Image: TU Vienna

Ther­mo­electrics: From heat to electricity

Researchers at TU Wien have devel­oped a new con­cept to con­vert ther­mal ener­gy into elec­tri­cal ener­gy more effi­cient­ly. Ener­gy loss­es can thus be minimized.

A lot of heat is lost dur­ing the con­ver­sion of ener­gy. Accord­ing to esti­mates, even more than 70%. How­ev­er, in ther­mo­elec­tric mate­ri­als, such as those stud­ied at the Insti­tute of Sol­id State Physics at the Vien­na Uni­ver­si­ty of Tech­nol­o­gy, heat can be con­vert­ed direct­ly into elec­tri­cal ener­gy. This effect (See­beck effect) can be used in numer­ous appli­ca­tions in indus­try but also in every­day life.

Recent­ly, Ernst Bauer’s research team made an excit­ing dis­cov­ery in a ther­mo­elec­tric con­sist­ing of iron, vana­di­um and alu­minum (Fe2VAl). The researchers recent­ly pub­lished their results in the renowned jour­nal “Nature Communications”.

The ide­al thermoelectric

To achieve the great­est pos­si­ble ener­gy con­ver­sion effect, researchers are look­ing for mate­ri­als that ful­fill a num­ber of prop­er­ties: They should have a large See­beck effect, high elec­tri­cal con­duc­tiv­i­ty and low ther­mal con­duc­tiv­i­ty. How­ev­er, this is extreme­ly dif­fi­cult because the prop­er­ties are inter­re­lat­ed and inter­de­pen­dent. The researchers there­fore asked them­selves what a mate­r­i­al would have to look like phys­i­cal­ly in order to ful­fill all these con­di­tions in the best pos­si­ble way.

Thus, physi­cists at the Vien­na Uni­ver­si­ty of Tech­nol­o­gy have suc­ceed­ed in find­ing a new con­cept to resolve this con­tra­dic­tion and opti­mize all ther­mo­elec­tric prop­er­ties in one mate­r­i­al at the same time. “At the so-called Ander­son tran­si­tion, a quan­tum phase tran­si­tion from local­ized to mobile elec­tron states, the con­di­tions for the ide­al ther­mo­elec­tric are giv­en. This means that all con­duc­tive elec­trons have approx­i­mate­ly the same ener­gy,” reports Fabi­an Garm­rou­di, first author of the study.

The Ander­son tran­si­tion occurs when impu­ri­ty atoms are added to a semi­con­duc­tor, which strong­ly bind their elec­trons. “Anal­o­gous to ice floes in the sea, these are ini­tial­ly iso­lat­ed from each oth­er and can­not be walked on. How­ev­er, if the num­ber of ice floes is large enough, you have a con­tin­u­ous con­nec­tion across which you can cross the sea,” Fabi­an Garm­rou­di draws a com­par­i­son. This hap­pens in a sim­i­lar way in the sol­id state: if the num­ber of impu­ri­ty atoms exceeds a crit­i­cal val­ue, the elec­trons can sud­den­ly move freely from one atom to anoth­er and cur­rent can flow.

Atoms exchange places when things get hot

The Ander­son tran­si­tion was demon­strat­ed in close col­lab­o­ra­tion with researchers from Swe­den and Japan as well as the Uni­ver­si­ty of Vien­na, and for the first time was asso­ci­at­ed with a sig­nif­i­cant change in ther­mo­elec­tric prop­er­ties. The team made the excit­ing dis­cov­ery when they heat­ed the mate­r­i­al to very high tem­per­a­tures, close to its melt­ing point.

“At high tem­per­a­tures, the atoms vibrate so strong­ly that they occa­sion­al­ly swap their lat­tice posi­tions. For exam­ple, iron atoms are then locat­ed where vana­di­um atoms were before. We suc­ceed­ed in freez­ing this ‘atom­ic con­fu­sion’, which occurs at high tem­per­a­tures, by so-called ‘quench­ing’, i.e. rapid cool­ing in a water bath,” reports Ernst Bauer. These irreg­u­lar defects serve exact­ly the same pur­pose as the for­eign atoms men­tioned ear­li­er, with­out the need to change the chem­i­cal com­po­si­tion of the mate­r­i­al to do so.

Ener­gy con­ver­sion thanks to disorder

In many research areas of sol­id-state physics, one is inter­est­ed in mate­ri­als that are as pure as pos­si­ble and have an ide­al crys­tal struc­ture. The rea­son: The reg­u­lar­i­ty of the atoms sim­pli­fies a the­o­ret­i­cal descrip­tion of the phys­i­cal prop­er­ties. How­ev­er, in the case of Fe2VAl, it is the imper­fec­tions (the irreg­u­lar­i­ties) that account for most of the ther­mo­elec­tric per­for­mance. It has also already been shown in neigh­bor­ing dis­ci­plines that irreg­u­lar­i­ties can be advan­ta­geous: “Basic research on quan­tum mate­ri­als is a good exam­ple of this. There, sci­ence has already been able to show that dis­or­der is often the nec­es­sary spice in the ‘quan­tum soup’,” says Andrei Pus­to­gov, one of the co-authors, and reports: “Now this con­cept has also arrived in applied sol­id-state research.”