- Email: [email protected]
Synthesis of iodine filled CoSb3 with extremely low thermal conductivity
Accepted Manuscript Synthesis of iodine filled CoSb3 with extremely low thermal conductivity Xiaodong Li, Bo Xu, Long Zhang, Fenfen Duan, Xinlin Yan, Jianqing Yang, Yongjun Tian PII: DOI: Reference:
S0925-8388(14)01565-5 http://dx.doi.org/10.1016/j.jallcom.2014.06.198 JALCOM 31614
To appear in:
Journal of Alloys and Compounds
Received Date: Revised Date: Accepted Date:
8 May 2014 20 June 2014 28 June 2014
Please cite this article as: X. Li, B. Xu, L. Zhang, F. Duan, X. Yan, J. Yang, Y. Tian, Synthesis of iodine filled CoSb3 with extremely low thermal conductivity, Journal of Alloys and Compounds (2014), doi: http://dx.doi.org/ 10.1016/j.jallcom.2014.06.198
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Synthesis of iodine filled CoSb3 with extremely low thermal conductivity Xiaodong Lia, Bo Xua, Long Zhanga,*, Fenfen Duana, Xinlin Yanb, Jianqing Yang a, and Yongjun Tiana
a
State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao, Hebei 066004, China
b
Institute of Solid State Physics, Vienna University of Technology, Wiedner Hauptstr. 8-10, 1040 Vienna, Austria
* Author to whom correspondence should be addressed. Electronic mail: [email protected]
Abstract p-type iodine filled CoSb3 skutterudite, which is inaccessible under ambient pressure, was synthesized with high pressure synthesis technique. The successful filling of iodine into voids of CoSb3 crystal structure was verified by combined measurements including X-ray diffraction, Raman spectroscopy, and specific heat. The highest filling fraction of 0.79 was achieved for the sample with a nominal composition of I1.2Co4Sb12. Beneficial from such a high filling fraction of iodine and the strong phonon-1
-1
scattering from iodine fillers, the thermal conductivity reached as low as 0.7 Wm K for I0.79Co4Sb12, which is the lowest value among all elemental filled skutterudites. Introducing iodine fillers is thus a practical way to further suppress thermal conductivity for single- and multiple-filled skutterudites.
Keywords: Thermoelectric materials; High-pressure; Solid state reactionsa; Skutterudite
1. Introduction Thermoelectric (TE) materials attracted extensive attention for environment-friendly power generation and refrigeration applications. TE performance of TE materials is evaluated by the 2
dimensionless figure of merit, ZT = S T/(ρκ), where S, ρ, κ, and T are the Seebeck coefficient, electrical resistivity, thermal conductivity, and temperature, respectively. Filled CoSb3-based filled skutterudite with a chemical formula of RxCo4Sb12 is one of the most successful prototypes of phonon1
glass electron crystal (PGEC) concept proposed by Slack, and is likely the most potential TE material for waste heat recovery (power generation) with enhanced TE properties.2-7 A significant advantage for this class of materials comes from the loosely bonded ionic fillers, which can rattle in the vacancies of skutterudite and scatter phonons strongly to reduce the thermal conductivity. Different filling elements have been explored ranging from heavy to light elements, e.g., rare-earth,
8-10
alkaline
earth,11-14 or alkaline metal.15-17 Moreover, multiple-filling strategy demonstrated further suppression of thermal conductivity since multiple fillers with distinct vibrational frequencies can scatter a broader spectrum of phonons for lower thermal conductivity.
12, 18
Still, the thermal conductivity remains high for
elemental filled CoSb3-based skutterudites and exploration of new filling species is anticipated. Previously, a simple criteria for fillable elemental species was proposed, which claimed the electronegativity difference between Sb and the filling element must be larger than 0.8 for the filled CoSb3 to be stable. Mg,
20
19
Almost all the filling species in CoSb3 reported so far satisfied this criteria except
which has a electronegativity difference smaller than 0.8 and can be filled into CoSb3 voids
under high pressure. Recently, Fukuoka et al. synthesized I0.9Rh4Sb12 using high pressure synthesis (HPS) technique.
21
Compared with other filling species, iodine atoms were revealed in an anionic state
-
(I ) with a longer I−Sb separation. Such a longer separation would lead to a looser bond, and consequentially a lower thermal conductivity. Obviously, pressure played a vital role for Mg and I filling. High pressure can lower the reaction temperature, facilitate the synthesis of metastable phase, and acquire new compound inaccessible under ambient pressure by shifting reaction equilibrium.22 Our previous works highlighted high pressure as a fundamental thermodynamic variable to synthesize elemental (e.g. Li,
17, 23
Gd,
24
20
and Mg ) filled CoSb3, which cannot be filled under ambient
pressure. It is thus highly motivating to synthesize iodine filled CoSb3 under high pressure, and to investigate how anionic fillers such as I- would affect the thermoelectric properties. Here we report the
2
successful synthesis of I-filled CoSb3 by using HPS technique. Iodine shows a high filling fraction in CoSb3, resulting in strong phonon scattering and remarkably suppressed thermal conductivity. 2. Experimental procedures Pure I2 (99.8%), Co (99.8%), and Sb (99.999%) powder were mixed according to a nominal atomic ratio of yn:4:12 (yn = 0.4, 0.8, and 1.2). The mixture was loaded into a steel mould, shaped with a cold press method, inserted into an h-BN crucible, and loaded into a high pressure apparatus for HPS. The first step of HPS was carried out at 1073 K and 5 GPa for 0.5 h. The resulting ingot was ground into powder under argon atmosphere, shaped with cold press, and loaded into the high pressure apparatus again for the second step of HPS (753 K and 5 GPa for 3 h). The obtained product was ground into powder and washed in distilled water to remove the excessive I2. The final powders were sintered at 5 GPa and 573 K for 0.5 h into dense pellet and cut for the transport property measurements. The processes for structural optimization and total energy calculation of ICo4Sb12 can be found elsewhere.20 X-ray diffraction (XRD) measurements were carried out with a Rigaku D/MAX/2500/PC (Cu Kα). Rietveld refinements were performed using the FULLPROF program to determine the filling fraction (FF) and lattice parameter.
