Magnetic properties and spin kinetics in Kondo lattices Текст научной статьи по специальности «Физика»

ISSN 2072-5981

Volume 17, Issue 1 Paper No 15104, 1-9 pages 2015

http://mrsej.kpfu.ru http://mrsej.ksu.ru

Established and published by Kazan University Sponsored by International Society of Magnetic

Resonance (ISMAR) Registered by Russian Federation Committee on Press,

August 2, 1996 First Issue was appeared at July 25, 1997

© Kazan Federal University (KFU)*

"Magnetic Resonance in Solids. Electronic Journal" (MRSey) is a

peer-reviewed, all electronic journal, publishing articles which meet the highest standards of scientific quality in the field of basic research of a magnetic resonance in solids and related phenomena. MRSey is free for the authors (no page charges) as well as for the readers (no subscription fee). The language of MRSey is English. All exchanges of information will take place via Internet. Articles are submitted in electronic form and the refereeing process uses electronic mail. All accepted articles are immediately published by being made publicly available by Internet (http://MRSez.kpfu.ru).

Editors-in-Chief Jean Jeener (Universite Libre de Bruxelles, Brussels) Boris Kochelaev (KFU, Kazan) Raymond Orbach (University of California, Riverside)

Executive Editor Yurii Proshin (KFU, Kazan) [email protected]

[email protected]

Editors

Vadim Atsarkin (Institute of Radio Engineering and Electronics, Moscow) Yurij Bunkov (CNRS, Grenoble) Mikhail Eremin (KFU, Kazan) David Fushman (University of Maryland,

College Park)

Hugo Keller (University of Zürich, Zürich) Yoshio Kitaoka (Osaka University, Osaka) Boris Malkin (KFU, Kazan) Alexander Shengelaya (Tbilisi State University, Tbilisi) Jörg Sichelschmidt (Max Planck Institute for Chemical Physics of Solids, Dresden) Haruhiko Suzuki (Kanazawa University,

Kanazava) Murat Tagirov (KFU, Kazan) Dmitrii Tayurskii (KFU, Kazan) Valentin Zhikharev (KNRTU, Kazan)

In Kazan University the Electron Paramagnetic Resonance (EPR) was discovered by Zavoisky E.K. in 1944.

*

Magnetic properties and spin kinetics in Kondo lattices

S.I. Belov *, A.S. Kutuzov

Kazan Federal University, Kremlevskaya 18, Kazan 420008, Russia

*E-mail: [email protected]

(Received December 22, 2015; accepted December 26, 2015)

We present a theoretical model to describe unusual properties of Kondo lattices. The influence of the Kondo effect on the static magnetic susceptibility and electron spin resonance (ESR) parameters is studied in a simple molecular field approximation together with a scaling perturbative approach. Theoretical expressions well agree with the ESR and static magnetic susceptibility experimental data.

PACS: 72.15.Qm, 76.30.He, 75.30.Cr, 71.27.+a

Keywords: Kondo lattice, Kondo effect, ESR, static magnetic susceptibility

1. Introduction

The unexpected observation of the electron spin resonance (ESR) in the heavy fermion compounds YbRh2Si2 and YbIr2Si2 [1, 2] below the thermodynamically measured Kondo temperature TK [3, 4] (TK « 25 K for YbRh2Si2 and TK « 40 K for YbIr2Si2) stimulated a renewed interest in the theory of Kondo lattices. A series of theoretical approaches was proposed to understand the ESR signal existence from both a Fermi-liquid description [5, 6] and a picture of localized Kondo-ions moments [7-20]. The latter approach is supported by the ESR data on the resonance g-factor, linewidth and intensity which certainly reflect the tetragonal symmetry of the crystal electric field (CEF) at the Yb3+ ion position [1-2, 21-24]. The static magnetic susceptibility measurement also points out localized nature of f-electron motion in the CEF [11].

In a series of previous works [11, 12, 14-20] we investigated the static properties and spin kinetics in Kondo lattices at temperatures below TK basing on the entirely local properties of Yb3+ ions in the CEF. The theoretical expressions for the static magnetic susceptibility obtained in a simple molecular field approximation [11] give an excellent agreement with experimental data. In other works [12, 16-18] it was shown that the main reason of the ESR signal observability is the mutual cancelation of the large relaxation rates in the collective spin mode due to the Kondo effect. In this paper we give a brief review of the developed theory and present some new results on the ESR intensity and static magnetic susceptibility.

