# Full text of "Quasi-two-dimensional Fermi surfaces of the heavy-fermion superconductor Ce$_2$PdIn$_8$"

## See other formats

arXiv:1509.02657vl [cond-mat.str-el] 9 Sep 2015 Quasi-two-dimensional Fermi surfaces of the heavy-fermion superconductor Ce2PdIn8 K. Gotze/ J. Klotz,^ D. Gnida,^ H. Harima,^ D. A. Demuer,® S. Elgazzar/ J. Wosnitza/ D. Kaczorowski,^ and 1. Sheikin®’0 ^ Hochfeld-Magnetlabor Dresden (HLD-EMFL), Helmholtz-Zentrum Dresden-Rossendorf and TU Dresden, D-01314 Dresden, Germany ^Institute of Low Temperature and Structure Research, Polish Academy of Sciences, P.O. Box I 4 IO, PL-50-950 Wroclaw, Poland ^Graduate School of Science, Kobe University, Kobe 657-8501, Japan '^IMR, Tohoku University, Oarai, Ibaraki 311-1313, Japan ^INAG/SPSMS, GEA Grenoble, 38054 Grenoble, France ^Laboratoire National des Champs Magnetiques Intenses (LNCMI-EMFL), CNRS, UJF, 38042 Grenoble, France ^Highly Correlated Matter Research Group, Department of Physics, University of Johannesburg, P.O. Box 524, Auckland Park 2006, South Africa (Dated: September 10, 2015) We report low-temperature de Haas-van Alphen (dHvA) effect measurements in magnetic fields up to 35 T of the heavy-fermion superconductor Ce 2 PdIng. The comparison of the experimental results with band-structure calculations implies that the 4/ electrons are itinerant rather than localized. The cyclotron masses estimated at high field are only moderately enhanced, 8 and 14 mo, but are substantially larger than the corresponding band masses. The observed angular dependence of the dHvA frequencies suggests quasi-two-dimensional Fermi surfaces in agreement with band-structure calculations. However, the deviation from ideal two dimensionality is larger than in CeCoIns, with which Ce 2 PdIn 8 bears a lot of similarities. This subtle distinction accounts for the different superconducting critical temperatures of the two compounds. PACS numbers: 71.18.-|-y,71.27.-|-a,74.70.Tx The appearance of unconventional superconductivity in the vicinity of a quantum critical point (QCP) is a common trend in Ge-based heavy-fermion (HF) com¬ pounds. A more recent and still somewhat controversial issue is the influence of the Fermi-surface (FS) dimen¬ sionality on unconventional superconductivity. Indeed, reduced dimensionality of the FS leads to nesting-type magnetic instabilities [l| and thus enhances the supercon¬ ductivity ii- The exact knowledge of the FS topology of HF systems is, therefore, essential. In addition, this information allows distinguishing if the / electrons are itinerant or localized, i.e. whether they contribute to the FS or not. Ce 2 PdIn 8 is a recently discovered HF superconductor with Tc = 0.7 K and a non-magnetic ground state Si- Non-Fermi-liquid behavior was observed in both macro¬ scopic and microscopic 0[ll measurements at low temperature, implying that Ce 2 PdIn 8 is located very close to a QGP. It was further suggested that a two- dimensional (2D) SDW-type QCP is induced by m^- netic field near the upper critical field, Hc 2 ss 2 T [9|. Unconventional superconductivity was demonstrated to be due to antiferromagnetic quantum fluctuations [II- These unusual properties are strikingly similar to those of the well-studied HF superconductor CeCoIns [I4l4l7l| . which is also located very close to a QCP at ambient pressure. However, the superconducting critical temper¬ ature Tc = 2.3 K of CeCoIns [l^ is considerably higher than that of Ce 2 PdIn 8 . Ce 2 PdIn 8 crystallizes into a tetragonal Ho 2 CoGa 8 - type crystal structure with space group PA/mmm. It belongs to the larger family of Ce„TIn 3„+2 {T: transi¬ tion metal, n = 1, 2, and 00 ) systems, containing a se¬ quence of n Celns layers intercalated by a TIn 2 layer along the c axis. While cubic Celna (n = 00 ) is a completely isotropic system, the layered structures with n = 1 and 2 are expected to lead to anisotropic prop¬ erties and quasi-2D FSs. Indeed, quasi-2D FS sheets were observed in both n = 1 systems CeCoIns [11113, Celrins flijl. C eRhln.