these materials no attempts at single crystal TE devices could be made. The original objectives for WP2 in the proposal description of work were:
1. Purification and crystal growth of the previously synthesized tetracene (T) and TTT derivatives.
2. Optimization of these processes by variation of the preparation conditions; evaluation by yield of purified material and structural quality of single crystals; feed-back to consortium partners No. 1 and 5 working on TTT synthesis.
3. Providing high quality single crystals for implementation in thermoelectric device applications.
4. Input to our Roadmap to Commercialisation via defining demands on crystalline samples for application.
Within the time-frame of the project WP2 has attained all of these goals. In fact the original objectives have been surpassed and extended. In addition to preparing single crystals of the prime p-type candidate for the project: (TTT)2I3 it became clear that published (at least in open readily available sources) data on the conductivity (both electrical and thermal) and Seebeck coefficient as a function of temperature is challenging due to the small size of the crystals. New apparatus built by the Pflaum group have allowed this for the first time
The electrical conductivity attained is in line with the crystals of Shchegolev and Yegubskii. At room temperature (300 K) the electrical conductivity of the h2esot crystals varies from crystal-to-crystal. To the best of our knowledge the value for the thermal conductivity of (TTT)2I has so far not been analysed to a great extent in the open access literature – although values akin to organic polymers [0.1 to 1.5 W/(K m)] have been occasionally suggested, in passing, for all organic conductors. Applying such estimates would predict ZT300K of 0.06 to 0.97 (i.e. in line with the goals of the initial h2esot programme at the upper limit value). However, using the new Würzburg apparatus the thermal conductivity values deduced for (TTT)2I3 are appreciably higher limiting ZT. Thus, although (TTT)2I3 single crystals can be realised they are not, in their present form, suitable for high efficiency TE generators. Single crystals from TTT and TCNQ could be attained but only as very small crystal dimensions. Only trace quantities could be prepared, again not presently suitable for n-type TEG production. Final output from WP2 is summarised below.
Immediate impact: (i) Publications (3 at journal impact factors of 0.385 0.023 and 3.664 ). Two publications were joint with the Casian group in Moldova. (ii) Presentations (2) at an international meeting on organic metals. (iii) Other public output (German language web journals and facebook) have brought the activities of h2esot to the attention of the wider German public.
Near Future FET Impact: (i) The combined property measurement apparatus in Würzburg is the first, as far as we are aware, to be able to completely determine ZT directly for very small sample sizes (single crystals) as a function of temperature. Such an approach is of utility and will have wider implication in the discovery of additional new organic TE materials. (ii) A publication detailing these measurements is in preparation. (iii) Should it be funded in 2016, a new Nottingham-Würzburg collaboration will begin by October 2016. (iv) Two PhD students (A. Steeger, F. Hüwe) will graduate through research training provided by h2esot.
Longer term FET impact: Research applications with other h2esot partners (EU and national) are being considered. Further research is needed in how to attain optimal p and n type TTT materials – although enormous progress has been made in the last three years.
Work package 3: Improved properties for TTT derivatives - electronic and solubility (WP leader: Dimitrov, Bulgarian Acad. of Sci.; 4 existing experienced research associates for 36 person-months).
The IOCCP team has provided the consortium with an optimized ‘classical’ route to bulk of tetracene and after this to explore the possibility of preparing derivatives with improved properties, as outlined in the original objectives:
1. Initial TTT production and TTT2I3 samples and supply to the consortium for purification and processing.
2. Access to substituted T and TTT and identification of substitution variants, with improved solubility and electronic properties (including interaction with WP2, 4-6).
3. Supply of optimal TTT derivatives.
All of the above goals have been attained. Fortunately, the parent TTT was determined to be optimal. Large batches of it could be attained by simple chemistry. Initial tetracene (T) can attained quickly, at low cost and in ca. 80-90% purity. Reaction of this T with sulfur and applying a purification procedure (described in D7.2) allows access to TTT at decagram scales. This approach is faster than that of WP1, but somewhat more expensive in reagents. By applying variations of this synthetic route four additional substituted TTT derivatives could be attained. However, in these cases the cost is significantly higher as a greater number of synthetic steps must be carried out to attain the necessary starting materials for producing TTT derivatives. None of these derivatives offer any significant advantage over (TTT) in sublimation processing.
Immediate impact: (i) The Sofia group has been the major provider of TTT to the project whose latter stages could not have been realized without their input.
Near Future FET Impact: (i) Publications: two are planned, due to its efficiency, speed and suitabilityfor use in academic labs one route has been suggested to the editorial board of Organic Syntheses (impact factor 1.607) as a suitable article. At least one publication describing the formation of the derivatives is in progress.
Longer term FET impact: We believe that the Seebeck performance of some of the derivatives will vary compared to the patent TTT, this may have future applications.
Workpackage 4: Crystalline thin film preparation and characterization: TTT, p-type (TTT)2I3, and n-type TTT(TCNQ)2(WP leader: Rutkis Group, Uni. Latvia, Solid State Physics; postdoc 36 person-months [Rutkis Group], 9.4 person-months postdoc or PhD equivalents [Pflaum Group]).
The goal of the group at Riga was to develop conditions for the preparation of TTT-thin-films and their use in devices. This was to be supported by some Organic Molecular Beam Deposition (OMBD) work by the group at Würzburg. The goals of WP4 as outlined in the original application were:
1. Design, upgrade and commission new techniques and equipment upgrades.
2. Development of crystalline film deposition techniques for TTT p-type and n-type derivatives.
3. Characterize physical properties of films and potential p/n junctions.
4. Work out recommendations for TE device fabrication out of TTT
Although only polycrystalline films could be attained all of the project goals could be attained and because of this the electrical conductivities attained are much lower than in the single crystals of WP2. The best p-type films are formed from co-deposition of TTT and I2 (described in D7.2 and D4.3). Similarly, co-deposition of TTT and TCNQ allowed access to n-type thin films. By using suitable masking approaches (see D4.3) in-plane devices could be fabricated by a successive deposition approach. Because the films are very thin (~1 micrometre) and polycrystalline their performance is poor and the electrical power output is very modest – of the order of 10-50 pW. However, such powers are still outputted even when the temperature difference across the p/n junction is rather small (10 oC).
Immediate impact: (i) Publications (1 at journal impact factor of 1.888 ), co-prepared with the Nottingham group. (ii) Presentations international meetings (7 in total). (iii) Web-based reporting.
Near Future FET Impact: (i) Further publications – it is expected that 1-3 further publications will result from the project. (ii) A PhD student (K. Pudzs) will graduate through research training provided by h2esot.