2. Basic Hamiltonian

We start from the atomic properties of an Yb3+ ion. A free Yb3+ ion has a 4f13 configuration with a single term 2F. The spin orbital interaction splits the 2F term into two multiplets: 2F7/2 with J = 7/2 and 2F5/2 with J = 5/2, where J denotes the value of the total momentum J = L + S with L and S as the orbital and spin momentum of the ion. Since the excited multiplet 2F5/2 is separated from the ground one 2F7/2 by the energy interval 1 eV, which is very large compared with the CEF energy, we consider in the following the ground multiplet only. Within the lowest multiplet the spin and orbital momentums of the ion are expressed via its total electronic momentum: S = (gj -1)J, L = (2 - gj)J, where gj = 8/7 is the Lande g-factor for the 2F7/2 multiplet.

The basic model includes the Zeeman energy, the Kondo interaction of Yb3+ ions with conduction electrons, the coupling between Yb3+ ions via conduction electrons and the kinetic energy of conduction electrons. The energy of Yb3+ ions in external magnetic field H can be written as follows

Zeeman = Mbgj X HJi . (1)

The Kondo exchange coupling of Yb3+ ions with conduction electrons and the indirect interaction between Yb3+ ions via conduction electrons (RKKY interaction) are also expressed in term of the total momentum J = (Jx, Jy, Jz):

HKondo = Mgj _ 1)! J O , (2)

i

hrkky = (gj -1)2!irkkyJiJj . (3)

j

Here A0 denotes the zero order term of the Kondo exchange integral expansion in multipoles [25-27], Oi is the spin density of the conduction electrons at the i-th ion site and I^KY denotes the constant of the RKKY interaction between two Kondo ions.

The tetragonal CEF splits the ground multiplet 2F7/2 into four Kramers doublets, each one described by the wave functions of the type y± = !c±m I ±M> [11, 14, 15], where M is the eigenvalue of

M

operator Jz (z-axis coincides with tetragonal axis of CEF). Within every Kramers doublet the total electronic momentum of the Yb3+ ion can be represented by the effective spin S = 12:

Jz =y Sz, Jx y =Y Sx y, (4)

where S is the effective spin operator, y and yj are given by

Y= 2{¥+IJzI¥+>, Yi= 2<^+|Jx|^_> . (5)

Since the first excited level is separated from the ground one by 17 meV (197 K) [28] and 18 meV (209 K) [29] in the cases of YbRh2Si2 and YbIr2Si2 the low temperature physics (T ■ 200 K) are well described by the lowest Kramers doublet. After projection onto the ground state the Zeeman energy, the Kondo interaction and the RKKY interaction take the form

H zs = Mb ![gj (HXSX + HySy) + glHzSz ], gj,,, = gJY^,,, (6)

i

Hsa=![_Jj(SX&X + S?v>) + JSZVZ] , Jj,,, = Jgjj,,,, J = A(gJ _ 1)1 gJ , (7)

i

Hss =![Ij(SXSX + S?SJ) + I{SZSZ] , j = (gj,n)2/j, Iij = /RKKY(gj _ 1)7gj2. (8) ij

The anisotropies of the Kondo and RKKY interactions are evidently related to that of the g-factor:

Ji/ J i=gj/g is, j1! =( gj/gi )2. (9)

Note, that the coordinate axes x, y, z in expressions (4)-(8) coincide with the crystallographic axes a, b, c. The kinetic energy of conduction electrons and their Zeeman energy can be written as

hc = ! tyCaCjx _ m! c+xc iX , (10)

ijX iX

Hza = mbg^! HOi . (11)

i

Here X = ±1 labels the orientation of the conduction electrons spin, m is the chemical potential, g° denotes the g-factor of the conduction electrons. The conduction electrons density is expressed in terms of the creation and annihilation operators

"» = X sxx' caca, (12)

xx

where sxx/ are the matrix elements of spin operators s = 1/2.

