Thermoelectric properties of Ag 2 Sb 2 Ge 46 − x Dy x Te 50 alloys with high power factor

2016

Phys. Rev. B

Self Cite

Thermoelectric materials utilize the Seebeck effect to convert thermal to electrical energy. The Seebeck coefficient (thermopower), S, depends on the free (mobile) carrier concentration, n, and effective mass, m*, as In some materials the influence of m* on the Seebeck coefficient can be significant and its determination is important for better understanding of thermoelectric materials. 125 Te nuclear magnetic resonance (NMR) [4][5][6][7][8][9] can be used to study complex tellurides and derive the carrier concentration from spin-lattice relaxation time, T 1 . In a theoretical model by Selbach et al. [10], the coefficient of proportionality between T 1 and n, "Bloembergen constant", C B , depends on m*. Hence, if the effective mass in a series of materials is not constant, it can affect C B , and, therefore, the carrier concentration deriving from 125 Te NMR T 1 measurements should be different compared to that obtained if m* is constant.Here it is shown that the combination of the (Table 1) were synthesized by a direct reaction of the constituent elements in 10 mm diameter fused silica ampoules in argon backfilled up to ~17 kPa pressure atmosphere. For synthesis we used Ge (Materion G-1038, 99.999% purity), and Te (Alfa Aesar, 99.9999% purity, product #12607, lot 004). Ampoules with the constituent elements were heated up to 1323 K to melt the constituents, and after ~4 hours the melt was cooled down with the furnace at the rate of 100 K/hour. Each ingot had a mass of 20 g, diameter of 10 mm, length of ~40 mm. Analysis of each material was performed by X-ray diffraction, scanning electron microscopy, and energy dispersive spectroscopy, and showed that all materials are nearly single phase; only ~1 at.% Ge is present in the samples matrix due to naturally occurring process of Ge rejection from the melt [8,14]. 125 Te NMR experiments were performed at 126 MHz using a Bruker 400WB spectrometer with TOPSPIN software in a magnetic field of 9.4 T without sample spinning 4 (static regime). Powder samples were prepared from ingots and placed in 4-mm diameter insert. 125 Te NMR spin-lattice relaxation measurements were used to obtain the spin-lattice relaxation time, T 1 , using GeTe as a reference material [5,8]; the uncertainty of T 1 measurements is less than ±0.2 ms [5].More experimental details including pulse sequence for 125 Te NMR can be found in Refs. 5,8,9,14. Samples for the Seebeck coefficient measurements were prepared by cutting the ingot with a diamond saw [9]. The Seebeck coefficient was measured by a LSR-3 measuring system (Linseis Inc.) in a helium atmosphere relative to the Pt legs of a Pt-(Pt+Rh) thermocouple and then the absolute Seebeck coefficient was calculated; the uncertainty of measurements is about ±3% [9]. Table 1). 125 Te NMR spectra for these two materials are shown on inset in Fig. 1(a), demonstrating that signals can be easily detected; spectra for the rest of materials also are located between those for GeTe and TAGS-85.Relaxation curves shown on Fig. 1(a) as well as for ...

“…1(a) as well as for the rest of materials can be fit by one component [5,14], i.e., GeTe and TAGS-85 and other materials studied here are electronically homogeneous [4,5,14].…”

Section: Introductionmentioning

confidence: 89%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Charge carrier effective mass and concentration derived from combination of Seebeck coefficient andTe125NMR measurements in complex tellurides

2016

Phys. Rev. B

Self Cite

“…Recycle delay time varies from 0.1 ms to 150 ms and measuring times for each samples were about 22 h. 125 Te NMR chemical shifts were referenced to Te(OH) 6 in solution and chemical shifts relative to (CH 3 ) 2 Te in benzene were larger by +712 ppm [20]. 125 Te NMR spin-lattice relaxation measurements were conducted in order to obtain the values of spin-lattice relaxation time, T 1 , and estimate free (mobile) charge carrier concentration in tellurides using the approach reported for 207 Pb NMR [21] and for 125 Te NMR [19,20]. The spin-lattice relaxation time was determined by fitting the dependence of the signal intensity vs. time delay; the uncertainty of T 1 determination was 0.2 ms [20].…”