Longer term FET impact: An additional new FET technology opportunity arose out of this work, but this is still bound by confidentiality agreements at this stage.
Workpackage 5: Device Fabrication and integration into a system (WP leader: Simpson, European Thermodynamics (SME); 3 industrial staff, 37 person-months).
European Thermodynamics task represents the final stage of the h2esot project – construction of one or more ThermoElectrical Generator (TEG) test devices with the aim of showing ‘proof of principle’ and attaining new technology with an as high Technology Readiness Level (TRL) as possible. The original proposal set the following objectives:
1. Construct organic device test rigs for characterisation vs. temperature. Measurements of bulk materials, through characterisation equipment: Seebeck, electrical conductivity and thermal diffusivity.
2. Preliminary investigation of viability of uni-couple and p/n analogues.
3. Utilisation of optimal p/n device in a FET viability test.
In mid-2015 it was apparent that the poorer performance of the single crystalline TTT materials material would significantly limit final device performance. With the agreement of the Project Officer it was agreed that: (i) the finalised version of the TEG material should, on a power-by-weight criterion, be able to reach a level of at least 20% of that of current commercial materials (BiTe); (ii) the TEG should provide strategies towards/attempts to attain mW levels of sustained performance by the end of the project. By the end of the project 4 demonstration devices based on TTT2I3 crystals provided by the Würzburg group had been prepared (unileg architectures using a constantan ‘dummy’ n-leg, see deliverable D7.2). These had the following desired features:
• All the devices were fully functional, producing power, and the TEGs could be assembled in a reproducible manner.
• The behaviour of the devices could be predicted by the thermoelectric properties measured on the individual crystals (values attained at Würzburg).
• The devices are thermally robust and can be subjected to many heating-cooling cycles.
However, there are significant obstacles to be overcome to attain beneficial FET. Firstly, the power output per crystal is too small ~µW at a temperature difference of 75 oC. Current fully commercial devices using ‘active’ BiTe alloys typically provide 130 W/kg.27 Thus, only the very best crystals in the best device approach a value within 20% of current commercial technology – but these are the first viable organic single crystal TEG devices ever realised.
Attempts were made to upgrade the power output of one test device via capacitor charging with the aim of attaining transient mW power – but this was unsuccessful (see D5.4). One way that higher power outputs could be attained would be by use of a large number of crystals. For example, 10000 crystals would present a ‘footprint’ of 2-3 cm2 and provide a mW output at a 25 oC temperature difference – even with the currently poorer performing crystals.
Immediate impact: (i) The success of the World’s first single-crystal TEG announced via Twitter.
Near Future FET Impact: (i) Demonstration of the device will be presented on the h2esot web page.
Longer term FET impact: To attain something approaching a commercial device requires some very significant improvements. Most desirable would be an improvement in the underlying crystal performance (closer to that predicted by the theory of WP6).
Workpackage 6: Linking theory to properties – predicting new TTT materials
(WP leader: Casian, Uni. Moldova; 3 existing academic staff equivalent to 20 person-months, PhD student 36 person-months); 56 person-months total.
The goals of WP6 have been associated with modelling σ, S, power factor and ZT in the parent TTT2I3, initially and then extension of this to TTT(TCNQ)2. Specifically the goals set were:
1. Elaboration of a more complete crystal model than has been used so far, in order to model more precisely the electric and energy transport.
2. Calculation of ZT for (TTT)2I3 crystals of p-type with different degrees of purity and perfection in order to determine the expected values and assist the experimental physics work.
3. Modelling of electrical conductivity, Seebeck coefficient, the electronic thermal conductivity and ZT in prospect crystals of n-type, elaboration of recommendations for experimentalists (see especially WP5).
Multi-parameter models have been used to model the behaviour of the parent TTT2I3. These models take no account of the structure of the tetrathiotetracene core which is treated simply as a source of ‘hole’ (positive charge) carriers. In weak electrical fields, when only the energy states in the lowest energy band are involved, this is applicable. The most important (of the many) parameters used in the model are
• w1 the carrier transfer energy along the 1D axis (estimated to be 0.16 eV in TTT2I3).
• Do introduced to model the reduction of (hole) carrier efficiency via scattering defects and ‘non perfect’ lattice/crystal defects.
• epsilonF = EF/2w1, a dimensionless parameter where the TTT2I3 Fermi energy is measured in units of twice the 1D carrier transfer energy – as it is useful in graphical visualisations, parameters outputted from the model such as epsilonF are unit-less .
Crudely summation of ensembles of carriers over 1, 2 or 3D lattices have been achieved, along with their interactions with simulated phonon modes in the lattice model. A similar approach has been employed in simulation of TTT(TCNQ)2. The model predicts very high electrical conductivity at stoichiometric (dimensionless) Fermi energy. At the same value of epsilonF relatively low values of S are predicted in line with experiments performed by the Würzburg and Riga partners. Unfortunately, the very high conductivity predicted in the model does not match the experimentally observed 1D electrical conductivity of crystals prepared in Würzburg. The reasons for this discrepancy are not fully understood at present, but might be connected with a lower degree of perfection of grown crystals.
Immediate impact: (i) Publications (7 in journals with impact factors of 0.023,13 , 0.358,12 1.798 , and a new journal with no measured IF ). Two publications joint with the Pflaum group, Würzburg. (ii) Presentations at international meetings (7 in total). (iii) Web based comments from other electrical materials publications on the impact of this work.33
Near Future FET Impact: (i) Finalisation of h2esot work at the Technical University of Moldova will result in the award of two PhD qualifications (I. Sanduleac – he is expected to graduate in early 2016, and S. Andronic – she is expected to graduate in early 2017) and potentially 1-2 additional publications/conference proceedings.
Longer term FET impact: Through the h2esot project the future of parameterised low dimensional theoretical modelling has been assured and interactions with experimentalists invigorated and refreshed.
Work package 7: Roadmap to commercialisation(WP leaders: Simpson, European Thermodynamics (SME) and Woodward, Uni. Nottingham, 1 x Technology Transfer Officer, 6 person-months).
The key aim of WP7 was to assemble all of the ‘best-of-the-best’ technical output from the whole h2esot programme into a form that made it easy to make a judgment call as to the ‘deploy-ability’ of the derived FET into near market opportunities. Specifically the project specification called for:
1. Develop/expand the associate External Advisory Board and harness their input in developing both foreseen and new applications/market opportunities of TTT at production scales well beyond the preliminary devices scope of H2ESOT.