Finally, we represent the total Hamiltonian as H = H0 + Hint with

h0 = Hc + Hzs + hza, Hint = Hsa + Hss . (13)

3. Static magnetic susceptibility

At first we consider the effects of the Kondo and RKKY interactions on the static magnetic susceptibility and g-factors in a simple molecular field approximation. In this case the Zeeman energy of the Yb3+ ions and conduction electrons is renormalized and the total Hamiltonian (13) is reduced to

H = Hc + H zs + H za (14)

with

#Zs = Ub X H g %, Hza=Mb X H la0i. (15)

i i

The renormalized g-tensors gs and ga are expressed via the static magnetic susceptibility by the relations

gs = (1 - Xj r - x Xs) gs, ga=(1 - Xj Xs) ga, (16)

where

Xj , x = TU XI*, (17)

Njubga Njub i

J and IiJ are introduced in (7) and (8), N is the number of Yb3+ ions. The static magnetic susceptibility tensors are defined in a usual way:

x _d<Msa a

tap mp

H_0

_d<Mp

xap dHp

, ms _-ugsXS;, M*_->UB*, (18)

H_0

where (...) means the thermodynamical average and a,( = x,y,z.

In the molecular field approximation the conduction electrons and Yb3+ ions susceptibilities follow

a system of self-consistent equations

'(i+ X X0s ) Xs + X X0 s r_ ic°s,

1 s . ~a "Off

Xj XX XX + XX _ XX .

(19)

Here X°s and £0p are the static susceptibilities unaffected by the Kondo and RKKY interactions. The elementary calculations with the Hamiltonian H° leads to the relations xOPp _ & apXaa and

xOp _ ô' apX0a The non-zero comPonents x« _ Xyy _ Xi and XXz _ X\\s are the well-known Curie susceptibility and %0a is the usual Pauli susceptibility:

„0s _ C±,ll n0 N(Ub gl j|)2 11 _ T ' _ 4£B '

X°a_ 2N(Ubg")2P, (21)

where p denotes the conduction electrons density of states per lattice site at the Fermi surface.

The symmetry of x°s and x°a tensors implies the relations ^ = Sap%saa and xlp = SapXL for the solution of the system (19). The non-zero components of the Yb3+ ions susceptibility

xXx =XSyy = xl and xlz = X\\ follow the Curie-Weiss law

C

xl II =-— (22)

1 t+el,ll

with the Curie constant and Weiss temperature

Ci,|I = Clj, (1 -Jx"a), ^uI = Clj, (a,-Ajx°a). (23)

The conduction electrons susceptibility becomes anisotropic and temperature dependent as affected by the Kondo and RKKY interactions:

X\ | | = X

( n \

,0a

1 -Aj- ClJ

T

(24)

It is evident that the main contribution to the total susceptibility x = Xs + X° is given by the Yb3+ ions

susceptibility due to the small value of the ratio Jxs ~ pT.

Substituting (22)-(24) into (16) we obtain the g-factors of Yb3+ ions and conduction electrons which are also anisotropic and temperature dependent:

gl,I I = gl,i I (1 -AjX0a) + e , ggl,| | =.

T

'T-

1 -A^^^ J t+e

(25)

In previous works [11, 19, 20] we studied the static magnetic susceptibility taking into account the excited states of the CEF as well, which results in an additional temperature independent Van Vleck term. The Curie-Weiss susceptibility together with Van Vleck part is well fitted to experimental data on the static susceptibility [11]. On the other hand the measurements of the ESR intensity [22], which is proportional to the static susceptibility, show the Curie-Weiss law alone. If we are interested in the ESR study it is sufficient to consider the Zeeman-splitted ground doublet in resonance with the alternating magnetic field. The Curie-Weiss term can be interpreted as a resonant part of the total static susceptibility. Although the molecular field approximation provides a good agreement with experiment, the fitting parameters C and e turned to be different for low and high temperature [11]. This temperature dependence may indicate the Kondo effect which leads to an additional renormalization of the static susceptibility.

A perturbation theory might seem a first step to go beyond the molecular field approximation. However, the standard perturbation expansion in the Kondo interaction is not sufficient at low temperatures: the second order calculations reveal the logarithmic divergencies of the type ln (T/W),

where W is a conduction electron bandwidth. This trouble can be overcome by means of the Anderson's "poor man's scaling" method [30] which allows one to extend the lowest order perturbation result and effectively sum the leading order logarithmic terms. The idea of this approach is to take into account the effect of the high energy excitations on the low energy physics by a renormalization of coupling constants. The original Kondo interaction Hsa (7) is projected onto the low energy conduction electrons states yielding a Hamiltonian H'sa with new Kondo couplings Jl and J' [16-20]. The renormalized dimensionless parameters Ul= (pJl)' and U = (pj)' become temperature dependent:

Ul= U/sin (p, U\ = Ucot (p, (p = U ln(T/TGK), U = pJl - J2, (26)

where TGK denotes a characteristic temperature given as follows

tgk = w exp

—=■ arccos

U

(gi/gi)

(27)

(the abbreviation "GK" indicates the Kramers ground state). The quantities TGK and U are scaling invariants which do not change with renormalizing the Hamiltonian Hsa.