Section: Nuclear Magnetic Resonance (Nmr) 125 Te Nmr Experiments Wermentioning

confidence: 99%

Crystal structure, magnetization, 125Te NMR, and Seebeck coefficient of Ge49Te50R1 (R = La, Pr, Gd, Dy, and Yb)

Materials Chemistry and Physics

Cooling

Bud'ko

et al. 2017

GeTe, a self-doping semiconductor, is a well-known base compound for thermoelectric and phase-change materials. It is known, that replacement of Ge in Ag 6.5 Sb 6.5 Ge 37 Te 50 (TAGS-85) material by rare earth Dy significantly enhances both the power factor and thermoelectric figure of merit. Here we demonstrate how replacement of Ge in GeTe by rare earths with different atomic size and localized magnetic moments affect XRD patterns, magnetization, 125 Te NMR spectra and spin-lattice relaxation, and the Seebeck coefficient of the alloys with a nominal composition of Ge 49 Te 50 R 1 (R = La, Pr, Gd, Dy, and Yb). SEM, EDS and WDS data show that rare earth atoms in the matrix are present at smaller extent compared to a nominal composition, whereas rare earth also is present in the inclusions. Rare earths affect the Seebeck coefficient, which is a result of interplay between the reduction due to higher carrier concentration and enhancement due to magnetic contribution. The effect of replacement of Ge in GeTe by Dy on the Seebeck coefficient is smaller than that observed in Ag 6.5 Sb 6.5 Ge 36 Te 50 Dy 1. This can be explained by larger amount of rare earth, which can be embedded into the lattice of materials containing [Ag+Sb] atomic pairs and possible effect from these pairs.

“…Tellurides have been well known for a long time because of their very good thermoelectric properties and they are very attractive as model systems. GeTe-based, Ag x Sb x Ge 50−2x Te 50 materials, 5,6 typically written as (GeTe) m (AgSbTe 2 ) 100−m (T _ellurium-A _ntimony-G _ermanium-S _ilver, acronym TAGSm), [7][8][9][10] represent one of the well-known groups of thermoelectric materials. These materials, including the high-efficiency TAGS-85 (with approximate composition of Ag 6.5 Sb 6.5 Ge 37 Te 50 ) were used in various applications, 7 but the origin of their high efficiency has never been adequately explained.…”

Section: Introductionmentioning

confidence: 99%

Role of chemically and thermally induced crystal lattice distortion in enhancing the Seebeck coefficient in complex tellurides

Kramer

2016

CrystEngComm

Self Cite

Composition and crystal structure of complex materials can significantly change the Seebeck effect, i.e., heat to electrical energy conversion, which is utilized in thermoelectric materials. Despite decades of studies of various thermoelectric materials and their application, the fundamental understanding of this effect still is limited. One of the most efficient groups of thermoelectric materials is based on GeTe, where Ge is replaced by [Ag + Sb], i.e., Ag x Sb x Ge 50−2x Te 50 alloys, traditionally shown as (GeTe) m (AgSbTe 2 ) 100−m (TAGSm series). Here we report on the discovery of two unique phenomena in TAGS materials attributed to the effects from [Ag + Sb] atoms: (i) a linear relation between the Seebeck coefficient and rhombohedral lattice distortion, and (ii) resonance-like temperature-induced behavior of the contribution to the Seebeck coefficient produced by [Ag + Sb] atoms. Our findings show that heat to electrical energy conversion strongly depends on the temperature-and compositionally-induced rhombohedral to cubic transformation where [Ag + Sb] atoms play a crucial mediating role.