2. Maximise industrial opportunity access to licensed IP for precursors, materials, processes and devices (in agreement with the project’s Consortium) – simplifying future industrial involvement.
3. To significantly increase the critical mass of researchers and industrialists in organic thermoelectric materials.
At the end of the project immediate commercial thermoelectric activities are limited by the very modest performance of the TEG devices that are attained (~1-20%max performance of current commercial BiTe technology). The costs for the T and TTT starting materials for our final organic TE devices compare well with the costs of equivalent high grade tellurium (€5/g) and are very significantly lower than in current commercial routes (costs of €90-200+/g) for tetracene derivatives. Additionally, while we do not believe there is currently a market position for a TTT-based TEG sales of tetracene derivatives may be possible for other applications.
Immediate impact: (i) The roadmap has been successfully assembled and passed to all consortium members.
Near Future FET Impact: (i) We have identified a partner in our external advisory board who will attempt sales of the parent tetracene (T) in 2016 (which will be extended to derivatives if successful).
Longer term FET impact: (i) It is believed that the FET of this project can lead the way towards a new alternative FET technology. Not discussed here for commercial reasons.
Work package 8: Consortium management of H2ESOT (WP leader: Woodward, Uni. Nottingham, plus all other partners, 1 x administrator, 7 person-months).
Management of the project conducted nominally under WP8 has worked pretty well over the course of the project. It should be remembered that none of the partners had met prior to the kick-off meeting and had be selected purely on their potentials to carry out WP1-7 in the project. The management goals of WP8 identified at the genesis of the project were:
To arrange an initial ‘kick-off’ and 3 subsequent meetings to allow consortium planning in H2ESOT. To attain effective year-on-year management of the project, to perform a mid-term review and to disseminate scientific and user information to the wider community of potential future collaborators and end users of the FET. To manage IP and licensing issues and prepare/circulate all reports/associated materials.
By holding a monthly Skype review meeting all of the partners have been kept abreast of all developments and FET opportunities across the consortium. Appropriate inter-group Skype research discussions also took place. This has made the Coordinator’s job relatively easy: his assessment of the timing, delivery and efficiency of the whole programme is that nearly all elements of the program behaved as expected. From a project structure point of view all deliverables have been met and all milestones were attained. From a practical point of view two tasks hit significant difficulties and needed to be re-defined. In one chemistry work package (WP1) none of the originally defined routes to tetracene (T) proved practical at scales >0.25 g. Fortunately, new chemistry was identified and adopted by the programme leading to highly efficient syntheses by its end. For reasons still not clear to us we could not attain crystals with the very high ZT values predicted by the theory that drove this application. If experimentation is a fault then further improving purification may be possible – but highly challenging. Attempting an alternative theoretical approach might provide a way of checking the underlying theory. Either way through h2esot the global community of scientists interested in low dimensional organic conductors has significantly increased. The lower values of ZT associated with our single crystals of (TTT)2I3 meant that the highly ambitious goals of Task 4 in WP5 had to be redefined. While the desired ‘proof of concept’ TEG could be manufactured it will not be suitable for a commercial TEG device. However, in an unforeseen twist alternative FET has been identified (confidential at present) and this is being pursed.
For an overview of h2esot see the ‘executive summary’ at the start of this document. Outputs and other impacts are defined in Templates A1-A2.
Potential Impact:
Our h2esot programme has taken the first steps in demonstrating the ability of single crystalline materials to directly turn waste heat energy into electrical power. We find that not only is this a scientific idea that is appealing and potentially offers global benefits it is also a concept that the general public can easily engage with and become enthusiastic about. In addition to the impacts that are defined in Templates A1-A2 we identify the web resources below that have provided additional publicity/societal impact for the project.
(a) Presentation of route of original route at the Kick off meeting of UK Thermoelectric Network (http://www.thermoelectricnetwork.com/)
(b) Presentation of ‘spin off’ 8-ring chemistry developed in h2esot at the Royal Society of Chemistry 30th Postgraduate Symposium, Lilly UK Research and Development, Windlesham, UK, 10 September 2015. Lee Eccleshare wins prize for best ‘flash presentation’ (https://mobile.twitter.com/HudSAS/status/627155312932487168 http://blogs.rsc.org/ob/2015/12/16/heterocyclic-and-synthesis-group-postgraduate-meeting-prize-winners/).
(c) Presentation of 8-ring chemistry at XI NOST Conference for Research Scholars (JNOST), NISER, Bhubaneswar, India 14-17 December 2015 (http://www.nost.in/annuncements.html). Lee Eccleshare is on the back row, 9th from the left (http://www.nost.in/JNOST-2015.jpg).
(d) Understanding Anionic Chugaev Elimination in Pericyclic Tetracene Formation, L. Burroughs, J. Ritchie, S. Woodward, Tetrahedron 2016, in production.
(e) Tetracene Driven Materials for Organic Electronics and the Energy Agenda, application to Engineering and Physical Sciences Research Council (EPSRC), UK on 05.10.2015. This includes some involvement from the Pflaum and Simpson groups of h2esot.
(f) Energy exchange between phononic and electronic subsystem governing the nonlinear conduction in DCNQI2Cu, F. Huewe, A. Steeger, I. Bauer, S. Doerrich, P. Strohriegl, J. Pflaum (http://www.iscom2015.de/).
(g) News article: ‘Ein neuer Ansatz zum Energiesparen’ (A new approach to save energy) initial announcement of h2esot by Universität Würzburg commented on by Green-Tech Germany, 14.02.2013 (http://www.greentech-germany.com/ein-neuer-ansatz-zum-energiesparen-a240778) and Innovations Report, 05.02.2013 (http://www.innovations-report.de/html/berichte/energie-elektrotechnik/neuer-ansatz-energiesparen-209057.html). Additionally, one article commented on in Facebook (https://www.facebook.com/sonnenseite/posts/328334740620603).
(h) News article: ‘Strom aus Abfallwärme’ (Power from waste heat) covered by INGENIEUR.de 11.01.2016 (http://www.ingenieur.de/Fachbereiche/Umwelt-Recyclingtechnik/Strom-Abfallwaerme).
(i) News article: ‘Organische Halbleiter: Thermogeneratoren aus Molekülkristallen’ (Organic semiconductors: thermogenerators from molecular crystals) covered by elektroniknet.de 27.02.2013 (http://www.elektroniknet.de/power/energy-harvesting/artikel/95374/).