Using the "poor man's scaling" method we find the static magnetic susceptibilities and the g-factors renormalized by the high energy conduction electrons excitations:

C'

(XII)' = : Ci,"

T + i

(xji) =X

0a

'i,II

1 -a

C'

T

(28)

and

(gi ,II )' = gi ,II (1 -Xj X°a)

T

T+

i,II

where

with

Ci ,II = Ci,II Zi,II,

(gi,n )'=.

9i ,ii=&i,u zi,I

1 -a

C'

T+

(29)

(30)

(31)

Zi = 1 + U (1/sin (p0 - 1/sin (p), Z|| = 1 + U (cot (p0 - cot (p). Here p0 = (p(T = W) = arccos(g\\/gi); U and (p are introduced in (26).

Formally, the expressions (28) and (29) are the same as that obtained in the molecular field approximation but the Curie constants and Weiss temperature convert to temperature dependent functions. The new Curie and Weiss parameters decrease upon lowering temperature and their asymptotic behavior at T TGK exp(1/U) follows

Ci ,II = Ci,I

1+ PJi,ii -

1

ln(T/TGK)

1+PJ1,II-

1

ln(^TGK)

(32)

Such a reduction of the Curie and Weiss "constants" at low temperature is in a qualitative agreement with experimental data [11].

4. The ESR parameters

To study ESR we add to the Hamiltonian (13) the interaction of conduction electrons and localized moments with an external alternating magnetic field perpendicular to the static magnetic field:

H mw = UB gaX h mw «i +UB X[ gi ( SX + Kw Si ) + g^ SZ

(33)

where hmw = h0cos®i; h0 and a> are the amplitude and frequency of the microwave field, respectively. The ESR response due to the microwave magnetic field perturbation (33) is given by the transverse dynamical susceptibility

X(p) = -{{Mh\Mh)

(34)

where Mh = Msh + Ma is the total magnetic moment along the direction of the microwave field, (A | Bj) means the Fourier transform of a retarded Green function

(A|B)^ = -i jdt ([A(t),B])exp(i®t).

For the sake of simplicity we consider the case of the static magnetic field parallel to the crystal symmetry axis c (z), although the calculations are principally the same for an arbitrary orientation of the static and alternating fields to the crystallographic axes. In this case

X (®) = ZXap(®), a p = s,G

ap

(36)

with partial susceptibilities Xap(®):

Xss = -(Mb g± )2( S+ )), X - S

= -mB g lgG(( S -

Xgs = -mb g'±gG

Xgg=-(Mbg^)2i G

(37)

Here S and o are the spin operators of Yb3+ ions and conduction electrons, respectively;

S± = (Sx ±iSy)/V2, g± = (ox ±ioy)/V2.

The collective spin motion of conduction electrons and localized moments is described by a set of coupled equations

a„„ a,.

\f

\aos aoo j

\

X ss X sG \ X GS X GG j

(P.. 0 ^

0 P

GG

with

aaa=®-®a+2«a+2aL (a = s, g) asG=^jX±®s -g±esG, aGs =AjX°g®g-^t ^o

g ±

(38)

(39)

Pss =-X±®S , Pgg=-X0g®g.

Here a>s and a>G are the resonant frequencies of Yb3+ ions and conduction electrons, which include only the molecular field shifts due to the Kondo and RKKY interactions:

=

Mb g||

H,

®g = Mb g| |

H;

(40)

It

g,S and gG are defined in (17) and (25). The imaginary parts of kinetic coefficients rap = Imi^ap) represent the partial relaxation rates in the system of Yb3+ ions and conduction electrons and their real parts give additional shifts to corresponding resonant frequencies.