25
Electrical resistivity and Seebeck coefficient were measured with
ZEM-3 (Ulvac-Riko), and thermal conductivity was measured with TC-7000H (Ulvac-Riko). The densities were measured using the Archimedes method in ethanol, in which the repeatability was 99.5%. The heat capacity measurements were carried out with the heat capacity option of the Physical Property Measurement System. Raman scattering measurements were performed by using a Renishaw inVia system at room temperature. 3. Results and discussion Figure 1 shows the XRD patterns of the final products with the characteristic peaks indexed. All samples were dominated with a skutterudite structure of Im 3 symmetry. The vanishing of (211) peak for the samples with higher iodine content verifies the successful filling of iodine atoms into skutterudite cages.26 The formation enthalpy of ICo4Sb12 (defined as ∆H = EICo4Sb12－E I－4ECo－12ESb, inset in Fig. 1) decreases almost linearly with increasing pressure, indicating a pressure-driven reaction. This is the origin we performed the synthesis experiments at 5 GPa pressure. The diffraction peaks (e.g., 2θ = 59° marked with a dashed line in Fig. 1) shift to smaller angle with increasing iodine content, indicating the expansion of the unit cell after iodine filling. In Table 1, we list room
3
temperature structural parameters of the skutterudite phases from Rietveld refinement. The FFs for yn = 0.4, 0.8, and 1.2 are 0.13, 0.72, and 0.79, respectively, indicating a saturated FF of iodine near 0.8, which is the highest FF reported in filled CoSb3 to date. We noted the lattice parameter increases significantly with increasing iodine content and the values over 9.1211(1) Å are the biggest one for filled CoSb3. This large lattice parameter obviously is connected to the large size of the filling anions. Also listed in Table 1 are the Seebeck coefficient and electrical resistivity for all samples at room temperature. The positive Seebeck coefficient indicates p-type charge carrier of hole in the systems due to the formation of iodine anion (electron acceptor). Increasing the FF of iodine can increase hole concentration. As a result, both the Seebeck coefficient and electrical resistivity should decrease with increasing FF of iodine. However, the Seebeck coefficient is almost constant with variable FFs. To clarify this issue, we calculated the band structures of CoSb3 and fully filled ICo4Sb12 along with the corresponding density of states (DOS) near the Fermi energy, as shown in Fig. 2. The calculated electronic structures of ICo4Sb12 are very similar to that of CoSb3. Iodine filling shows limited influence on the DOS of CoSb3 except for moving the Fermi level downward into the valence band and steepening the DOS near the Fermi level. In the Energy range of 0 to -0.2 eV, the plateau of DOS is 43 electrons/eV for ICo4Sb12 (Fig. 2b), while only 32 electrons/eV for CoSb3 (Fig. 2d). The steeper profile of DOS can enhance the Seebeck coefficient, and thereby compensates the negative effect induced by increasing hole concentration at higher FF of iodine. Therefore, the Seebeck coefficient keeps nearly invariable for various compositions. Figure 3 shows the temperature dependent thermal conductivity for IyCo4Sb12 compared with the 15
27
28
29
lowest thermal conductivity in each investigated series for Ay Co4Sb12 (A = K , Ba , Ce , In ). The thermal conductivity follows approximately the 1/T relation for I0.13Co4Sb12, while shows a quadratic temperature-dependence for I0.72Co4Sb12 and I0.79Co4Sb12 arising from the bipolar effect. Over the measurement temperature range, κ are in the range of 1.79−1.44 W/mK for I0.13Co4Sb12, and 1−0.7 W/mK for the samples with higher FFs (y = 0.72 and 0.79). κ reduced dramatically with higher FF of iodine. Moreover, these values are the lowest ones in filled skutterudites to the best of our knowledge. The significant suppression of the thermal conductivity for iodine filled CoSb3 compared with AyCo4Sb12 (A = K, Ba, Ce and In) are obviously demonstrated in Fig. 3. The room temperature lattice thermal conductivity κl versus filling fraction for IyCo4Sb12 is displayed in Fig. 4. κl was derived by subtracting electron contribution κl = LT/ρ from the total thermal
4
-8
2
-2
conductivity, where L = 2.44×10 V K is the Lorenz number and ρ is the electrical resistivity. The 17
lattice thermal conductivity for IyCo4Sb12 again is greatly suppressed compared with those for Li- , K15
27
28, 30
, Ba- , Ce-
31
, Yb- , Tl-
32
and In-filled
29
CoSb3. These facts indicate that the phonons are
scattered much more strongly by iodine fillers than by any other fillers. We speculate that such intensive scattering is related to the very high FF of iodine as well as the loosely bonding of iodine due 21
to the longer I−Sb separation than that in other filled skutterudites . The very low thermal conductivity of iodine filled CoSb3 can be attributed to the high filling fraction of iodine. In addition, after iodine atoms filled into the voids of CoSb3, the large radius of iodine atoms would induce lattice distortions locally, which may further depress the lattice thermal conductivity. To investigate the perturbation to Sb4-rings after iodine filling, Raman spectra were collected as a function of iodine content (Fig. 5). With higher iodine filling fraction, the lower-energy Ag mode is intensively broadened and flattened to overlap with the background, while the higher-energy one almost unchanged. The broadening of Raman peak is mainly due to the distortion of the structure unit. The lower-energy Ag mode is presumably caused by the stretching of the longer Sb−Sb bond and the higher-energy one the shorter 33
Sb−Sb bond . Raman results indicate the deformation of the longer Sb−Sb bond is easier than that of the shorter Sb−Sb bond. It is also noted that the peak of the higher-energy Ag mode shifts to higher position after iodine filling. These results confirm the success of iodine filling and evidence the severe distortion of Sb4-rings by iodine filling. To clarify the existence of localized incoherent vibrations associated with iodine anion in the voids of the host lattice, heat capacity of I0.79Co4Sb12 were measured from 2 to 50 K. Our previous heat capacity data for CoSb3 were used as the reference.