(j) 10th International Conference on Functional Materials and Nanotechnologies (FM&NT 2015) (link), 05-08.10.2015 Vilnius, Lithuania, Book of Abstract p.78 (http://www.fmnt.ff.vu.lt/wp-content/uploads/2015/10/konf_foto.jpg).
List of Websites:
Additional information on the h2esot program can be found at: www.h2esot.com
The partners were:
Simon Woodward, School of Chemistry, University of Nottingham, United Kingdom (Synthetic Chemistry and Coordinator)
http://www.nottingham.ac.uk/~pczsw/SWGroup/
Jens Pflaum, Experimentelle Physik 6, University of Wuezburg, Germany (Experimental Physics)
http://www.physik.uni-wuerzburg.de/ep6pflaum/
Vladimir Dimitrov, IOCCP, Bulgarian Academy of Sciences, Bulgaria (Synthetic Chemistry)
http://www.orgchm.bas.bg/list/oss/index.html
Martins Rutkis, Laboratory of Organic Materials, Institute of Solid State Physics, University of Latvia, Latvia (Thin Film Physics)
http://www.cfi.lu.lv/eng/about-issp/staff/persona/125/
Kevin Simpson, Technical Director, European Thermodynamics Ltd (Thermo-electrical Engineering)
http://www.europeanthermodynamics.com/about/people
Anatolie Casian, Technical University of Moldova, Chisinau, Moldova (Theoretical Physics)
http://www.utm.md/
Entry paragraph for the General Public: Could you turn waste heat directly into electricity? Our program was about achieving that (we can do it, but not efficiently at present...if you want the bottom line already...). We do this using the thermoelectric (TE) effect - Let’s explain that: Everyone knows that if you put one end of a metal bar in a fire the heat spreads down the bar to the cold end. It does this by the atoms in the lattice vibrating, transferring the energy from one to another (we call these vibrations ‘phonons’). What very few people realise though is that some of the energy can also be transferred by the movement of electrons (or other charge carriers) down the conductor temperature gradient (we call these ‘polarons’). The more ‘polaron’ energy transported down the temperature gradient you get (rather than ‘phonon’ energy transfer) the more electrical power can get out from a better thermoelectric effect.
H2esot arose out of curiosity to investigate ad hoc and unpublished suggestions that the molecule tetrathiotetracene [TTT, CAS No. 193-44-2, see www.h2esot.com or http://www.nottingham.ac.uk/~pczsw/SWGroup/] could provide thermoelectric (TE) devices of very high efficiencies if suitably doped single crystals could be fashioned from it. Thermoelectric devices can generate electrical power directly from a temperature gradient and their efficiency at doing so is measured by a dimensionless comparator ZT, where T indicates the temperature of operation (in K). In theoretical models 1D conduction in single crystals of TTT2I3 can be predicted to afford ZT values >2. Employing materials this efficient, powerful new ways to trap and use profitably waste heat become viable. Over three years from 2013, 6 partners in chemistry, experimental and theoretical physics and electrical engineering came together to test the hypothesis that TTT2I3 and related materials could be exceptional thermoelectric materials. None of the experimental groups had worked together in the field of thermoelectrics prior to h2esot.
Traditionally TTT has been a very hard chemical to prepare efficiently and even more challenging to purify, and this has always limited interest in using the compound. H2esot chemists have developed two different robust routes to prepare both TTT and its derivatives in large scales via tetracene (see https://en.wikipedia.org/wiki/Tetracene). These routes are quick, cost efficient and amenable to large scales (5-50 g in academic laboratories with the potential for future industrial scale up). Aside from its ability to form TTT, tetracene is a widely used fragment in organic electronics and h2esot’s methodologies for preparing it quickly constitute a significant contribution. By using gas-streaming sublimation techniques it has been possible to prepare gram quantities of highly pure TTT and lustrous gold coloured crystals of TTT2I3 with average electrical conductivities of >1000 S cm-1 (for comparison: Cu is 6 × 10^5, Hg 1 × 10^4, graphite 20, Si 1.6 × 10^-5, rubber 10^-16 S cm-1). A new apparatus to measure this on small crystals was achieved. Although the h2esot crystals produce significant (for an organic solid) ZT efficiencies at room temperature the absolute values attained are very much smaller than those predicted by theory. The values attained are however, very significant given the short duration of the h2esot program and the fact that none of the team had worked together before. Nonetheless several ‘proof of concept’ TE generators were built using the crystals – the first time this has been achieved with TTT. An average crystal produces only small levels of power which is not commercially competitive with current systems, based on bismuth telluride, even though the latter is itself expensive.
Aside from potential future technological impacts the h2esot program has, thus far, generated 12 published papers (all open access by one method or another), >12 conference contributions, 5 popular science outputs (web news articles, facebook, Twitter) and contributed to the training of six PhD students, some supported independently of h2esot. The h2esot partners have continued to collaborate after the program.
Project Context and Objectives:
Based on its original ‘Description of Work’ Annex the key goals of the h2esot project were defined by six objectives - these are listed below together with the reasons why these ‘bottlenecks’ existed [given in brackets] are shown below:
• O1: Fast preparation of TTT and its parent tetracene (T) and derivatives thereof. [Known TTT preparations require many separate chemical transformations to be carried out to access the needed tetracene (T) precursors efficiently (often >5)].
• O2: Efficient purification of TTT to >99.99% purity. Preparation of crystalline samples of (TTT)aXb (X = I, a =2, b = 3; X = TCNQ, a = 1, b = 2) and properties thereof. [Simple TTT has a very poor solubility (10-4 M) and has been traditionally purified by (ineffective) laborious precipitations and static vacuum sublimations. Therefore, single crystal devices do not perform at their intrinsic limit so far].
• O3: Preparation of more process-able (on bulk scales) TTT-derivatives with improved solubility, n-type or other improved electronic behavior. [Few (<30) derivatives of TTT have ever been prepared – little or no correlation data between substituent effects and properties (electronic, solubility or otherwise) exists/is published].
• O4: Crystalline film deposition and characterisation of TTT derivatives. [The poor solubility of TTT invalidates solution processing. FET required knowledge on TTT-derivative film deposition and growth control does not exist].
• O5: Fabrication and quantification of devices built from new TTT-based materials. Defining industrially usable materials with respect to performance and stability under operation. [The lack of any commercial TTT prevents experimental physics and industrial electronics groups investigating device construction for industrial needs].
• O6: Underlying understanding of mechanisms resulting in the exceptional behavior of TTT-based systems. Defining theoretical aspects of new n-type TTT systems. [There is no present understanding in how the changing of TTT (placing a substituent at positions 1-4/6-10) affects the device behavior. Understanding of optimized TTT n-type thermoelectrics for devices need to be developed ‘from scratch’].