The kinetic coefficients r ss and rGG are the well-known Korringa and Overhauser relaxation rates (Yb3+ ions relax to the conduction electrons being in the thermodynamical equilibrium and vice versa). Two additional coefficients rs L and roL describe the relaxation of Kondo ions and conduction

electrons to the thermal bath ("lattice"). Finally, the coefficients rsG and rGS provide the coupling between the transvers magnetizations of Yb3+ and conduction electrons. This coupling is especially important if the relaxation rate of the conduction electrons toward the Kondo ions is much faster than to the lattice and the resonant frequencies are close to one another ("bottleneck" regime):

Too » G

(41)

The poles of the total susceptibility are determined by the condition aSSaGG - aSGaGS = 0 which leads to two complex roots. Their real parts represent resonant frequencies and their imaginary parts represent the relaxation rates. Under condition of the strong bottleneck regime (41) the imaginary parts of the resonant poles read as follows

Im(l) = 0 (42)

Im(®2) = riL +raL , ^ =raL ir^, ^ = T. (43)

r aa aa

The expressions (42) and (43) were investigated in detail in the works [12, 16-18]. In the presence of the Kondo effect the partial relaxation rates diverge upon lowering temperature to TGK. To the leading order in logarithmic terms

nT n

r = r =-—- r = r =----(44)

SS OS ln2 (TTk ), aa sa 2pln2 {T/Tgk ). ( ) The imaginary parts of the resonant poles take the form

n

^ = 2 , 2fT/T ), (45) 2pln (T/Tgk )

Im(®2) = rsL + 2pTraL +nTU4 ln2 (TTgk). (46)

One can see that the first pole corresponds to the individual subsystems motion with the ESR linewidth too large to be experimentally observable. On the contrary, the second pole describes the strongly coupled spin motion of conduction electrons and Yb3+ ions when the individual subsystems features are not important. In this case the effective relaxation rate Tcoll = Im(a2) is greatly reduced and determined mainly by the relaxation of Kondo ions to the lattice. The divergent parts of different kinetic coefficients cancel each other in the collective spin mode due to the existence of the common energy scale TGK regulating their temperature dependence at T ^ TGK.

A similar cancelation of the logarithmic terms of the kinetic coefficients Eap occurs for the

resonant frequency determined as a real part of the pole a2. The molecular field shifts arising from the Kondo interaction also disappear from the collective spin mode. However, the Knight shift due to RKKY interaction still takes place and the resonant frequency contains the Kondo anomaly term via the renormalized Weiss temperature. For the corresponding resonant g-factor g res = Re(®2)/jUB Hres we obtain

T

re^ _ s J (Al\

g|1 = g|1 T + q,(1 + pjl{ - U cot (P)' where and p are introduced in (23) and (26). In the absence of the RKKY interaction the resonant g-factor coincides with the usual g-factor of an Yb3+ ion in the tetragonal CEF.

Now we consider the ESR intensity, which is determined by integrating the absorbed power of the microwave magnetic field hmw. The latter is expressed via the transverse dynamical susceptibility by the relation

P = 1 h^aIm [x(® + /0)]. (48)

Among the partial susceptibilities xap(a) the main contribution to the total susceptibility is made by

the term xss (a) due to the negligibly small value of the ratio xa IXs ~ pT. In this approximation the elementary integrating gives

I ~ als xh, (49)

where ares is the microwave field frequency and Xh denotes the static magnetic susceptibility of Yb3+ Magnetic Resonance in Solids. Electronic Journal. 2015, Vol. 17, No 1, 15104 (9 pp.) 7

ions along the direction of the alternating field. Note, that the expression (49) is valid for arbitrary directions of the static and alternating magnetic fields (the condition H ' h is always fulfilled).

Since the microwave frequency is constant in the ESR experiment the temperature dependence of the ESR intensity is determined by the static susceptibility only. As a result we have the Curie-Weiss law with the temperature dependent Curie and Weiss parameters:

C'

1 , I I

where C'1 y and d'L y are defined by the expressions (30)-(32). Such a dependence qualitatively agrees with experimental data [22].

5. Summary

We theoretically investigated magnetic properties and spin kinetics of heavy fermion Kondo lattices YbRh2Si2 and YbIr2Si2 at temperatures well below thermodynamically measured Kondo temperature 1k. It turned out that the local properties of f-electrons in the CEF and the Kondo effect due to the strong interaction of Yb3+ ions with conduction electrons can explain many features of Kondo lattices with heavy fermions. It was shown that the Kondo effect plays an important part in the behavior of the Kondo lattice systems. The Kondo anomalies are manifested in the static magnetic susceptibility, the ESR linewidth, the resonant g-factor and the ESR intensity. The temperature dependence of these parameters obtained in the framework of proposed model well agrees with experimental data.