^ [ 2 I [23j | and the n = 2 compound Ce9RhIns |24 HI. The degree of two dimensionality is expected to be larger in monolayer systems with alter¬ nating layers of Celna and TIn 2 than in their bilayer counterparts, in which two Celna layers are separated by one TIn 2 layer. In this paper, we report high-held dHvA measurements of Ce 2 PdIn 8 . The observed FS is quasi-2D, however, we hnd that the more three-dimensional crystal structure of Ce 2 PdIn 8 relative to CeCoIns leads to a reduced two dimensionality of the FS topology. We argue that this can explain the difference in the superconducting critical temperatures of the two compounds. Single crystals were grown by the self-hux method Q] , and we have conhrmed by specihc-heat measurements that they are not contaminated by Celna. The dHvA measurements were performed using a torque cantilever magnetometer mounted in a top-loading dilution refriger¬ ator equipped with a low-temperature rotator. Magnetic helds, B, up to 35 T generated by LNCMI-Grenoble re¬ sistive magnets were applied at different angles between 2 the [001] and [100] directions. FIG. 1. Fourier spectrum of the high-held dHvA oscillations (inset) in Ce 2 PdIn 8 for magnetic held applied at 4° off the c axis at 30 mK. Figure [T] shows the oscillatory torque after subtract¬ ing a non-oscillating background and the corresponding Fourier transform in Ce 2 PdIn 8 . Four fundamental fre¬ quencies, denoted C; Vj crii ^-nd a 2 , are observed when B is applied close to the c axis. The oscillations were traced up to 60°, where their amplitude decreased below the noise level. To figure out whether the / electrons are itinerant or localized in Ce 2 PdIng, we performed band-structure cal¬ culations for both Ce 2 PdIn 8 and La 2 PdIn 8 , the latter corresponding to the Ce compound with localized / elec¬ trons. For both compounds, the calculations were car¬ ried out using a full potential augmented plane wave method with the local density approximation (LDA) for the exchange-correlation potential. As crystals of La 2 PdIn 8 are currently unavailable, the lattice parame¬ ters of Ce 2 PdIn 8 were used for the La 2 PdIn 8 calculations. The resulting FSs are shown in Fig. [2j Given the layered crystal structure, it is not surpris¬ ing that some of the calculated FS sheets are quasi-2D in both Ce 2 PdIn 8 and La 2 PdIn 8 . The details of the FSs are, however, clearly different. In contrast, the topol¬ ogy of the 4/-itinerant FS of CeCoIng is similar to the 4/-localized FS of LaRhlng and CeRhIns 22, 23|, where the two FSs differ mainly by size. It should be noted that CeCoIns is a compensated metal with equal car¬ rier numbers of electrons and holes, while LaCoIus is an uncompensated metal. On the other hand, both Ce 2 PdIn 8 and La 2 PdIn 8 are compensated metals. As seen in Fig.