20
The contribution of iodine to the heat capacity
has been determined form the difference in the molar heat capacity of I0.79Co4Sb12 and CoSb3 (Fig. 6). The solid line through the data represents a fit to the equation modeled by the Einstein law: ∆C p = C p , I
0.79 Co4 Sb12
− C p ,Co Sb = γ T + Ax 2 e x 4
12
(e
x
2
−1 ,
)
23
(1) -2
-1
where γT is the electron contribution to the molar heat capacity and, γ = 0.040±0.004 JK mol , A = 18±1 JK-1mol -1 and x = ΘE/T with an Einstein temperature ΘE = 84±1 K. The filling fraction determined from A/3R (R is the ideal gas constant) is 0.72, which is consistent with the Rietveld refined value 0.79. This is the highest filling fraction in filled CoSb3 and benefit the suppression of thermal conductivity. 4. Conclusions
5
In summary, theoretical investigations revealed a pressure-driven reaction for ICo4Sb12. Anion filled p-type IyCo4Sb12 skutterudites, which are inaccessible at ambient pressure, were synthesized by a stepwise synthesis method using HPS technique at 5 GPa. A saturated filling fraction of 0.79 was evaluated from Rietveld refinement and further confirmed by the heat capacity measurement, which is the highest filling fraction ever reported in elemental filled CoSb3 materials. Beneficial from such a high filling fraction, the lowest thermal conductivity is achieved among all skutterudites. Interestingly, even I0.13Co4Sb12 with a relatively low filling fraction shows a very small thermal conductivity compared with other elemental filled CoSb3, indicating I-filling strongly scatters phonons and thus significantly suppresses thermal conductivity. The investigations of incorporation of iodine fillers in multiple-filled skutterudite are ongoing. Acknowledgments This work was supported by the National Science Foundation of China (51201149, 51121061, 51072175), the Natural Science Foundation for Distinguished Young Scholars of Hebei Province of China (E2014203150), and the Key Basic Research Project of Hebei (14961013D).
6
References 1. G.A. Slack, in: D.M. Rowe (Ed.), CRC Handbook of Thermoelectrics, CRC Press, Boca Raton, FL, 1995, pp. 407−440. 2. C. Uher, Semiconduct. Semimet. 69 (2001) 139−253. 3. G.J. Snyder, E.S. Toberer, Nature Mater. 7 (2008) 105−114. 4. J.F. Li, W.S. Liu, L.D. Zhao, M. Zhou, NPG Asia Mater. 2 (2010) 152−158. 5. H. Kleinke, Chem. Mater. 22 (2010) 604−611. 6. J.S. Sakamoto, H. Schock, T. Caillat, J.P. Fleurial, R. Maloney, M. Lyle, T. Ruckle, E. Timm, L. Zhang, Sci. Adv. Mater. 3 (2011) 621−632. 7. H. Schock, G. Brereton, E. Case, J. D'Angelo, T. Hogan, et al., J. Energ. Resour-ASME 135 (2013) 022001. 8. J. Yang, D. Morelli, G. Meisner, W. Chen, J. Dyck, C. Uher, Phys. Rev. B 67 (2003) 165207. 9. H. Li, X. Tang, X. Su, Q. Zhang, Appl. Phys. Lett. 92 (2008) 202114. 10. S.Q. Bai, X. Shi, L.D. Chen, Appl. Phys. Lett. 96 (2010) 202102. 11. J.R. Salvador, J. Yang, H. Wang, X. Shi, J. Appl. Phys. 107 (2010) 043705. 12. X. Shi, J. Yang, J.R. Salvador, M. Chi, J.Y. Cho, H. Wang, S. Bai, J. Yang, W. Zhang, L. Chen, J. Am. Chem. Soc. 133 (2011) 7837−7846. 13. M. Puyet, B. Lenoir, A. Dauscher, C. Candolfi, J. Hejtmanek, C. Stiewe, E. Müller, Appl. Phys. Lett. 101 (2012) 222105. 