thermoelectricity and other effects in rare earths
play

Thermoelectricity and other effects in Rare Earths compounds with - PowerPoint PPT Presentation

Thermoelectricity and other effects in Rare Earths compounds with high magnetic disorder. Jos C. Gmez Sal Universidad de Cantabria. Santander. Spain IICAM and ECOM Workshop on correlated thermoelectric materials. Hvar. 26 September 2005


  1. Thermoelectricity and other effects in Rare Earths compounds with high magnetic disorder. José C. Gómez Sal Universidad de Cantabria. Santander. Spain IICAM and ECOM Workshop on correlated thermoelectric materials. Hvar. 26 September 2005

  2. Analysis of thermoelectric power for the study of different problems . Origin of the changes from ferro-to antiferromagnetism in Rare-Earth compounds. The role of the density of states. (RENi 1-x Cu x – REPt 1-x Cu x ) Thermoelectricity and Kondo ferromagnets CePt 1-x Ni x . The problem of highly magnetically disordered compounds. May be an open problem for the thermoelecricity CeNi 1-x Cu x

  3. Origin of the changes from ferro-to antiferromagnetism in Rare-Earth compounds. The role of the density of states. (RENi 1-x Cu x – REPt 1-x Cu x ) • The starting point was the wide phenomenology of compounds changing from Ferro to Antiferro, with the same crystalline structure. In particular 4f -3d metals, with only Rare Earth magnetic atoms. • Pseudobinary RE-3d with RKKY interactions. At that time (75-80) the change was supposed to be related to the distances between magnetic ions. • Single ground state magnetism. Example Orthorhombic. Pnma pseudobinary compounds . Inconmensurate magnetic structures, stable at very low temperatures depending the Kramers or non Kramers character of the RE ion. D. Gignoux and J.C. Gómez Sal "Temperature dependence of the incommensurate magnetic structure of ErNi0.6Cu0.4 compound“. Physics Letters, 50 A, 1, 63-64 (1974) . "Magnetic properties and structures of the rare earth compounds RNi1-xCux with FeB type structure" J. of Magn. Magn. Mat., 1, 203-213 (1976)

  4. A. Hernando, J.M. Rojo, J.C. Gómez Sal & J.M. Novo, J. Appl. Phys. 79, 4815 (1996) RKKY Perturbative, discrete calculation 2 nd order 1 st order Free electrons polarization term static term FM ü Spin redistribution ü Spin polarization ü Positive ü Negative baseline E AFM - E FM E AFM - E FM E AFM - E FM E AFM - E FM ∝ n(E F ) AFM .- A.Hernándo , J.M.Rojo, J.C.Gómez Sal and J.M.Barandiarán. “Density of states and indirect exchange in metallic systems”. Acta Phys.Polonica A, 90, 1227-1234, (1997) A.Hernándo , J.M.Barandiarán, J.M.Rojo and J.C.Gómez Sal. “About the effect of pressure and volume expansion on the transition from antiferromagnetic to ferromagnetic state in some metal alloys: a simplified view”.J.Magn.Magn.Matter, 174, 181-184, (1997)

  5. � A. Señas, J. Rodríguez Fernández, J. C. Gómez Sal , J. Campo.and J.Rodríguez-Carvajal. “From ferromagnetism to inconmensurate magnetic structures: a neutron diffraction study of the chemical substitution effects in TbPt1-xCux .” Phys.Rev.B 70, 184425, (2004) � J.Garcia Soldevilla, J.Blanco, J.Rodriguez Fernandez, J.Espeso, J.C.Gomez Sal, M.T.Fernandez Diaz, J.Rodriguez Carvajal and D.Paccard. “ Complex magnetic ordering in Nd Ni1-x Cux: magnetic structures”. Phys.Rev.B, 70, 224411, (2004) � � RKKY objectives band volume Origin of magnetism in RE Chemical substitution Pressure üdiscrete tuning ücontinuous tuning üsubstitution of ions increase or üvolume reduction decrease the number of conduction üno introduction of impurities electrons

  6. Δ V, Δ e - 28 29 63.54 58.71 Periodic table of [Ar] 3d 8 4s 2 [Ar] 3d 10 4s 1 Cu Ni the elements FM AFM FeB → CsCl CrB → FeB 46 47 106.4 107.87 [Kr] 3d 10 [Kr] 4d 10 5s 1 Pd Ag FM AFM CrB → FeB CsCl ü Gd 3+ : huge neutronic absorption 195.09 196.97 79 78 [Xe] 4f 14 5d 10 6s 1 [Xe] 4f 14 5d 9 6s 1 ü Tb 3+ : high magnetic moment Pt Au AFM FM ü TbNi 1-x Cu x and TbPt 1-x Cu x CrB → FeB FeB → CsCl Δ V

  7. electronic effects ñ In both series FM-AFM change without or with volume change, respectively ñAFM occurs in TbNi 1-x Cu x for highest cell volumes ñ XPS shows that n(E F ) diminishes as Cu increases in TbPt 1-x Cu x

  8. TbPt 1-x Cu x magnetic structures annealed D1B ILL & G4.1 LLB TbPt TbCu 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 z,c z,c x T � Q � type y,b µ B Tb 0 <56 0 -CxFz 7.2 x,a AM (x,y,z) Pt 0.2 <47 0 -CxFz 8.2 x T Q type µ B x,a y,b -CxFz 0.4 <36 0, 0.23, 0 AMxyz 9.0 “limit compound” 0.6 <37 0.15, 0, 0.22 AMxyz 8.9 z T 21-39 K AM k ||y (x,y,z) 129º-130.2º ≤ 30 K –CxFz x