[5J the charge-carrier number given by the FS volume in Ce 2 PdIn 8 is about two times smaller than that in La 2 PdIn 8 . Figure [3l[a) shows the experimentally observed angu¬ lar dependence of the dHvA frequencies in Ce 2 PdIn 8 together with the results of band-structure calculations based on the 4/-itinerant band model. Experimental and calculated frequencies and effective masses are also shown Branch Experiment F (kT) m* jm^ Calculation F (kT) mbimo 7 — — 0.34 1.55 5 — — 0.43 0.42 c 0.78 8.4±0.4 0.69 1.24 n 1.2 — 1.07 2.27 (y.2 3.26 — 2.82 0.81 4.82 14±1 4.82 2.12 P2 — — 7.27 2.81 bi — — 13.08 5.2 ^ The effective masses were measured with magnetic field applied at 4° off the c axis TABLE 1. Experimental and calculated dHvA frequencies and effective masses in Ce 2 PdIn 8 for magnetic field along the c axis. in Table ID The agreement between the experimentally observed a branches and those of the calculations is ex¬ cellent. Not only the angular dependences are the same, but even the absolute values agree very well. This implies that both the topology and the size of the calculated FS sheet reproduce the experimental results exceptionally well. The a branches correspond to the quasi-2D FS of the band 73, as shown in Fig. [H Regarding the mea¬ sured lower (< 2 kT) dHvA frequencies, there is also a very good agreement between the experimental and cal¬ culated branches. These branches originate mostly from rather isotropic parts of the FS in band 72. The cal¬ culated /3 branches originating from complicated sheets in band 72 were, however, not observed in the experi¬ ment. This is probably caused by a strongly enhanced effective mass or an unfavorable curvature factor for de¬ tecting the dHvA signal. For comparison, in Fig. |31[b) we plot the experimental results obtained in Ce 2 PdIn 8 together with band-structure calculations for La 2 PdIn 8 . In this case, the two are obviously at odds with each other. In particular, only one quasi-2D FS was observed in the experiment, while the calculations predict two of them for La 2 PdIn 8 , which should be easy to detect. The comparison of the experimentally observed dHvA frequencies with the results of the LDA band-structure calculations thus gives clear evidence for a quasi-2D FS with itinerant / electrons in Ce^P dlns . The same con¬ clusion was drawn for CeCoIns [13, l20| . We, alternatively, calculated the band structure using the local spin density approximation with the relativis¬ tic version of the full-potential local orbital method 261 for Ce 2 PdIn 8 . This also suggests quasi-2D FS, but the agreement with the experimental results is not as good. The effective masses shown in Table |D were determined by fitting the temperature dependence of the oscilla¬ tory amplitude by the standard Lifshitz-Kosevich for¬ mula [27|. This was done for the magnetic field applied at 4° off the c axis. Due to the small amplitudes of the 3 FIG. 2. Calculated FSs of Ce 2 PdIn 8 (left). Calculated FSs of La 2 PdIn 8 (right) are also shown for comparison. oscillations the field range from 28 to 34.5 T was used for the analysis. Even for such high fields, the effective masses of only two branches, ( and ai, could be reliably determined. The obtained values are 8.4±0.4 and 14±1 mo, respectively. The effective mass of the q.\ branch cor¬ responding to the quasi-2D sheet of the FS is comparable to the values, 8 - 18 mg, reported for the quasi-2D FS of CeCoIns [l^l^[2^. This implies a similar degree of hy¬ bridization between the / and conduction electrons. The detected effective masses are, however, by far too small to account for the huge value of the electronic specific heat coefficient, 7 , of the order of 1 ^K^mol just above the superconducting transition iii- Presumably, the effective masses of the /3 branches, which are not ob¬ served here, are strongly enhanced. Indeed, already the calculated band masses of the /3 branches are higher than those of the other branches (see Table