14. G. Rogl, Z. Aabdin, E. Schafler, J. Horky, D. Setman, M. Zehetbauer, M. Kriegisch, O. Eibl, A. Grytsiv, E. Bauer, M. Reinecker, W. Schranz, P. Rogl, J. Alloys Compd. 537 (2012) 183−189. 15. Y.Z. Pei, L.D. Chen, W. Zhang, X. Shi, S. Q. Bai, X.Y. Zhao, Z.G. Mei, X.Y. Li, Appl. Phys. Lett. 89 (2006) 221107. 16. Y.Z. Pei, J. Yang, L.D. Chen, W. Zhang, J.R. Salvador, J. Yang, Appl. Phys. Lett. 95 (2009) 042101. 17. J. Zhang, B. Xu, L.-M. Wang, D. Yu, Z. Liu, J. He, Y. Tian, Appl. Phys. Lett. 98 (2011) 072109. 18. L. Zhang, A. Grytsiv, P. Rogl, E. Bauer, M. Zehetbauer, J. Phys. D: Appl. Phys. 42 (2009) 225405. 19. X. Shi, W. Zhang, L. Chen, J. Yang, Phys. Rev. Lett. 95 (2005) 185503. 20. J. Yang, L. Zhang, Y. Liu, C. Chen, J. Li, D. Yu, J. He, Z. Liu, Y. Tian, B. Xu, J. Appl. Phys. 113 (2013) 113703. 21. H. Fukuoka, S. Yamanaka, Chem. Mater. 22 (2010) 47−51. 22. J.V. Badding, Annu. Rev. Mater. Sci. 28 (1998) 631−658. 23. J. Zhang, B. Xu, L.-M. Wang, D. Yu, J. Yang, F. Yu, Z. Liu, J. He, B. Wen, Y. Tian, Acta Mater. 60 (2012) 1246−1251. 24. J. Yang, B. Xu, L. Zhang, Y. Liu, D. Yu, Z. Liu, J. He, Y. Tian, Mater. Lett. 98 (2013) 171−173. 25. J. Rodriguez-Carvajal, Physica B 192 (1993) 55−69. 26. G.S. Nolas, G.A. Slack, D.T. Morelli, T.M. Tritt, A.C. Ehrlich, J. Appl. Phys. 79 (1996) 4002−4008. 27. L.D. Chen, T. Kawahara, X.F. Tang, T. Goto, T. Hirai, J.S. Dyck, W. Chen, C. Uher, J. Appl. Phys. 90 (2001) 1864−1868. 28. P. Qiu, X. Shi, Y. Qiu, X. Huang, S. Wan, W. Zhang, L. Chen, J. Yang, Appl. Phys. Lett. 103 (2013) 062103. 29. T. He, J. Chen, H. D. Rosenfeld, M. A. Subramanian, Chem. Mater. 18 (2006) 759−762. 30. D.T. Morelli, G.P. Meisner, B. Chen, S. Hu, C. Uher, Phys. Rev. B 56 (1997) 7376−7383. 31. H. Li, X. Tang, Q. Zhang, C. Uher, Appl. Phys. Lett. 93 (2008) 252109. 32. B.C. Sales, B.C. Chakoumakos, D. Mandrus, Phys. Rev. B 61 (2000) 2475−2481. 33. G.S Nolas, C.A. Kendziora, Phys. Rev. B 59 (1999) 6189−6192.
Table I. Nominal composition, relative density (den.), filling fraction (FF), room temperature lattice parameter (a), positional parameters (y, z) of Sb, Seebeck coefficient (S), electrical resistivity (ρ), and thermal conductivity (κ) for IyCo4Sb12. Nominal composition
den. [%]
I0.4Co4Sb12
97.4
I0.8Co4Sb12 I1.2Co4Sb12
S [µV/K]
ρ [µΩ/m]
[W/mK]
0.13(2) 9.0374(1) 0.3360(2) 0.1580(2)
60
253
1.79
98.6
0.72(2) 9.1211(1) 0.3364(1) 0.1577(2)
58
122
0.88
98.2
0.79(2) 9.1217(1) 0.3372(2) 0.1571(3)
63
87
0.79
FF
a [Å]
y
z
κ
Figure 1. XRD patterns of IyCo4Sb12. The dashed line at 2θ = 59° is a guide for the eye to see the shift of the peaks. The inset shows the calculated formation enthalpy as a function of applied pressure.
Figure 2. The band structures and density of states (DOS) near the Fermi level (EF, set as zero) for ICo4Sb12 and CoSb3.
8
Figure 3. Temperature dependent thermal conductivity κ of IyCo4Sb12 compared with the data from 15 27 28 29 K0.45Co4Sb12 , Ba0.38Co4Sb12 , Ce0.11Co4Sb12 and In0.05Co4Sb12 .
Figure 4. Lattice thermal conductivity κl of IyCo4Sb12 versus filling fraction y compared with the data 17 15 27 28,30 31 32 29 from Li- , K- , Ba- , Ce, Yb- , Tl- and In-filled CoSb3 at room temperature.
9
Figure 5. Raman spectra of IyCo4Sb12 at room temperature.
Figure 6. Molar heat capacity divided by temperature versus temperature square for IyCo4Sb12 and CoSb3 (see text for details).