  9. TbPt/Ni 1-x Cu x under pressure volume effects .LLB I.Goncharenko TbPt TbCu 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 PROBLEMS: TbNi üTexture üPreferred orientations üInhomogeneities üFocalisation üInstrument z,c +26º x,a In the -CxFz Ferromagnetic moments rotation towards AFM x -axis

  10. ñ Detailed sample characterisation. ñ Complex analysis due to signal to noise ratio. ñ Pnma seems stable up to 65 kbar . Soft changes induced by hydrostatic pressures ñPressure favours AFM ñNi-series in spite of V increase, AFM appears with increasing Cu-e- ñPt-series no V changes but AFM also appears with increasing Cu-e- ñElectronic rather than volume effects control the sign of the RKKY interaction ñXPS indicates n(E F ) as the key parameter

  11. � Thermopower measurements Institut de Cienciès de Materials de Barcelona ICMB. J.Fontcuberta (77-300 K) by using a differential method. RNi 1-x Cu x (R=Gd and Nd) and GdCu 1-x Pt x . obtain density of states information from transport properties, and thermopower in particular, is not an easy task, we can argue that at temperatures well above the magnetic ordering temperature, only the diffusive and phonon contributions of S(T) should be present . Furthermore a linear dependence of S(T) is typically found in those systems in which the diffusive term dominates. In this case, we can write • S(T)= - ( π 2 k B 2 T / 3e) {dlnl(E)/dE + dlnA(E)/dE} E=EF l is the mean free path of conduction electrons, A(E) the area in k-space of the energy surface having energy E [proportional to N(E)] generally for d and f metals, dlnl(E)/dE < dlnA(E)/dE the slope dS(T)/dT in the linear part of the S(T) plot depends on {N (E) -1 . dN(E)/dE} E=EF . Considering a general shape N(E) = E p , and that the value of dN(E)/dE is not expected to approximate proportionality between the (dS/dT) slope and N(E F ) -1 for the same series compounds .

  12. Experimental results dS/dT in all the samples studied is positive , N(E)/dE has to be negative essential features of the high temperature thermopower can be described on the basis of hole like conduction GdNi1-xCux and NdNi1-xCux series linear range (150 - 300 K), inverse slopes of the S(T) curves can be derived GdPt1-xCux , no distinct behaviour can be clearly deduced from the S(T) curves, within the experimental accuracy A general comment should be stressed, FM GdPt0.8Cu0.4 compound shows smaller slopes in the whole temperature range than those of the GdPt0.4Cu0.6 and GdPt0.2Cu0.8 AF compounds.

  13. � � � • Compound Crystalline Magnetic 1/(dS/dT) • Structure � Character � K 2 µ V -1 • GdAg* FeB AF 12.8 • GdNi CrB FM 124.3 • GdNi0.7Cu0.3 FeB FM 59.6 • GdNi0.4Cu0.6 FeB AF 36.8 • NdNi0.6Cu0.4 FeB FM 222.7 • NdNi0.3Cu0.7 FeB FM 154.2 • NdNi0.2Cu0.8 FeB FM 294.1 • NdCu FeB AF 20.0 • GdNi0.8Co0.2** CrB FM 93.7 Other systems in which the rule of distance seems to fail: for example GdM (M= Cu, Ag, Au) compounds are AF for Cu and Ag but FM for Au whereas the lattice constant of GdAu is intermediate between the other two. Only indications are obtained from Thermopower, surely, the temperature range was not the appropriate and in any case is very difficult to obtain information about density of states due to the many effects involved However the complete study (neutron diffraction, XPS) is conclusive about the preponderant role of electronic effects in this problem .

  14. � � CePt 1-x Ni x s k RKKY-Kondo competition : J K J i J i +1 DONIACH PHASE DIAGRAM J i Long-range RKKY Hybridisation localised 4f–conduction electron T K ∝ exp(-1/J kf N(E F )) magnetic interaction T H Kondo = -J K · s · Ji NON MAGNETIC GROUND MAGN. STATE GROUND T RkkY ∝ J Kf 2 n(E F ) STATE J K n(E F ) cr T C → 0 K HEAVY T C FERMION …….. Anomalous behavior: NFL at low T FL - CeCu 6-x Au x - (YU)Pd 3 -UCu 5-x Pd x J k n(E F ) and many more…… J k n(E F ) cr Interest of f-intermetallic compounds with competing interactions J K n(E F ) experimentally tuned by -Disorder effects - Applied pressure -Electronic effects - Chemical pressure : Chemical Substitutional systems →

  15. The case of CePt 1-x Ni x D. Gignoux, J. C. Gómez Sal, Phys. Rev. B 30, 3967 (1984) J. A. Blanco, M. de Podesta, J. I. Espeso, J. C. Gómez Sal, C. Lester, K. A. McEwen, N. Patrikios and J. Rodriguez. Fernandez, Phys. Rev. B 15126 (1994) CeNi is intermediate valence, CePt is one of the few Kondo-ferromagnet. Tc = 6K. It has been demonstrated that the Ce3+ state is increasingly screened by Kondo effect with the rise of Ni concentration, x, leading to a gradually decrease of the magnetic moment. The disappearance of the magnetic order occurs at x = 0.95. These features are due to the increasing importance of the 4f- conduction band hybridization respect to the magnetic RKKY interactions and are interpreted on the ground of the diagram proposed by Doniach .

Recommend


More recommend