• Against O1: This has been fully achieved. Two different routes, but equally efficient (3 step), routes to the precursor tetracene (T) [developed by the Sofia and Nottingham groups]. These work on 5-10 g scales in an academic lab but the potential for scale-up to 100g and multi-kg levels appears practical. The Sofia group has optimized a procedure to prepare TTT in a chemically pure form (ca. 98-99% pure).
• Against O2: Gas-phase sublimation [developed at Wuerzburg] has provided the required purity of TTT (>99.99%, based on its residual electrical conductivity). Similarly a process for TTT2I3 could be developed, but not for TCNQ salts.
• Against O3: Some initial functionalized TTTs were prepared but these had properties (solubility in particular) that was not superior to the parent. As the latter performed well in gas-phase processing it was used as the major material investigated. Nevertheless at least one of the routes identified in O1 would allow access to such molecules if they became of interest in the future.
• Against O4: New apparatus allowed the preparation of thin films of TTT from small amounts of material. By appropriate doping a viable film TE device could be fabricated. While the TTT materials used do form crystals there are still issues with isolated crystallites/grain boundaries and this affects device performance.
• Against O5: Entirely new methods for the fabrication of TTT-based single crystal TEG devices have been developed for the first time. Most encouragingly the devices appear thermally stable to many heating/cooling cycles. However, the techniques developed are only appropriate for the research lab, not for production, and thus the TRL is only 3. The behavior of the final device is very significantly less than that predicted by theory at present.
• Against O6: Existing 1D mathematical modeling of the electrical, thermal and Seebeck performance of TTT-based materials has been extended to 2 and 3D systems and to n-type materials. The behavior of these models is in line with previous systems but, presently these do not converge fully with the experimental results. Similarly, due to the complexity of the model it does not allow predictions of optimal TTT derivatives preventing a strong link between theory, organic synthesis and material properties.
Overall the project has achieved fully the base-level objectives it set itself and demonstrated new potential FET at a technology readiness level (TRL) of 3. The absence of n-type materials and poor crystal efficiency has prevented us from attaining a TRL of 6 (the upper level of what could be expected from the project) presently. However, the risk of that was recognized at the inception of this FET project and such a goal is extremely ambitious for a 3 year program for an entirely new consortium with no previous links.
Project Results:
Report overview. This section of the final review of h2esot describes the position of the project at its final 36 month position. It is based on an analysis of the key outputs of the work packages 1-8 as measured against the objectives defined in the original Description of Work (DoW).
Review of Final Work package Positions
Work package 1: Efficient chemical synthesis of T and TTT and their derivatives (WP leader: Woodward, Uni. Nottingham; 1 x postdoc, 36 person-months)
The goals of WP1 was the development of a new routes to tetracene (T), and through that tetrathiotetracene, (TTT) that are of low cost and amenable to large scale production as defined in the original project objectives:
1. Identification of new synthetic organic chemistry able to allow rapid access to tetracene (T) and tetrathiotetracene (TTT) derivatives.
2. Comparison of the efficiency of these processes against traditional approaches to T and TTT (with WP3).
3. Supply of suitable samples to the consortium for purification and device fabrication (to WP2 and WP4).
4. Input into defining synthetic chemistry element to our Roadmap to Commercialisation (WP7).
Within the time-frame of the project WP1 has attained all of these goals; although the routes to the derivatives are less developed than the parent and the route to TTT (developed in WP3) could not be bettered. None of the original proposed routes proved effective, although one route was published and un-expected ‘spin-off’ discovery of the formation of 8-ring compounds was reported. The work package was thus redefined at the mid-term review, with the agreement of the Project Officer, to a new route (the subject of commercial discussions.
Full (currently confidential) experimental details are available presented in WP7 (Technology Roadmap), all of the steps have been trialled/optimised, many at large scales. The cost of tetracene (T) from this approach is highly competitive against current commercial suppliers. The final output from WP1 is summarised below.
Immediate impact: (i) Publications (2 at journal impact factors of 2.7571 and 11.3362). (ii) Presentations at three UK national and one international meeting (in addition to the three review meetings of the h2esot itself).
Near Future FET Impact: (i) Finalisation of the chemistry is continuing under a PhD studentship (J. Richie) that has run in parallel to h2esot on independent funding. He is expected to graduate in early 2017. (ii) Additional publications (one in production, one in preparation). (iii) One application for national UK funding on development of the work has been submitted.
Longer term FET impact: The Nottingham group will continue to interact with the h2esot partners and provide tetracene materials (from stock) where possible on route to developing more efficient tetracene-based TE devices. We have also begun trialling our tetracenes as components in other devices due to their potential low cost and ready availability.
Workpackage 2: Processing of TTT and derivatives, purification, growth of single crystals and characterisation (WP leader: Pflaum, Julius-Maximilian Uni; 1 x postdoc, 18 person-months; or PhD equivalents).
The role of WP2 in producing highly purified (>99.9%) tetrathiotetracene (TTT) and highly conducting single crystals of (TTT)2I3 was central to the project as without these materials no attempts at single crystal TE devices could be made. The original objectives for WP2 in the proposal description of work were:
1. Purification and crystal growth of the previously synthesized tetracene (T) and TTT derivatives.
2. Optimization of these processes by variation of the preparation conditions; evaluation by yield of purified material and structural quality of single crystals; feed-back to consortium partners No. 1 and 5 working on TTT synthesis.
3. Providing high quality single crystals for implementation in thermoelectric device applications.
4. Input to our Roadmap to Commercialisation via defining demands on crystalline samples for application.
Within the time-frame of the project WP2 has attained all of these goals. In fact the original objectives have been surpassed and extended. In addition to preparing single crystals of the prime p-type candidate for the project: (TTT)2I3 it became clear that published (at least in open readily available sources) data on the conductivity (both electrical and thermal) and Seebeck coefficient as a function of temperature is challenging due to the small size of the crystals. New apparatus built by the Pflaum group have allowed this for the first time
The electrical conductivity attained is in line with the crystals of Shchegolev and Yegubskii. At room temperature (300 K) the electrical conductivity of the h2esot crystals varies from crystal-to-crystal. To the best of our knowledge the value for the thermal conductivity of (TTT)2I has so far not been analysed to a great extent in the open access literature – although values akin to organic polymers [0.1 to 1.5 W/(K m)] have been occasionally suggested, in passing, for all organic conductors. Applying such estimates would predict ZT300K of 0.06 to 0.97 (i.e. in line with the goals of the initial h2esot programme at the upper limit value). However, using the new Würzburg apparatus the thermal conductivity values deduced for (TTT)2I3 are appreciably higher limiting ZT. Thus, although (TTT)2I3 single crystals can be realised they are not, in their present form, suitable for high efficiency TE generators. Single crystals from TTT and TCNQ could be attained but only as very small crystal dimensions. Only trace quantities could be prepared, again not presently suitable for n-type TEG production. Final output from WP2 is summarised below.