10
1) Iodine filled IyCo4Sb12 was synthesized for the first time. 2) The highest filling fraction ever reported in elemental filled CoSb3 materials. 3) Very strong phonon scattering effect for iodine fillers. 4) The lowest thermal conductivity among all skutterudites.
11
S0925-8388(14)01565-5 http://dx.doi.org/10.1016/j.jallcom.2014.06.198 JALCOM 31614
To appear in:
Journal of Alloys and Compounds
Received Date: Revised Date: Accepted Date:
8 May 2014 20 June 2014 28 June 2014
Please cite this article as: X. Li, B. Xu, L. Zhang, F. Duan, X. Yan, J. Yang, Y. Tian, Synthesis of iodine filled CoSb3 with extremely low thermal conductivity, Journal of Alloys and Compounds (2014), doi: http://dx.doi.org/ 10.1016/j.jallcom.2014.06.198
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Synthesis of iodine filled CoSb3 with extremely low thermal conductivity Xiaodong Lia, Bo Xua, Long Zhanga,*, Fenfen Duana, Xinlin Yanb, Jianqing Yang a, and Yongjun Tiana
a
State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao, Hebei 066004, China
b
Institute of Solid State Physics, Vienna University of Technology, Wiedner Hauptstr. 8-10, 1040 Vienna, Austria
* Author to whom correspondence should be addressed. Electronic mail: [email protected]
Abstract p-type iodine filled CoSb3 skutterudite, which is inaccessible under ambient pressure, was synthesized with high pressure synthesis technique. The successful filling of iodine into voids of CoSb3 crystal structure was verified by combined measurements including X-ray diffraction, Raman spectroscopy, and specific heat. The highest filling fraction of 0.79 was achieved for the sample with a nominal composition of I1.2Co4Sb12. Beneficial from such a high filling fraction of iodine and the strong phonon-1
-1
scattering from iodine fillers, the thermal conductivity reached as low as 0.7 Wm K for I0.79Co4Sb12, which is the lowest value among all elemental filled skutterudites. Introducing iodine fillers is thus a practical way to further suppress thermal conductivity for single- and multiple-filled skutterudites.
Keywords: Thermoelectric materials; High-pressure; Solid state reactionsa; Skutterudite
1. Introduction Thermoelectric (TE) materials attracted extensive attention for environment-friendly power generation and refrigeration applications. TE performance of TE materials is evaluated by the 2
dimensionless figure of merit, ZT = S T/(ρκ), where S, ρ, κ, and T are the Seebeck coefficient, electrical resistivity, thermal conductivity, and temperature, respectively. Filled CoSb3-based filled skutterudite with a chemical formula of RxCo4Sb12 is one of the most successful prototypes of phonon1
glass electron crystal (PGEC) concept proposed by Slack, and is likely the most potential TE material for waste heat recovery (power generation) with enhanced TE properties.2-7 A significant advantage for this class of materials comes from the loosely bonded ionic fillers, which can rattle in the vacancies of skutterudite and scatter phonons strongly to reduce the thermal conductivity. Different filling elements have been explored ranging from heavy to light elements, e.g., rare-earth,
8-10
alkaline
earth,11-14 or alkaline metal.15-17 Moreover, multiple-filling strategy demonstrated further suppression of thermal conductivity since multiple fillers with distinct vibrational frequencies can scatter a broader spectrum of phonons for lower thermal conductivity.
12, 18
Still, the thermal conductivity remains high for
elemental filled CoSb3-based skutterudites and exploration of new filling species is anticipated. Previously, a simple criteria for fillable elemental species was proposed, which claimed the electronegativity difference between Sb and the filling element must be larger than 0.8 for the filled CoSb3 to be stable. Mg,
20
19
Almost all the filling species in CoSb3 reported so far satisfied this criteria except
which has a electronegativity difference smaller than 0.8 and can be filled into CoSb3 voids
under high pressure. Recently, Fukuoka et al. synthesized I0.9Rh4Sb12 using high pressure synthesis (HPS) technique.
21
Compared with other filling species, iodine atoms were revealed in an anionic state
-
(I ) with a longer I−Sb separation. Such a longer separation would lead to a looser bond, and consequentially a lower thermal conductivity. Obviously, pressure played a vital role for Mg and I filling. High pressure can lower the reaction temperature, facilitate the synthesis of metastable phase, and acquire new compound inaccessible under ambient pressure by shifting reaction equilibrium.22 Our previous works highlighted high pressure as a fundamental thermodynamic variable to synthesize elemental (e.g. Li,
17, 23
Gd,
24
20
and Mg ) filled CoSb3, which cannot be filled under ambient
pressure. It is thus highly motivating to synthesize iodine filled CoSb3 under high pressure, and to investigate how anionic fillers such as I- would affect the thermoelectric properties. Here we report the
2
successful synthesis of I-filled CoSb3 by using HPS technique. Iodine shows a high filling fraction in CoSb3, resulting in strong phonon scattering and remarkably suppressed thermal conductivity. 2. Experimental procedures Pure I2 (99.8%), Co (99.8%), and Sb (99.999%) powder were mixed according to a nominal atomic ratio of yn:4:12 (yn = 0.4, 0.8, and 1.2). The mixture was loaded into a steel mould, shaped with a cold press method, inserted into an h-BN crucible, and loaded into a high pressure apparatus for HPS. The first step of HPS was carried out at 1073 K and 5 GPa for 0.5 h. The resulting ingot was ground into powder under argon atmosphere, shaped with cold press, and loaded into the high pressure apparatus again for the second step of HPS (753 K and 5 GPa for 3 h). The obtained product was ground into powder and washed in distilled water to remove the excessive I2. The final powders were sintered at 5 GPa and 573 K for 0.5 h into dense pellet and cut for the transport property measurements. The processes for structural optimization and total energy calculation of ICo4Sb12 can be found elsewhere.20 X-ray diffraction (XRD) measurements were carried out with a Rigaku D/MAX/2500/PC (Cu Kα). Rietveld refinements were performed using the FULLPROF program to determine the filling fraction (FF) and lattice parameter.
