TESIL (completed)

• Shyikira, Antoine de Padoue; Akhtar, Naureen; Skomedal, Gunstein; Sætre, Tor Oskar & Middleton, Peter Hugh (2021). High temperature oxidation of higher manganese silicides. Corrosion Science. ISSN 0010-938X. 185. doi: 10.1016/j.corsci.2021.109327. Full text in Research Archive
• Graff, Joachim Seland; Schuler, Raphael; Song, Xin; Castillo-Hernandez, Gustavo; Skomedal, Gunstein & Enebakk, Erik [Show all 11 contributors for this article] (2021). Fabrication of a Silicide Thermoelectric Module Employing Fractional Factorial Design Principles. Journal of Electronic Materials. ISSN 0361-5235. 50, p. 4041–4049. doi: 10.1007/s11664-021-08902-y. Full text in Research Archive Show summary

  Thermoelectric modules can be used in waste heat harvesting, sensing, and cooling applications. Here, we report on the fabrication and performance of a four-leg module based on abundant silicide materials. While previously optimized Mg2Si0.3Sn0.675Bi0.025 is used as the n-type leg, we employ a fractional factorial design based on the Taguchi methods mapping out a four-dimensional parameter space among Mnx-εMoεSi1.75−δGeδ higher manganese silicide compositions for the p-type material. The module is assembled using a scalable fabrication process, using a Cu metallization layer and a Pb-based soldering paste. The maximum power output density of 53 μW cm–2 is achieved at a hot-side temperature of 250 °C and a temperature difference of 100 °C. This low thermoelectric output is related to the high contact resistance between the thermoelectric materials and the metallic contacts, underlining the importance of improved metallization schemes for thermoelectric module assembly.

• Løvvik, Ole Martin; Flage-Larsen, Espen & Skomedal, Gunstein (2020). Screening of thermoelectric silicides with atomistic transport calculations. Journal of Applied Physics. ISSN 0021-8979. 128. doi: 10.1063/5.0008198. Full text in Research Archive Show summary

  More than 1000 crystalline silicide materials have been screened for thermoelectric properties using first-principles atomistic calculations coupled with the semi-classical Boltzmann transport equation. Compounds that contain radioactive, toxic, rare, and expensive elements as well as oxides, hydrides, carbides, nitrides, and halides have been neglected in the study. The already well-known silicides with good thermoelectric properties, such as SiGe, Mg2Si, and MnSix, are successfully predicted to be promising compounds along with a number of other binary and ternary silicide compositions. Some of these materials have only been scarcely studied in the literature, with no thermoelectric properties being reported in experimental papers. These novel materials can be very interesting for thermoelectric applications provided that they can be heavily doped to give a sufficiently high charge carrier concentration and that they can be alloyed with isoelectronic elements to achieve adequately low phonon thermal conductivity. The study concludes with a list of the most promising silicide compounds that are recommended for further experimental and theoretical investigations.

• Zhang, H.; Hippalgaonkar, K.; Buonassisi, T.; Løvvik, Ole Martin; Sagvolden, Espen & Ding, D. (2019). Machine Learning for Novel Thermal-Materials Discovery: Early Successes, Opportunities, and Challenges. Engineered Science. ISSN 2576-988X. 2, p. 1–8. doi: 10.30919/esee8c209.
• Tabib, Mandar; Løvvik, Ole Martin; Johannessen, Kjetil Andre; Rasheed, Adil; Sagvolden, Espen & Rustad, Anne Marthine (2018). Discovering Thermoelectric Materials Using Machine Learning: Insights and Challenges. Lecture Notes in Computer Science (LNCS). ISSN 0302-9743. 11139 LNCS, p. 392–401. doi: 10.1007/978-3-030-01418-6_39. Full text in Research Archive Show summary

  This work involves the use of combined forces of data-driven machine learning models and high fidelity density functional theory for the identification of new potential thermoelectric materials. The traditional method of thermoelectric material discovery from an almost limitless search space of chemical compounds involves expensive and time consuming experiments. In the current work, the density functional theory (DFT) simulations are used to compute the descriptors (features) and thermoelectric characteristics (labels) of a set of compounds. The DFT simulations are computationally very expensive and hence the database is not very exhaustive. With an anticipation that the important features can be learned by machine learning (ML) from the limited database and the knowledge could be used to predict the behavior of any new compound, the current work adds knowledge related to (a) understanding the impact of selection of influence of training/test data, (b) influence of complexity of ML algorithms, and (c) computational efficiency of combined DFT-ML methodology.

• Løvvik, Ole Martin & Berland, Kristian (2018). Predicting the thermoelectric figure-of-merit from first principles. Materials Today: Proceedings. ISSN 2214-7853. 5(4), p. 10227–10234. doi: 10.1016/j.matpr.2017.12.269.