Immediate impact: (i) Publications (3 at journal impact factors of 0.385 0.023 and 3.664 ). Two publications were joint with the Casian group in Moldova. (ii) Presentations (2) at an international meeting on organic metals. (iii) Other public output (German language web journals and facebook) have brought the activities of h2esot to the attention of the wider German public.
Near Future FET Impact: (i) The combined property measurement apparatus in Würzburg is the first, as far as we are aware, to be able to completely determine ZT directly for very small sample sizes (single crystals) as a function of temperature. Such an approach is of utility and will have wider implication in the discovery of additional new organic TE materials. (ii) A publication detailing these measurements is in preparation. (iii) Should it be funded in 2016, a new Nottingham-Würzburg collaboration will begin by October 2016. (iv) Two PhD students (A. Steeger, F. Hüwe) will graduate through research training provided by h2esot.
Longer term FET impact: Research applications with other h2esot partners (EU and national) are being considered. Further research is needed in how to attain optimal p and n type TTT materials – although enormous progress has been made in the last three years.
Work package 3: Improved properties for TTT derivatives - electronic and solubility (WP leader: Dimitrov, Bulgarian Acad. of Sci.; 4 existing experienced research associates for 36 person-months).
The IOCCP team has provided the consortium with an optimized ‘classical’ route to bulk of tetracene and after this to explore the possibility of preparing derivatives with improved properties, as outlined in the original objectives:
1. Initial TTT production and TTT2I3 samples and supply to the consortium for purification and processing.
2. Access to substituted T and TTT and identification of substitution variants, with improved solubility and electronic properties (including interaction with WP2, 4-6).
3. Supply of optimal TTT derivatives.
All of the above goals have been attained. Fortunately, the parent TTT was determined to be optimal. Large batches of it could be attained by simple chemistry. Initial tetracene (T) can attained quickly, at low cost and in ca. 80-90% purity. Reaction of this T with sulfur and applying a purification procedure (described in D7.2) allows access to TTT at decagram scales. This approach is faster than that of WP1, but somewhat more expensive in reagents. By applying variations of this synthetic route four additional substituted TTT derivatives could be attained. However, in these cases the cost is significantly higher as a greater number of synthetic steps must be carried out to attain the necessary starting materials for producing TTT derivatives. None of these derivatives offer any significant advantage over (TTT) in sublimation processing.
Immediate impact: (i) The Sofia group has been the major provider of TTT to the project whose latter stages could not have been realized without their input.
Near Future FET Impact: (i) Publications: two are planned, due to its efficiency, speed and suitabilityfor use in academic labs one route has been suggested to the editorial board of Organic Syntheses (impact factor 1.607) as a suitable article. At least one publication describing the formation of the derivatives is in progress.
Longer term FET impact: We believe that the Seebeck performance of some of the derivatives will vary compared to the patent TTT, this may have future applications.
Workpackage 4: Crystalline thin film preparation and characterization: TTT, p-type (TTT)2I3, and n-type TTT(TCNQ)2(WP leader: Rutkis Group, Uni. Latvia, Solid State Physics; postdoc 36 person-months [Rutkis Group], 9.4 person-months postdoc or PhD equivalents [Pflaum Group]).
The goal of the group at Riga was to develop conditions for the preparation of TTT-thin-films and their use in devices. This was to be supported by some Organic Molecular Beam Deposition (OMBD) work by the group at Würzburg. The goals of WP4 as outlined in the original application were:
1. Design, upgrade and commission new techniques and equipment upgrades.
2. Development of crystalline film deposition techniques for TTT p-type and n-type derivatives.
3. Characterize physical properties of films and potential p/n junctions.
4. Work out recommendations for TE device fabrication out of TTT
Although only polycrystalline films could be attained all of the project goals could be attained and because of this the electrical conductivities attained are much lower than in the single crystals of WP2. The best p-type films are formed from co-deposition of TTT and I2 (described in D7.2 and D4.3). Similarly, co-deposition of TTT and TCNQ allowed access to n-type thin films. By using suitable masking approaches (see D4.3) in-plane devices could be fabricated by a successive deposition approach. Because the films are very thin (~1 micrometre) and polycrystalline their performance is poor and the electrical power output is very modest – of the order of 10-50 pW. However, such powers are still outputted even when the temperature difference across the p/n junction is rather small (10 oC).
Immediate impact: (i) Publications (1 at journal impact factor of 1.888 ), co-prepared with the Nottingham group. (ii) Presentations international meetings (7 in total). (iii) Web-based reporting.
Near Future FET Impact: (i) Further publications – it is expected that 1-3 further publications will result from the project. (ii) A PhD student (K. Pudzs) will graduate through research training provided by h2esot.
Longer term FET impact: An additional new FET technology opportunity arose out of this work, but this is still bound by confidentiality agreements at this stage.
Workpackage 5: Device Fabrication and integration into a system (WP leader: Simpson, European Thermodynamics (SME); 3 industrial staff, 37 person-months).
European Thermodynamics task represents the final stage of the h2esot project – construction of one or more ThermoElectrical Generator (TEG) test devices with the aim of showing ‘proof of principle’ and attaining new technology with an as high Technology Readiness Level (TRL) as possible. The original proposal set the following objectives:
1. Construct organic device test rigs for characterisation vs. temperature. Measurements of bulk materials, through characterisation equipment: Seebeck, electrical conductivity and thermal diffusivity.
2. Preliminary investigation of viability of uni-couple and p/n analogues.
3. Utilisation of optimal p/n device in a FET viability test.
In mid-2015 it was apparent that the poorer performance of the single crystalline TTT materials material would significantly limit final device performance. With the agreement of the Project Officer it was agreed that: (i) the finalised version of the TEG material should, on a power-by-weight criterion, be able to reach a level of at least 20% of that of current commercial materials (BiTe); (ii) the TEG should provide strategies towards/attempts to attain mW levels of sustained performance by the end of the project. By the end of the project 4 demonstration devices based on TTT2I3 crystals provided by the Würzburg group had been prepared (unileg architectures using a constantan ‘dummy’ n-leg, see deliverable D7.2). These had the following desired features:
• All the devices were fully functional, producing power, and the TEGs could be assembled in a reproducible manner.