25
Electrical resistivity and Seebeck coefficient were measured with
ZEM-3 (Ulvac-Riko), and thermal conductivity was measured with TC-7000H (Ulvac-Riko). The densities were measured using the Archimedes method in ethanol, in which the repeatability was 99.5%. The heat capacity measurements were carried out with the heat capacity option of the Physical Property Measurement System. Raman scattering measurements were performed by using a Renishaw inVia system at room temperature. 3. Results and discussion Figure 1 shows the XRD patterns of the final products with the characteristic peaks indexed. All samples were dominated with a skutterudite structure of Im 3 symmetry. The vanishing of (211) peak for the samples with higher iodine content verifies the successful filling of iodine atoms into skutterudite cages.26 The formation enthalpy of ICo4Sb12 (defined as ∆H = EICo4Sb12－E I－4ECo－12ESb, inset in Fig. 1) decreases almost linearly with increasing pressure, indicating a pressure-driven reaction. This is the origin we performed the synthesis experiments at 5 GPa pressure. The diffraction peaks (e.g., 2θ = 59° marked with a dashed line in Fig. 1) shift to smaller angle with increasing iodine content, indicating the expansion of the unit cell after iodine filling. In Table 1, we list room
3
temperature structural parameters of the skutterudite phases from Rietveld refinement. The FFs for yn = 0.4, 0.8, and 1.2 are 0.13, 0.72, and 0.79, respectively, indicating a saturated FF of iodine near 0.8, which is the highest FF reported in filled CoSb3 to date. We noted the lattice parameter increases significantly with increasing iodine content and the values over 9.1211(1) Å are the biggest one for filled CoSb3. This large lattice parameter obviously is connected to the large size of the filling anions. Also listed in Table 1 are the Seebeck coefficient and electrical resistivity for all samples at room temperature. The positive Seebeck coefficient indicates p-type charge carrier of hole in the systems due to the formation of iodine anion (electron acceptor). Increasing the FF of iodine can increase hole concentration. As a result, both the Seebeck coefficient and electrical resistivity should decrease with increasing FF of iodine. However, the Seebeck coefficient is almost constant with variable FFs. To clarify this issue, we calculated the band structures of CoSb3 and fully filled ICo4Sb12 along with the corresponding density of states (DOS) near the Fermi energy, as shown in Fig. 2. The calculated electronic structures of ICo4Sb12 are very similar to that of CoSb3. Iodine filling shows limited influence on the DOS of CoSb3 except for moving the Fermi level downward into the valence band and steepening the DOS near the Fermi level. In the Energy range of 0 to -0.2 eV, the plateau of DOS is 43 electrons/eV for ICo4Sb12 (Fig. 2b), while only 32 electrons/eV for CoSb3 (Fig. 2d). The steeper profile of DOS can enhance the Seebeck coefficient, and thereby compensates the negative effect induced by increasing hole concentration at higher FF of iodine. Therefore, the Seebeck coefficient keeps nearly invariable for various compositions. Figure 3 shows the temperature dependent thermal conductivity for IyCo4Sb12 compared with the 15
27
28
29
lowest thermal conductivity in each investigated series for Ay Co4Sb12 (A = K , Ba , Ce , In ). The thermal conductivity follows approximately the 1/T relation for I0.13Co4Sb12, while shows a quadratic temperature-dependence for I0.72Co4Sb12 and I0.79Co4Sb12 arising from the bipolar effect. Over the measurement temperature range, κ are in the range of 1.79−1.44 W/mK for I0.13Co4Sb12, and 1−0.7 W/mK for the samples with higher FFs (y = 0.72 and 0.79). κ reduced dramatically with higher FF of iodine. Moreover, these values are the lowest ones in filled skutterudites to the best of our knowledge. The significant suppression of the thermal conductivity for iodine filled CoSb3 compared with AyCo4Sb12 (A = K, Ba, Ce and In) are obviously demonstrated in Fig. 3. The room temperature lattice thermal conductivity κl versus filling fraction for IyCo4Sb12 is displayed in Fig. 4. κl was derived by subtracting electron contribution κl = LT/ρ from the total thermal
4
-8
2
-2
conductivity, where L = 2.44×10 V K is the Lorenz number and ρ is the electrical resistivity. The 17
lattice thermal conductivity for IyCo4Sb12 again is greatly suppressed compared with those for Li- , K15
27
28, 30
, Ba- , Ce-
31
, Yb- , Tl-
32
and In-filled
29
CoSb3. These facts indicate that the phonons are
scattered much more strongly by iodine fillers than by any other fillers. We speculate that such intensive scattering is related to the very high FF of iodine as well as the loosely bonding of iodine due 21
to the longer I−Sb separation than that in other filled skutterudites . The very low thermal conductivity of iodine filled CoSb3 can be attributed to the high filling fraction of iodine. In addition, after iodine atoms filled into the voids of CoSb3, the large radius of iodine atoms would induce lattice distortions locally, which may further depress the lattice thermal conductivity. To investigate the perturbation to Sb4-rings after iodine filling, Raman spectra were collected as a function of iodine content (Fig. 5). With higher iodine filling fraction, the lower-energy Ag mode is intensively broadened and flattened to overlap with the background, while the higher-energy one almost unchanged. The broadening of Raman peak is mainly due to the distortion of the structure unit. The lower-energy Ag mode is presumably caused by the stretching of the longer Sb−Sb bond and the higher-energy one the shorter 33
Sb−Sb bond . Raman results indicate the deformation of the longer Sb−Sb bond is easier than that of the shorter Sb−Sb bond. It is also noted that the peak of the higher-energy Ag mode shifts to higher position after iodine filling. These results confirm the success of iodine filling and evidence the severe distortion of Sb4-rings by iodine filling. To clarify the existence of localized incoherent vibrations associated with iodine anion in the voids of the host lattice, heat capacity of I0.79Co4Sb12 were measured from 2 to 50 K. Our previous heat capacity data for CoSb3 were used as the reference.