Acknowledgments

This work was funded by the subsidy allocated to Kazan Federal University for the state assignment in the sphere of scientific activities. A.S.K. is also thankful to the Russian Government Program of Competitive Growth of Kazan Federal University.

References

1. Sichelschmidt J., Ivanshin V.A., Ferstl J., Geibel C., Steglich F. Phys. Rev. Lett. 91, 156401 (2003)

2. Sichelschmidt J., Wykhoff J., Krug von Nidda H-A., Fazlishanov 1.1., Hossain Z., Krellner C., Geibel C., Steglich F. J. Phys.: Condens. Matter 19, 016211 (2007)

3. Trovarelli O., Geibel C., Mederle S., Langhammer C., Grosche F.M., Gegenwart P., Lang M., Sparn G., Steglich F. Phys. Rev. Lett. 85, 626 (2000)

4. Köhler U., Oeschler N., Steglich F., Maquilon S., Fisk Z. Phys. Rev. B 77, 104412 (2008)

5. Abrahams E., Wölfle P. Phys. Rev. B 78, 104423 (2008)

6. Wölfle P., Abrahams E. Phys. Rev. B 80, 235112 (2009)

7. Schlottmann P. Phys. Rev. B 79, 045104 (2009)

8. Huber D.L. J. Phys.: Condens. Matter 21, 322203 (2009)

9. Huber D.L. J. Phys.: Condens. Matter 24, 226001 (2012)

10. Huber D.L. Mod. Phys. Lett. B 26, 1230021 (2012)

11. Kutuzov A.S., Skvortsova A.M., Belov S.I., Sichelschmidt J., Wykhoff J., Eremin I., Krellner C., Geibel C., Kochelaev B.I. J. Phys.: Condens. Matter 20, 455208 (2008)

12. Kochelaev B.I., Belov S.I., Skvortsova A.M., Kutuzov A.S., Sichelschmidt J., Wykhoff J., Geibel C., Steglich F. Eur. Phys. J. B 72, 485 (2009)

13. Sichelschmidt J., Kambe T., Fazlishanov I., Zakharov D., Krug von Nidda H.A., Wykhoff J., Skvortsova A., Belov S., Kutuzov A., Kochelaev B., Pashchenko V., Lang M., Krellner C., Geibel C., Steglich F. Phys. Status Solidi B 247, 747 (2010)

14. Kutuzov A.S., Skvortsova A.M. Magn. Reson. Solids 11, 7 (2009)

15. Kutuzov A.S., Skvortsova A.M. J. Phys.: Conf. Ser. 324, 012039 (2011)

16. Belov S.I., Kutuzov A.S., Kochelaev B.I. J. Phys.: Conf. Ser. 324, 012017 (2011)

17. Belov S.I., Kutuzov A.S., Kochelaev B.I. Sichelschmidt J. J. Phys.: Condens. Matter 24, 365601 (2012)

18. Belov S.I., Kutuzov A.S.Magn. Reson. Solids 14, 12103 (2012)

19. Belov S.I., Kutuzov A.S.Magn. Reson. Solids 16, 14103 (2014)

20. Belov S.I., Kutuzov A.S. Appl. Magn. Reson. 45, 1179 (2014)

21. Sichelschmidt J., Wykhoff J., Krug von Nidda H.A., Ferstl J., Geibel C., Steglich F. J. Phys.: Condens. Matter 19, 116204 (2007)

22. Gruner T., Wykhoff J., Sichelschmidt J., Krellner C., Geibel C., Steglich F. J. Phys.: Condens. Matter 22, 135602 (2010)

23. Gruner T., Sichelschmidt J., Klingner C., Krellner C., Geibel C., Steglich F. Phys. Rev. B 85, 035119 (2012)

24. Sichelschmidt J., Jeevan H.S., Mchalwat M., Gegenwart P. Phys. Status Solidi B 250, 495 (2013)

25. Taylor K.N.R., Darby M.I. Physics of Rare Earth Solids, Chapman and Hall, London (1972)

26. Liu S.H. Phys. Rev. 121, 451 (1961)

27. Liu S.H. Phys. Rev. 123, 470 (1961)

28. Stockert O., Koza M.M., Ferstl J., Murani A.P., Geibel C., Steglich F. Physica B 378-380, 157 (2006)

29. Hiess A., Stockert O., Koza M.M., Hossain Z., Geibel C. Physica B 378-380, 748 (2006)

30. Anderson P.W. J. Phys. C 3, 2436 (1970)