View all works in Cristin

• Løvvik, Ole Martin (2021). Casting Materials Acceleration Platform and the accelerated search for thermoelectric materials.
• Løvvik, Ole Martin (2021). Screening of materials for thermoelectric conversion of heat into electricity.
• Løvvik, Ole Martin (2020). Elektronikk i varme omgivelser og termoelektriske generatorer. [Radio]. Intervju i Abels tårn, NRK P2.
• Løvvik, Ole Martin; Graff, Joachim Seland; Schrade, Matthias; Sunding, Martin Fleissner; Wright, Daniel Nilsen & Song, Xin [Show all 10 contributors for this article] (2020). Silicide module produced in the TESil project.
• Løvvik, Ole Martin (2019). Klimasending fra Eidsvolls plass om bærekraft og fornybar energi. [Radio]. Intervju i Abels tårn, NRK2.
• Løvvik, Ole Martin; Flage-Larsen, Espen & Skomedal, Gunstein (2019). Atomistic simulations and machine learning methods for development of thermoelectric materials.
• Shyikira, Antoine de Padoue; Skomedal, Gunstein & Middleton, Peter Hugh (2019). Performance and stability of silicide based thermoelectric modules. Show summary

  Long-term studies on thermoelectric generator uni-couple based on N-type magnesium silicide (Mg2.01Si0.49Sn0.5Sb0.01) and P-type higher manganese silicide (Mn0.98Mo0.02Si1.73Ge0.02) materials are presented, in the operating temperature range of 2000C-4000C. Emphasis is put on the performance and reliability of the current collector configuration, especially, on the hot side of the module and on the thermomechanical stresses that are created during operation and lifetime testing, as a result of large temperature gradients experienced across the thermoelectric legs. With silver (Ag) paste as contact material, the long term-stability of the uni-couples were carried out on non-metalized legs and gold metalized legs under ambient conditions. Under isothermal and thermocycling tests, the non-metalized legs showed a gradual decrease in open circuit voltage (after a period of 200hours) and increase in internal resistance. In contrary, the metalized leg-based module was robust and stable for the same isothermal period, however, after 300cycles the n-type material showed mechanical failure (cracks) but the p-type resisted. Post-operation analysis by SEM/EDS and mechanical testing revealed that oxidation, adherence of the contact material and diffusion of the bonding material were the cause of performance degradation of the unicouples.

• Løvvik, Ole Martin (2019). Nye løsninger for å lage strøm av spillvarme kan gi muligheter for norsk industri. Kommentarer i artikkelen. Teknisk Ukeblad. ISSN 0040-2354.
• Løvvik, Ole Martin; Berland, Kristian; Remonato, Filippo; Sagvolden, Espen; Flage-Larsen, Espen & Schrade, Matthias [Show all 13 contributors for this article] (2019). Screening thermoelectric materials with ab initio atomistic modelling and machine learning techniques.
• Løvvik, Ole Martin (2018). Thermoelectric materials for energy harvesting – new modelling tools with predictive power.
• Løvvik, Ole Martin; Flage-Larsen, Espen & Skomedal, Gunstein (2018). Screening silicide thermoelectric materials using ab initio transport calculations.
• Løvvik, Ole Martin (2018). Følger av Trumps forhold til forskning. [Radio]. Intervju i Ekko, NRK P2.
• Løvvik, Ole Martin (2018). Trump ut av storsatsing på fornybar energi. [Radio]. Intervju i Ekko, NRK P2.
• Løvvik, Ole Martin (2018). Machine learning of energy materials.
• Graff, Joachim Seland (2018). Next Generation Thermoelectric Materials based on Silicon.
• Løvvik, Ole Martin (2018). Thermoelectric materials for energy harvesting – high-throughput screening using new tools.
• Shyikira, Antoine de Padoue; Skomedal, Gunstein; Middleton, Hugh P. & Sætre, Tor Oskar (2018). High Temperature Oxidation of Higher Manganese silicide and Alloys.
• Løvvik, Ole Martin (2018). Moderne metoder innen modellering og maskinlæring kan gi nye løsninger for energieffektivisering og utnyttelse av spillvarme.
• Riis, Henrik; Song, Xin; Prytz, Øystein & Finstad, Terje (2018). Characterization of Cu/ZnSb interfaces.
• Song, Xin; Graff, Joachim Seland; Götz, Adam; Skomedal, Gunstein & Finstad, Terje (2018). Investigating the influence of synthesis approach on the microstructure and thermoelectric properties for HMS.
• Løvvik, Ole Martin (2017). Hva skjer hvis du setter et kjøleskap i fryseren? [Radio]. Intervju i Ekko, NRK P2.
• Løvvik, Ole Martin (2017). Californias miljøkamp inspirerer. [Radio]. Intervju i Ekko, NRK P2.
• Løvvik, Ole Martin; Shulumba, Nina; Hellman, Olle; Persson, Clas & Berland, Kristian (2017). Predicted figure-of-merit of half-Heusler alloys - importance of scattering mechanisms.

View all works in Cristin