20
The contribution of iodine to the heat capacity
has been determined form the difference in the molar heat capacity of I0.79Co4Sb12 and CoSb3 (Fig. 6). The solid line through the data represents a fit to the equation modeled by the Einstein law: ∆C p = C p , I
0.79 Co4 Sb12
− C p ,Co Sb = γ T + Ax 2 e x 4
12
(e
x
2
−1 ,
)
23
(1) -2
-1
where γT is the electron contribution to the molar heat capacity and, γ = 0.040±0.004 JK mol , A = 18±1 JK-1mol -1 and x = ΘE/T with an Einstein temperature ΘE = 84±1 K. The filling fraction determined from A/3R (R is the ideal gas constant) is 0.72, which is consistent with the Rietveld refined value 0.79. This is the highest filling fraction in filled CoSb3 and benefit the suppression of thermal conductivity. 4. Conclusions
5
In summary, theoretical investigations revealed a pressure-driven reaction for ICo4Sb12. Anion filled p-type IyCo4Sb12 skutterudites, which are inaccessible at ambient pressure, were synthesized by a stepwise synthesis method using HPS technique at 5 GPa. A saturated filling fraction of 0.79 was evaluated from Rietveld refinement and further confirmed by the heat capacity measurement, which is the highest filling fraction ever reported in elemental filled CoSb3 materials. Beneficial from such a high filling fraction, the lowest thermal conductivity is achieved among all skutterudites. Interestingly, even I0.13Co4Sb12 with a relatively low filling fraction shows a very small thermal conductivity compared with other elemental filled CoSb3, indicating I-filling strongly scatters phonons and thus significantly suppresses thermal conductivity. The investigations of incorporation of iodine fillers in multiple-filled skutterudite are ongoing. Acknowledgments This work was supported by the National Science Foundation of China (51201149, 51121061, 51072175), the Natural Science Foundation for Distinguished Young Scholars of Hebei Province of China (E2014203150), and the Key Basic Research Project of Hebei (14961013D).
6
References 1. G.A. Slack, in: D.M. Rowe (Ed.), CRC Handbook of Thermoelectrics, CRC Press, Boca Raton, FL, 1995, pp. 407−440. 2. C. Uher, Semiconduct. Semimet. 69 (2001) 139−253. 3. G.J. Snyder, E.S. Toberer, Nature Mater. 7 (2008) 105−114. 4. J.F. Li, W.S. Liu, L.D. Zhao, M. Zhou, NPG Asia Mater. 2 (2010) 152−158. 5. H. Kleinke, Chem. Mater. 22 (2010) 604−611. 6. J.S. Sakamoto, H. Schock, T. Caillat, J.P. Fleurial, R. Maloney, M. Lyle, T. Ruckle, E. Timm, L. Zhang, Sci. Adv. Mater. 3 (2011) 621−632. 7. H. Schock, G. Brereton, E. Case, J. D'Angelo, T. Hogan, et al., J. Energ. Resour-ASME 135 (2013) 022001. 8. J. Yang, D. Morelli, G. Meisner, W. Chen, J. Dyck, C. Uher, Phys. Rev. B 67 (2003) 165207. 9. H. Li, X. Tang, X. Su, Q. Zhang, Appl. Phys. Lett. 92 (2008) 202114. 10. S.Q. Bai, X. Shi, L.D. Chen, Appl. Phys. Lett. 96 (2010) 202102. 11. J.R. Salvador, J. Yang, H. Wang, X. Shi, J. Appl. Phys. 107 (2010) 043705. 12. X. Shi, J. Yang, J.R. Salvador, M. Chi, J.Y. Cho, H. Wang, S. Bai, J. Yang, W. Zhang, L. Chen, J. Am. Chem. Soc. 133 (2011) 7837−7846. 13. M. Puyet, B. Lenoir, A. Dauscher, C. Candolfi, J. Hejtmanek, C. Stiewe, E. Müller, Appl. Phys. Lett. 101 (2012) 222105. 14. G. Rogl, Z. Aabdin, E. Schafler, J. Horky, D. Setman, M. Zehetbauer, M. Kriegisch, O. Eibl, A. Grytsiv, E. Bauer, M. Reinecker, W. Schranz, P. Rogl, J. Alloys Compd. 537 (2012) 183−189. 