• The behaviour of the devices could be predicted by the thermoelectric properties measured on the individual crystals (values attained at Würzburg).
• The devices are thermally robust and can be subjected to many heating-cooling cycles.
However, there are significant obstacles to be overcome to attain beneficial FET. Firstly, the power output per crystal is too small ~µW at a temperature difference of 75 oC. Current fully commercial devices using ‘active’ BiTe alloys typically provide 130 W/kg.27 Thus, only the very best crystals in the best device approach a value within 20% of current commercial technology – but these are the first viable organic single crystal TEG devices ever realised.
Attempts were made to upgrade the power output of one test device via capacitor charging with the aim of attaining transient mW power – but this was unsuccessful (see D5.4). One way that higher power outputs could be attained would be by use of a large number of crystals. For example, 10000 crystals would present a ‘footprint’ of 2-3 cm2 and provide a mW output at a 25 oC temperature difference – even with the currently poorer performing crystals.
Immediate impact: (i) The success of the World’s first single-crystal TEG announced via Twitter.
Near Future FET Impact: (i) Demonstration of the device will be presented on the h2esot web page.
Longer term FET impact: To attain something approaching a commercial device requires some very significant improvements. Most desirable would be an improvement in the underlying crystal performance (closer to that predicted by the theory of WP6).
Workpackage 6: Linking theory to properties – predicting new TTT materials
(WP leader: Casian, Uni. Moldova; 3 existing academic staff equivalent to 20 person-months, PhD student 36 person-months); 56 person-months total.
The goals of WP6 have been associated with modelling σ, S, power factor and ZT in the parent TTT2I3, initially and then extension of this to TTT(TCNQ)2. Specifically the goals set were:
1. Elaboration of a more complete crystal model than has been used so far, in order to model more precisely the electric and energy transport.
2. Calculation of ZT for (TTT)2I3 crystals of p-type with different degrees of purity and perfection in order to determine the expected values and assist the experimental physics work.
3. Modelling of electrical conductivity, Seebeck coefficient, the electronic thermal conductivity and ZT in prospect crystals of n-type, elaboration of recommendations for experimentalists (see especially WP5).
Multi-parameter models have been used to model the behaviour of the parent TTT2I3. These models take no account of the structure of the tetrathiotetracene core which is treated simply as a source of ‘hole’ (positive charge) carriers. In weak electrical fields, when only the energy states in the lowest energy band are involved, this is applicable. The most important (of the many) parameters used in the model are
• w1 the carrier transfer energy along the 1D axis (estimated to be 0.16 eV in TTT2I3).
• Do introduced to model the reduction of (hole) carrier efficiency via scattering defects and ‘non perfect’ lattice/crystal defects.
• epsilonF = EF/2w1, a dimensionless parameter where the TTT2I3 Fermi energy is measured in units of twice the 1D carrier transfer energy – as it is useful in graphical visualisations, parameters outputted from the model such as epsilonF are unit-less .
Crudely summation of ensembles of carriers over 1, 2 or 3D lattices have been achieved, along with their interactions with simulated phonon modes in the lattice model. A similar approach has been employed in simulation of TTT(TCNQ)2. The model predicts very high electrical conductivity at stoichiometric (dimensionless) Fermi energy. At the same value of epsilonF relatively low values of S are predicted in line with experiments performed by the Würzburg and Riga partners. Unfortunately, the very high conductivity predicted in the model does not match the experimentally observed 1D electrical conductivity of crystals prepared in Würzburg. The reasons for this discrepancy are not fully understood at present, but might be connected with a lower degree of perfection of grown crystals.
Immediate impact: (i) Publications (7 in journals with impact factors of 0.023,13 , 0.358,12 1.798 , and a new journal with no measured IF ). Two publications joint with the Pflaum group, Würzburg. (ii) Presentations at international meetings (7 in total). (iii) Web based comments from other electrical materials publications on the impact of this work.33
Near Future FET Impact: (i) Finalisation of h2esot work at the Technical University of Moldova will result in the award of two PhD qualifications (I. Sanduleac – he is expected to graduate in early 2016, and S. Andronic – she is expected to graduate in early 2017) and potentially 1-2 additional publications/conference proceedings.
Longer term FET impact: Through the h2esot project the future of parameterised low dimensional theoretical modelling has been assured and interactions with experimentalists invigorated and refreshed.
Work package 7: Roadmap to commercialisation(WP leaders: Simpson, European Thermodynamics (SME) and Woodward, Uni. Nottingham, 1 x Technology Transfer Officer, 6 person-months).
The key aim of WP7 was to assemble all of the ‘best-of-the-best’ technical output from the whole h2esot programme into a form that made it easy to make a judgment call as to the ‘deploy-ability’ of the derived FET into near market opportunities. Specifically the project specification called for:
1. Develop/expand the associate External Advisory Board and harness their input in developing both foreseen and new applications/market opportunities of TTT at production scales well beyond the preliminary devices scope of H2ESOT.
2. Maximise industrial opportunity access to licensed IP for precursors, materials, processes and devices (in agreement with the project’s Consortium) – simplifying future industrial involvement.
3. To significantly increase the critical mass of researchers and industrialists in organic thermoelectric materials.
At the end of the project immediate commercial thermoelectric activities are limited by the very modest performance of the TEG devices that are attained (~1-20%max performance of current commercial BiTe technology). The costs for the T and TTT starting materials for our final organic TE devices compare well with the costs of equivalent high grade tellurium (€5/g) and are very significantly lower than in current commercial routes (costs of €90-200+/g) for tetracene derivatives. Additionally, while we do not believe there is currently a market position for a TTT-based TEG sales of tetracene derivatives may be possible for other applications.
Immediate impact: (i) The roadmap has been successfully assembled and passed to all consortium members.
Near Future FET Impact: (i) We have identified a partner in our external advisory board who will attempt sales of the parent tetracene (T) in 2016 (which will be extended to derivatives if successful).
Longer term FET impact: (i) It is believed that the FET of this project can lead the way towards a new alternative FET technology. Not discussed here for commercial reasons.
Work package 8: Consortium management of H2ESOT (WP leader: Woodward, Uni. Nottingham, plus all other partners, 1 x administrator, 7 person-months).