H]). On the other hand, the Sommerfeld coefficient of Ce 2 PdIn 8 is similar to that of CeCoIns [l^, where the effective masses were reported to strongly decrease with magnetic field 0 - While the observed dHvA oscillations in Ce 2 PdIn 8 are not strong enough to perform the field-dependent anal¬ ysis of the effective masses, they can also be expected to decrease with magnetic field. This assumption is sup¬ ported by the experimentally observed field dependence of the T^-coefficient in the resistivity 0 and of the Som¬ merfeld coefficient of the specific heat 9| above the upper critical field. As shown in Fig. 21 the major FS sheets of both Ce 2 PdIn 8 and CeCoIns are quasi-2D corrugated cylin¬ ders extending along the [001] direction. As many of the physical properties of HF materials strongly depend on the FS dimensionality, the key question here is which FS is more 2D, i.e. which amplitude of the corrugation is smaller. For both Ce 2 PdIn 8 and CeCoIns, the FSs experimentally determined through dHvA measurements are in excellent agreement with calculated ones. This is, however, not always the case. That is why we introduce a quantitative criterion of a quasi-2D FS deviation from an ideal cylinder: A = (S^ax - Srain)IS, where S^ax and Smin are the maximum and minimum extremal cross- sections respectively, and S is the average cross-section of the warped cylinder. For an ideal cylinder, A = 0. Since the extremal cross-sections, S'i, of the FS are pro¬ portional to the dHvA frequencies, F), S can be replaced by F measured with field along [001] to determine A experimentally. In Ce 2 PdIn 8 , with the two frequencies, a\ and a-i, listed in Table [H this yields A = 0.386. The dHvA effect measurements in CeCoIns revealed three ex¬ tremal cross-section of the quasi-2D FS 0,0,0 . The reported values of the dHvA frequencies are slightly dif¬ ferent, yielding the average value of A = 0.221. This implies that the deviation from the ideal 2D FS is much smaller in CeCoIns than in Ce 2 PdIn 8 . This is expected as well from the more 3D crystal structure of Ce 2 PdIn 8 as compared to CeCoIns. The corresponding larger anisotropy of CeCoIns as compared to Ce 2 PdIn 8 accounts for the higher supercon¬ ducting critical temperature of CeCoIns. In fact, 2.3 K in CeCoIns is the highest among all the known Ce-based 4 Field angle (deg.) FIG. 3. (color online) Angular dependence of the experimentally observed dHvA frequencies in Ce 2 PdIng (big open circles) is shown together with the results of band-structure calculations (small closed circles) performed for Ce 2 PdIng (a) and La 2 PdIng (b). The latter correspond to Ce 2 PdIng with localized 4/ electrons and are shown for comparison. Very low calculated dHvA frequencies that correspond to small FS pockets are not shown for clarity. HF materials. Remarkably, the FS of CeCoIns is the most 2D like as compared to its Ir and Rh analogs (^ . The FS of CeRhInsjhowever, changes at its critical pres¬ sure Pc — 2.4 GPa [29|, and the reported dHvA frequen¬ cies yield A = 0.17 above Pc- This value is similar to that of CeCoIns at ambient pressure. In CeRhIns, super¬ conductivity emerges around Pc, where Pc = 2.1 K [30j, a value close to that of CeCoIns. Regar ding Celrins, experimentally observed dHvA frequencies |2l| result in A = 0.269, a value in between those for CeCoIm and Ce 2 PdIn 8 . However, Tc = 0.4 K of Celrins 31| can not be compared directly to the critical temperatures of CeCoIus and Ce^Pdlug, as Celrins is located further away from a QCP [S^- When Celrins is tuned to a QCP by Rh substitution, Tc increases to about 