Stabilization of type-I antiferromagnetism in mixed-valence TmTe at P = 6 GPa

Article abstract of DOI:10.1016/S0304-8853(00)01134-3

Previous investigations of magnetic order in the pressure-induced mixed-valence state of TmTe have been extended to a maximum pressure of P = 6.0 GPa. At this pressure, the same type-I antiferromagnetic state as in TmSe at P = 0 is formed below T_N = 2.55K, with a comparable magnetic moment of μ = 1.9 ± 0.2 μ_B. The fact that this value is about 50% larger than at 5.4 GPa in the ferromagnetic phase suggests that the change in the magnetic structure is associated with an instability of the Tm 4f moment.

Full text of DOI:10.1016/S0304-8853(00)01134-3

Journalof Magnetism and Magnetic Materiasl 226 }230 (2001) 211}213
Stabilization of type-I antiferromagnetism in mixed-valence
TmTe at P"6 GPa
J.-M. Mignotꢀꢀ*, I.N. Goncharenkoꢀꢀꢁ, T. Matsumuraꢂ, T. Suzukiꢃ
ꢀLaboratoire Le&on Brillouin, CEA-CNRS, CEA/Saclay, F-91191 Gifsur Y vette, France
ꢁRussian Research Centre **Kurchatov Institut++, 123182 Moscow, Russia
ꢂDepartment ofPhysics, Tohoku Uni versity, Sendai 980-77, Japan
ꢃNational Research Institute for Metals, Tsukuba, Ibaraki 305-0047, Japan
Abstract
Previous investigations of magnetic order in the pressure-induced mixed-valence state of TmTe have been extended to
a maximum pressure of P"6.0 GPa. At this pressure, the same type-I antiferromagnetic state as in TmSe at P"0 is
formed below ¹ "2.55 K, with a comparable magnetic moment of ꢀ"1.9$0.2 ꢀ . The fact that this value is about
,
50% larger than at 5.4 GPa in the ferromagnetic phase suggests that the change in the magnetic structure is associated
with an instability of the Tm 4f moment. ꢀ 2001 Elsevier Science B.V. All rights reserved.
Keywords: Magnetic transition; Mixed valences; Neutron di!raction; High pressure; Pressure e!ect
TmTe belongs to the family of cubic (NaCl structure)
monochalcogenides TmX (X: S, Se, Te), which feature
unique magnetic and transport properties re#ecting the
instability of the Tm 4f shell [1] (for an extensive review
on these systems, see Ref. [2]). At ambient pressure,
this question, we have investigated the pressure}temper-
ature magnetic-phase diagram of TmTe using neutron
di!raction. In the "rst series of experiments reported in
Ref. [5], it was found that at P"2.7 GPa, i.e., just after
the MV regime sets in, the ordering temperature jumps to
TmTe is
a
rather conventionaldiveanl t magnetic
¹ "14 K and the structure becomes ferromagnetic with
!
semiconductor, with a band gap of about 350 meV from
opticaland transport measurements. The 4f ꢁꢂ level of
Tmꢃ> is located inside the gap, slightly below the bottom
of the empty 5d6s conduction band. The compound
a magnetic moment ꢀ "2.1$0.1 ꢀ . At higher pres-
2ꢄ
sure, ¹ decreases again to reach 4.5 K at 5.4 GPa, while
!
the ordered moment drops to 1.2 ꢀ . This pronounced
reduction of the magnetic moment was tentatively as-
orders antiferromagnetically [AF II, k"(ꢁ, ꢁ, ꢁ)] below
cribed to magnetic #uctuations, both in view of the an-
omalous scattering that develops in the ꢁ(¹) curves over
the same range of pressures [6] and because the ambi-
ent-pressure properties of TmSe had been previously
interpreted in terms of a Kondo-lattice state [7]. How-
ever, despite the rather high pressures achieved in those
experiments, evidence for the recovery of the AF I struc-
ture corresponding to the magnetic order in TmSe was
ꢃ ꢃ ꢃ
¹ +0.4 K [3], a rather low value which can be ascribed
,
to weak superexchange interactions. It was "rst shown
by Jayaraman et al. [4] that the transport gap closes with
pressure at a rate of approximately !210 meV/GPa,
leading to a MV state above P+2 GPa. The question
naturally arises of the relationship between this new
high-pressure phase and the well-known magnetic MV
state occurring in TmSe at ambient pressure. To answer
still eluding, and the pressure variation of ꢀ seemed to
2ꢄ
extrapolate to zero, rather than to the relatively large
moment (ꢀ"1.7$0.2 ꢀ [8] or 1.9 ꢀ [9]) reported for
the latter compound.
This prompted us to extend our neutron measure-
ments to a maximum pressure of 6.0GPa. The experi-
mentalconditions were the same as those previously used
*Corresponding author. Fax: #33-1-69-08-82-61.
E-mail address: jmignot@cea.fr, mignot@llb.saclay.cea.fr
(J.-M. Mignot).
0304-8853/01/$- see front matter ꢀ 2001 Elsevier Science B.V. All rights reserved.
PII: S 0 3 0 4 - 8 8 5 3 ( 0 0 ) 0 1 1 3 4 - 3