Management of the project conducted nominally under WP8 has worked pretty well over the course of the project. It should be remembered that none of the partners had met prior to the kick-off meeting and had be selected purely on their potentials to carry out WP1-7 in the project. The management goals of WP8 identified at the genesis of the project were:
To arrange an initial ‘kick-off’ and 3 subsequent meetings to allow consortium planning in H2ESOT. To attain effective year-on-year management of the project, to perform a mid-term review and to disseminate scientific and user information to the wider community of potential future collaborators and end users of the FET. To manage IP and licensing issues and prepare/circulate all reports/associated materials.
By holding a monthly Skype review meeting all of the partners have been kept abreast of all developments and FET opportunities across the consortium. Appropriate inter-group Skype research discussions also took place. This has made the Coordinator’s job relatively easy: his assessment of the timing, delivery and efficiency of the whole programme is that nearly all elements of the program behaved as expected. From a project structure point of view all deliverables have been met and all milestones were attained. From a practical point of view two tasks hit significant difficulties and needed to be re-defined. In one chemistry work package (WP1) none of the originally defined routes to tetracene (T) proved practical at scales >0.25 g. Fortunately, new chemistry was identified and adopted by the programme leading to highly efficient syntheses by its end. For reasons still not clear to us we could not attain crystals with the very high ZT values predicted by the theory that drove this application. If experimentation is a fault then further improving purification may be possible – but highly challenging. Attempting an alternative theoretical approach might provide a way of checking the underlying theory. Either way through h2esot the global community of scientists interested in low dimensional organic conductors has significantly increased. The lower values of ZT associated with our single crystals of (TTT)2I3 meant that the highly ambitious goals of Task 4 in WP5 had to be redefined. While the desired ‘proof of concept’ TEG could be manufactured it will not be suitable for a commercial TEG device. However, in an unforeseen twist alternative FET has been identified (confidential at present) and this is being pursed.
For an overview of h2esot see the ‘executive summary’ at the start of this document. Outputs and other impacts are defined in Templates A1-A2.
Potential Impact:
Our h2esot programme has taken the first steps in demonstrating the ability of single crystalline materials to directly turn waste heat energy into electrical power. We find that not only is this a scientific idea that is appealing and potentially offers global benefits it is also a concept that the general public can easily engage with and become enthusiastic about. In addition to the impacts that are defined in Templates A1-A2 we identify the web resources below that have provided additional publicity/societal impact for the project.
(a) Presentation of route of original route at the Kick off meeting of UK Thermoelectric Network (http://www.thermoelectricnetwork.com/)
(b) Presentation of ‘spin off’ 8-ring chemistry developed in h2esot at the Royal Society of Chemistry 30th Postgraduate Symposium, Lilly UK Research and Development, Windlesham, UK, 10 September 2015. Lee Eccleshare wins prize for best ‘flash presentation’ (https://mobile.twitter.com/HudSAS/status/627155312932487168 http://blogs.rsc.org/ob/2015/12/16/heterocyclic-and-synthesis-group-postgraduate-meeting-prize-winners/).
(c) Presentation of 8-ring chemistry at XI NOST Conference for Research Scholars (JNOST), NISER, Bhubaneswar, India 14-17 December 2015 (http://www.nost.in/annuncements.html). Lee Eccleshare is on the back row, 9th from the left (http://www.nost.in/JNOST-2015.jpg).
(d) Understanding Anionic Chugaev Elimination in Pericyclic Tetracene Formation, L. Burroughs, J. Ritchie, S. Woodward, Tetrahedron 2016, in production.
(e) Tetracene Driven Materials for Organic Electronics and the Energy Agenda, application to Engineering and Physical Sciences Research Council (EPSRC), UK on 05.10.2015. This includes some involvement from the Pflaum and Simpson groups of h2esot.
(f) Energy exchange between phononic and electronic subsystem governing the nonlinear conduction in DCNQI2Cu, F. Huewe, A. Steeger, I. Bauer, S. Doerrich, P. Strohriegl, J. Pflaum (http://www.iscom2015.de/).
(g) News article: ‘Ein neuer Ansatz zum Energiesparen’ (A new approach to save energy) initial announcement of h2esot by Universität Würzburg commented on by Green-Tech Germany, 14.02.2013 (http://www.greentech-germany.com/ein-neuer-ansatz-zum-energiesparen-a240778) and Innovations Report, 05.02.2013 (http://www.innovations-report.de/html/berichte/energie-elektrotechnik/neuer-ansatz-energiesparen-209057.html). Additionally, one article commented on in Facebook (https://www.facebook.com/sonnenseite/posts/328334740620603).
(h) News article: ‘Strom aus Abfallwärme’ (Power from waste heat) covered by INGENIEUR.de 11.01.2016 (http://www.ingenieur.de/Fachbereiche/Umwelt-Recyclingtechnik/Strom-Abfallwaerme).
(i) News article: ‘Organische Halbleiter: Thermogeneratoren aus Molekülkristallen’ (Organic semiconductors: thermogenerators from molecular crystals) covered by elektroniknet.de 27.02.2013 (http://www.elektroniknet.de/power/energy-harvesting/artikel/95374/).
(j) 10th International Conference on Functional Materials and Nanotechnologies (FM&NT 2015) (link), 05-08.10.2015 Vilnius, Lithuania, Book of Abstract p.78 (http://www.fmnt.ff.vu.lt/wp-content/uploads/2015/10/konf_foto.jpg).
List of Websites:
Additional information on the h2esot program can be found at: www.h2esot.com
The partners were:
Simon Woodward, School of Chemistry, University of Nottingham, United Kingdom (Synthetic Chemistry and Coordinator)
http://www.nottingham.ac.uk/~pczsw/SWGroup/
Jens Pflaum, Experimentelle Physik 6, University of Wuezburg, Germany (Experimental Physics)
http://www.physik.uni-wuerzburg.de/ep6pflaum/
Vladimir Dimitrov, IOCCP, Bulgarian Academy of Sciences, Bulgaria (Synthetic Chemistry)
http://www.orgchm.bas.bg/list/oss/index.html
Martins Rutkis, Laboratory of Organic Materials, Institute of Solid State Physics, University of Latvia, Latvia (Thin Film Physics)
http://www.cfi.lu.lv/eng/about-issp/staff/persona/125/
Kevin Simpson, Technical Director, European Thermodynamics Ltd (Thermo-electrical Engineering)
http://www.europeanthermodynamics.com/about/people
Anatolie Casian, Technical University of Moldova, Chisinau, Moldova (Theoretical Physics)
http://www.utm.md/