1 K [s^, and is likely to be reduced due to disorder as compared to pure compounds. Consistently with A, this value also falls in between those for CeCoIns and Ce 2 PdIn 8 . Un¬ fortunately, there is currently no information about the FS of Rh-substituted Celrins. While the FS dimension¬ ality is not the only factor that determines Tc in HF superconductors, it is certainly a significant one. Indeed, Tc = 18.5 K was reported for PuCoGas [s^, which is the highest among those yet observed in /-electron materi¬ als. Remarkably, the calculated FS of PuCoGas consists of three corrugated cylinders , although the degree of corrugation is relatively high with A being 0.448, 0.359 and 0.66 respectively. However, the results of these cal¬ culations are still to be confirmed experimentally. In summary, our high-held dHvA investigation of Ce 2 PdIn 8 combined with band-structure calculations ev¬ idence the existence of s quasi-2D FS with itinerant /- electrons in this compound. The comparison of the FS topology of Ce 2 PdIn 8 and CeCoIns implies that the FS of the latter compound is much closer to an ideal cylinder characteristic for a 2D case. The difference in the FS di¬ mensionality accounts for different superconducting crit¬ ical temperatures of the two compounds, which are both located in close vicinity to a QCP and have a similar degree of hybridization between the 4/ and conduction electrons. It would be interesting to apply the quantita¬ tive criterion of the FS two dimensionality we introduced here to other HF materials with quasi-2D FS. In particu¬ lar, the criterion can be used to verify the theoretical pre¬ diction about the influence of the FS dimensionality on the type of quantum criticality in HF compounds . Another interesting question is whether magnetic fields themselves affect the FS dimensionality in Ce 2 PdIn 8 in particular and other quasi-2D HF materials in general. Zero-held ARPES measurements in Ce 2 PdIn 8 would be very useful to address this issue. We are grateful to T. Maehira for sharing with us the details of the band-structure calculations in PuCoGas. KG acknowledges support from the DFG within GRK 1621. We acknowledge the support of the HLD-HZDR and the LNCMI-CNRS, members of the European Mag- 5 Ce2PdIng FIG. 4. Comparison of the calculated quasi-2D FSs of CeCoIns and Ce 2 PdIng. netic Field Laboratory (EMFL). and K. Ueda, G. Lonzarich, * ilya.sheikin@lncmi.cnrs.fr [1] T. Moriya, Y. Takahashi, J. Phys. Soc. Jpn. 59, 2905 (1990) [2] P. Monthoux and G. Phys. Rev. B 59, 14598 (1999) [3] P. Monthoux, J. Phys.: Condens. Matter 15, S1973 (2003) [4] D. Kaczorowski, A. P. Pikul, D. Gnida, and V. H. Tran, Phys. Rev. Lett. 103, 027003 (2009) and V. Tran, T. Y. Guan, S. Y. Zhou, D. Gnida, D. Kaczorowski, and S. Y. Li, Phys. Rev. X 1, 011010 (2011) [7] V. H. Tran, D. Kaczorowski, R. T. Khan, and E. Bauer, Phys. Rev. B 83, 064504 (2011) [8] D. Gnida, M. Matusiak, and D. Kaczorowski, Phys. Rev. B 85, 060508 (2012) [9] Y. Tokiwa, P. Gegenwart, D. Gnida, and D. Kac¬ zorowski, Phys. Rev. B 84, 140507 (2011) [10] M. Matusiak, D. Gnida, and D. Kaczorowski, Phys. Rev. B 84, 115110 (2011) [11] H. Fukazawa, R. Nagashima, S. Shimatani, Y. Kohori, and D. Kaczorowski, Phys. Rev. B 86, 094508 (2012) [12] V. H. Tran, A. D. Hillier, D. T. Adroja, and D. Kac¬ zorowski, Phys. Rev. B 86, 094525 (2012) [13] K. Hashimoto, Y. Mizukami, R. Katsumata, H. Shishido, M. Yamashita, H. Ikeda, Y. Mat- suda, J. A. Schlueter, J. D. Fletcher, A. Carring¬ ton, D. Gnida, D. Kaczorowski, and T. Shibauchi, Proc. Natl. Acad. Sci. U.S.A. 110, 3293 (2013) [14] J. Paglione, M. A. Tanatar, D. G. Hawthorn, E. Boaknin, R. W. Hill, F. Ronning, M. Suther¬ land, L. Taillefer, C. Petrovic, and P. C. Canfield, Phys. Rev. Lett. 91, 246405 (2003) [15] A. Bianchi, R. Movshovich, I. Vekhter, P. G. Pagliuso, and J. L. Sarrao, Phys. Rev. Lett. 91, 257001 (2003) [16] E. D. Bauer, C. Capan, F. Ronning, R. Movshovich, J. D. Thompson, and J. L. Sar¬ rao, Phys. Rev. Lett. 94, 047001 (2005) [17] F. Ronning, C. Capan, E. D. Bauer, J. D. Thompson, J. L. Sarrao, and R. Movshovich, Phys. Rev. B 73, 064519 (2006) [18] C. Petrovic, P. G. Pagliuso, M. F. Hund¬ ley, R. Movshovich, J. L. Sarrao, J. D. Thompson, Z. Fisk, and P. Monthoux, J. Phys.: Condens. Matter 13, L337 (2001) [19] R. Settai, H. Shishido, S. Ikeda, Y. Murakawa, M. Nakashima, D. Aoki, Y. Haga, H. Harima, and Y. Onuki, J. Phys.: Condens. Matter 13, L627 (2001) [20] D. Hall, E. C. Palm, T. P. Murphy, S. W. Tozer, Z. Fisk, U. Alver, R. G. Goodrich, J. L. Sarrao, P. G. Pagliuso, and T. Ebihara, Phys. Rev. B 64, 212508 (2001) [21] Y. Haga, Y. Inada, H. Harima, K. Oikawa, M. Murakawa, H. Nakawaki, Y. Tokiwa, D. Aoki, H. Shishido, S. Ikeda, N. Watanabe, and Y. Onuki, Phys. Rev. B 63, 060503 (2001) [22] H. Shishido, R. Settai, D. Aoki, S. Ikeda, H. Nakawaki, N. Nakamura, T. lizuka, Y. Inada, K. Sugiyama, T. Takeuchi, K. Kindo, T. Kobayashi, Y. Haga, H. Harima, Y. Aoki, T. Namiki, H. Sato, and Y. Onuki, J. Phys. Soc. Jpn. 71, 162 (2002) [23] D. Hall, T. P. Murphy, E. C. Palm, S. W. Tozer, Z. Fisk, N. Harrison, R. G. Goodrich, U. Alver, and J. L. Sarrao, Int. J. Mod. Phys. B 16, 3004 (2002) [24] T. Ueda, H. Shishido, S. Hashimoto, T. Okubo, M. Ya- mada, Y. Inada, R. Settai, H. Harima, A. Galatanu, E. Yamamoto, N. Nakamura, K. Sugiyama, T. Takeuchi, K. Kindo, T. Namiki, Y. Aoki, H. Sato, and Y. Onuki, J. Phys. Soc. Jpn. 73, 649 (2004) [25] R. Jiang, D. Mou, C. Liu, X. Zhao, Y. Yao, H. Ryu, C. Petrovic, K.-M. Ho, and A. Kaminski, Phys. Rev. B 91, 165101 (2015) [26] K. Koepernik and H. Eschrig, Phys. Rev. B 59, 1743 (1999) [27] D. Shoenberg, Magnetic oscillations in metals (Cam¬ bridge University Press, 1984). [28] A. Polyakov, O. Ignatchik, B. Bergk, K. Gotze, A. D. Bianchi, S. Blackburn, B. Prevost, G. Sey- farth, M. Cote, D. Hurt, C. Capan, Z. Fisk, R. G. Goodrich, I. Sheikin, M. Richter, and J. Wosnitza, Phys. Rev. B 85, 245119 (2012) [29] H. Shishido, R. Settai, H. Harima, and Y. Onuki, J. Phys. Soc. Jpn. 74, 1103 (2005) [30] H. Hegger, C. Petrovic, E. G. Moshopoulou, M. F. Hundley, J. L. Sarrao, Z. Fisk, and J. D. Thompson, Phys. Rev. Lett. 84, 4986 (2000) [31] C. Petrovic, R. Movshovich, M. Jaime, P. G. Pagliuso, M. F. Hundley, J. L. Sarrao, Z. Fisk, and J. D. Thomp¬ son, Europhys. Lett. 53, 354 (2001) [32] M. Matsumoto, M. J. Han, J. Otsuki, and S. Y. Savrasov, Phys. Rev. B 82, 180515 (2010) [33] G.-q. Zheng, N. Yamaguchi, H. Kan, Y. Kitaoka, J. L. 6 Sarrao, P. G. Pagliuso, N. O. Moreno, and J. D. Thomp¬ son, Phys. Rev. B 70, 014511 (2004) [34] J. L. Sarrao, L. A. Morales, J. D. Thompson, B. L. Scott, G. R. Stewart, F. Wastin, J. Rebizant, P. Boulet, E. Go- lineau, and G. H. Lander, Nature 420, 297 (2002). [35] T. Maehira, T. Hotta, K. Ueda, and A. Hasegawa, Phys. Rev. Lett. 90, 207007 (2003) [36] Q. Si, J. H. Pixley, E. Nica, S. J. Ya¬ mamoto, P. Goswami, R. Yu, and S. Kirchner, J. Phys. Soc. Jpn. 83, 061005 (2014) [37] Q. Si, Physica B 378-380, 23 (2006) [38] Q. Si, Phys. Status Solidi B 247, 476 (2010) [39] J. Gusters, K.-A. Lorenzer, M. Muller, A. Prokofiev, A. Sidorenko, H. Winkler, A. M. Strydom, Y. Shimura, T. Sakakibara, R. Yu, Q. Si, and S. Paschen, Nat. Mater. 11, 189 (2012)