212
J.-M. Mignot et al. / Journal ofMagnetism and Magnetic Materials 226 }230 (2001) 211}213
duces magnetic scattering]. It is consistent with previous
results for TmSe, in which the moments were shown to lie
along one of the four-fold axes perpendicular to the
k vector [8]. Under the same assumption, we can
estimate the Tm magnetic moment at 1.5 K to be ꢀ
"
2ꢄ
1.9$0.2ꢀ . The calculation takes into account di!er-
ences in magnetic domain populations due to non-hy-
drostatic pressure components (p /p +2/3;p +0, z
V
W
X
being parallel to the axis of the pressure cell), as deduced
from the intensities of the 100, 101, and 110 re#ections. In
TmSe, the Neeltemperature at P"0 is ¹ "3.46 K in
,
the best stoichiometric samples [7], but it was found to
increase rather steeply with pressure [9] and the smaller
value found here for TmTe may correspond to TmSe upon
a slight expansion of the lattice (`negative pressurea).
The Neeltemperature or the new AF I phase at
6.0GPa is plotted on the magnetic phase diagram in
Fig. 2, together with the corresponding value for the
magnetic moment. The previous neutron data points for
the low-pressure AF II phase and the high-pressure
F phase are displayed as squares and circles, respectively.
The present observation of a magnetic phase transition at
2.55K for P"6.0 GPa is consistent, within uncertainties
in experimentalpressure scales, with the resistivity anom-
aly observed recently at about 2.5 K for P"5.5 Gpa
[10]. The steep rise in ꢁ(¹) following the onset of AF
I order in the latter measurements provides additional
evidence that the electronic and magnetic state achieved
in TmTe at these pressures is essentially the same as in
TmSe at P"0. However, the sudden positive jump in
Fig. 1. Temperature dependence of the peak intensity of the
10 0 magnetic re#ection in TmTe at P"6.0 GPa (measured on
cooling).
ꢀ
between 5.4 and 6.0 GPa, which is not re#ected by to
2ꢄ
a corresponding increase of the ordering temperature,
shows that the change in the magnetic structure with
pressure must be accompanied by a more subtle instabil-
ity of the 4f ground state, and therefore cannot be re-
duced to a simple competition between di!erent terms in
the exchange interactions.
Fig. 2. Pressure}temperature magnetic-phase diagram of TmTe
(left scale); closed symbols represent the ordering temperatures
for the AF II (᭿), F (ⅷ), and AF I (᭜)phases, respectively. Open
symbols represent the ordered magnetic moment ꢀ measured
2ꢄ
at ¹ +1.5K (right scale).
ꢄꢅꢆ
The authors wish to express their thanks to
A. Gukasov, P. Fouilloux and Th. Beau"ls for their
help during the experiments, and to M. Ohashi for
communicating his data prior to publication.
and the single-crystal material, prepared by induction-
melting of high-purity components inside a vacuum-
sealed tungsten crucible, had
a lattice constant
a +6.21 A, characteristic of stoichiometric TmTe. Al-
s
ꢄ
though the increase in pressure with respect to Ref. [5]
was rather limited, the data indicate that the system
switches to a completely di!erent type of magnetic struc-
ture. No ferromagnetic intensity is found superimposed
on the weak 1 1 1-type nuclear peaks down to the lowest
temperature but, on the other hand, satellites corre-
sponding to the AF I wavevector k"(1, 0, 0) are clearly
observed. From the temperature dependence of the peak
intensity for Q"(1, 0, 0) Fig. 1, we deduce a Neeltemper-
ature of 2.55$0.10K. The observation of a "nite inten-
sity for the 1 0 0-type re#ections implies that the magnetic
moment is not parallel to k [only the moment component
perpendicular to the scattering vector Q"(1, 0, 0) pro-
References
[1] E. Bucher, K. Andres, F.J. di Salvo, J.P. Maita, A.C.
Gossard, A.S. Cooper, G.W. Hull Jr., Phys. Rev. B 11
(1975) 500.
[2] P. Wachter, in: K.A. Gschneider Jr., L. Eyring, G.H.
Lander, G.R. Choppin (Eds.), Handbook of the Physics
and Chemistry of Rare Earths, Vol. 19 } Lanthanides-
Actinides: Physics } II, Elsevier Science, Amsterdam, 1994,
p. 177.
[3] Y. Lassailly, C. Vettier, F. Holtzberg, A. Benoit,
J. Flouquet, Solid State Commun. 52 (1984) 717.

J.-M. Mignot et al. / Journal ofMagnetism and Magnetic Materials 226 }230 (2001) 211}213
213
[4] A. Jayaraman, E. Bucher, D.B. McWhan, in: T.H. Henrie,
R.E. Lindstrom (Eds.), Proceedings of the Eighth Rare-
Earth Research Conference, Vol. 1, US GPO, Washington
DC, 1970, p. 333.
[7] P. Haen, F. Lapierre, J.-M. Mignot, R. Tournier, F. Hol-
tzberg, Phys. Rev. Lett. 43 (1979) 304.
[8] H. Bjerrum-M+ller, S.M. Shapiro, R.J. Birgeneau, Phys.
Rev. Lett. 39 (1977) 1021.
[5] P. Link, I.N. Goncharenko, J.-M. Mignot, T. Matsumura,
T. Suzuki, Phys. Rev. Lett. 80 (1998) 173.
[9] C. Vettier, J. Flouquet, J.-M. Mignot, F. Holtzberg,
J. Magn. Magn. Mater. 15-18 (1980) 987.
[6] T. Matsumura, T. Kosaka, J. Tang, T. Matsumoto,
N. Mori, T. Suzuki, Phys. Rev. Lett. 78 (1997) 1138.
[10] M. Ohashi et al., Physica B, (these Proceedings).