15. Y.Z. Pei, L.D. Chen, W. Zhang, X. Shi, S. Q. Bai, X.Y. Zhao, Z.G. Mei, X.Y. Li, Appl. Phys. Lett. 89 (2006) 221107. 16. Y.Z. Pei, J. Yang, L.D. Chen, W. Zhang, J.R. Salvador, J. Yang, Appl. Phys. Lett. 95 (2009) 042101. 17. J. Zhang, B. Xu, L.-M. Wang, D. Yu, Z. Liu, J. He, Y. Tian, Appl. Phys. Lett. 98 (2011) 072109. 18. L. Zhang, A. Grytsiv, P. Rogl, E. Bauer, M. Zehetbauer, J. Phys. D: Appl. Phys. 42 (2009) 225405. 19. X. Shi, W. Zhang, L. Chen, J. Yang, Phys. Rev. Lett. 95 (2005) 185503. 20. J. Yang, L. Zhang, Y. Liu, C. Chen, J. Li, D. Yu, J. He, Z. Liu, Y. Tian, B. Xu, J. Appl. Phys. 113 (2013) 113703. 21. H. Fukuoka, S. Yamanaka, Chem. Mater. 22 (2010) 47−51. 22. J.V. Badding, Annu. Rev. Mater. Sci. 28 (1998) 631−658. 23. J. Zhang, B. Xu, L.-M. Wang, D. Yu, J. Yang, F. Yu, Z. Liu, J. He, B. Wen, Y. Tian, Acta Mater. 60 (2012) 1246−1251. 24. J. Yang, B. Xu, L. Zhang, Y. Liu, D. Yu, Z. Liu, J. He, Y. Tian, Mater. Lett. 98 (2013) 171−173. 25. J. Rodriguez-Carvajal, Physica B 192 (1993) 55−69. 26. G.S. Nolas, G.A. Slack, D.T. Morelli, T.M. Tritt, A.C. Ehrlich, J. Appl. Phys. 79 (1996) 4002−4008. 27. L.D. Chen, T. Kawahara, X.F. Tang, T. Goto, T. Hirai, J.S. Dyck, W. Chen, C. Uher, J. Appl. Phys. 90 (2001) 1864−1868. 28. P. Qiu, X. Shi, Y. Qiu, X. Huang, S. Wan, W. Zhang, L. Chen, J. Yang, Appl. Phys. Lett. 103 (2013) 062103. 29. T. He, J. Chen, H. D. Rosenfeld, M. A. Subramanian, Chem. Mater. 18 (2006) 759−762. 30. D.T. Morelli, G.P. Meisner, B. Chen, S. Hu, C. Uher, Phys. Rev. B 56 (1997) 7376−7383. 31. H. Li, X. Tang, Q. Zhang, C. Uher, Appl. Phys. Lett. 93 (2008) 252109. 32. B.C. Sales, B.C. Chakoumakos, D. Mandrus, Phys. Rev. B 61 (2000) 2475−2481. 33. G.S Nolas, C.A. Kendziora, Phys. Rev. B 59 (1999) 6189−6192.
Table I. Nominal composition, relative density (den.), filling fraction (FF), room temperature lattice parameter (a), positional parameters (y, z) of Sb, Seebeck coefficient (S), electrical resistivity (ρ), and thermal conductivity (κ) for IyCo4Sb12. Nominal composition
den. [%]
I0.4Co4Sb12
97.4
I0.8Co4Sb12 I1.2Co4Sb12
S [µV/K]
ρ [µΩ/m]
[W/mK]
0.13(2) 9.0374(1) 0.3360(2) 0.1580(2)
60
253
1.79
98.6
0.72(2) 9.1211(1) 0.3364(1) 0.1577(2)
58
122
0.88
98.2
0.79(2) 9.1217(1) 0.3372(2) 0.1571(3)
63
87
0.79
FF
a [Å]
y
z
κ
Figure 1. XRD patterns of IyCo4Sb12. The dashed line at 2θ = 59° is a guide for the eye to see the shift of the peaks. The inset shows the calculated formation enthalpy as a function of applied pressure.
Figure 2. The band structures and density of states (DOS) near the Fermi level (EF, set as zero) for ICo4Sb12 and CoSb3.
8
Figure 3. Temperature dependent thermal conductivity κ of IyCo4Sb12 compared with the data from 15 27 28 29 K0.45Co4Sb12 , Ba0.38Co4Sb12 , Ce0.11Co4Sb12 and In0.05Co4Sb12 .
Figure 4. Lattice thermal conductivity κl of IyCo4Sb12 versus filling fraction y compared with the data 17 15 27 28,30 31 32 29 from Li- , K- , Ba- , Ce, Yb- , Tl- and In-filled CoSb3 at room temperature.
9
Figure 5. Raman spectra of IyCo4Sb12 at room temperature.
Figure 6. Molar heat capacity divided by temperature versus temperature square for IyCo4Sb12 and CoSb3 (see text for details).
10
1) Iodine filled IyCo4Sb12 was synthesized for the first time. 2) The highest filling fraction ever reported in elemental filled CoSb3 materials. 3) Very strong phonon scattering effect for iodine fillers. 4) The lowest thermal conductivity among all skutterudites.
11