Magnetic properties of ErxY1-xF3 solid solutions

Article abstract of DOI:10.1016/j.ssc.2004.10.023

The Er_xY_1-x F₃ (x=0.1, 0.2, 0.7, 0.9, 1) solid solutions were synthesized and characterized by X-ray powder diffraction and magnetic measurements. The crystal structure refinements done by the Rietveld profile method show that no significant change of the structure parameters with the erbium concentration occurs. On the basis of DC susceptibility measurements in the 2-300 K range the lowest four crystal field levels have been determined, giving the ground level magnetic moment value of 6.7 μ_B. Results of M(H) studies point to the presence of complex exchange interactions between erbium ions.

Full text of DOI:10.1016/j.ssc.2004.10.023

Solid State Communications 133 (2005) 157–161
www.elsevier.com/locate/ssc
Magnetic properties of ErxY1KxF3 solid solutions
a,
Jovan Blanusa , Miodrag Mitric , Vladan Kusigerski , Vojislav Spasojevic ,
a
a
a
*
b
Zvonko Jaglicic , Janez Pirnat , Zvonko Trontelj
b
ˇ ´b
aCondensed Matter Physics Laboratory, The Vinca Institute, P.O. Box 522, 11001 Belgrade, Serbia and Montenegro
bInstitute of Mathematics, Physics and Mechanics, Jadranska 19, 1000 Ljubljana, Slovenia
¨
Received 6 September 2004; received in revised form 18 October 2004; accepted 19 October 2004 by H. von Lohneysen
Available online 5 November 2004
Abstract
The ErxY1KxF3 (xZ0.1, 0.2, 0.7, 0.9, 1) solid solutions were synthesized and characterized by X-ray powder diffraction and
magnetic measurements. The crystal structure refinements done by the Rietveld profile method show that no significant change
of the structure parameters with the erbium concentration occurs. On the basis of DC susceptibility measurements in the 2–
300 K range the lowest four crystal field levels have been determined, giving the ground level magnetic moment value of
6.7 mB. Results of M(H) studies point to the presence of complex exchange interactions between erbium ions.
q 2004 Elsevier Ltd. All rights reserved.
PACS: 61.10Ki; 71.30.E; 71.71.Ch
Keywords: A. Rare earths fluorides; B. Chemical synthesis; D. Crystal and ligand fields; D. Magnetic properties
1. Introduction
corresponding fluorides were found to have several possible
kinds of magnetic ordering. This, to some extent unusual
behavior, is believed to be due to the sensitive balance
between weak exchange interactions and dipol–dipol
interactions, which can even be affected with the method
of sample preparation, as it was found in the case of HoF3
[9]. Although, some of the investigations made so far
revealed an interesting result about their magnetic behavior,
it is still little known on that subject since very few systems
have been fully analysed. This is probably due to the
difficulties with the preparation of pure samples and, in
some cases, their high sensitivity to the moisture and oxygen
contamination.
We report the results of the crystal structure, magnetic
susceptibility and magnetization study on the ErxY1KxF3
(xZ0.1, 0.2, 0.7, 0.9, 1) solid solutions. In few previous
papers [1–3] we have discussed the crystal field influence on
the magnetic susceptibility of rare-earth Yb3C, Tm3C and
Er3C ions in the oxygen surroundings, i.e. in cubic
sesquioxides. In contrast to these compounds, which have
two available cation sites with a different local symmetry,
the lanthanide trifluorides REF3 (RE—rare earth) have only
one low symmetry site in both possible structure types, i.e.
in orthorhombic YF3 and hexagonal LaF3 type [4].
It is also known that some of the YF3 type rare-earth
trifluorides undergo a magnetic phase transition at very low
ordering temperatures [5–9]. Opposite to the sesquioxides
that are all exclusively antiferromagnets [10], the
The magnetic structure of ErF3 is determined from
neutron diffraction [7], as a three-dimensional antiferro-
´
magnetic ordering with the Neel temperature of 1.05 K. To
our best knowledge, no EPR, spectroscopic, magnetization
or wide temperature range susceptibility measurements have
been done on this compound yet. The aim of this work is to
provide new experimental data for both ErF3 and ErxY1KxF3
solid solutions.
* Corresponding author. Tel.: C381 11 8065828; fax: C381 11
8065829.
E-mail address: blanusa@vin.bg.ac.yu (J. Blanusa).
0038-1098/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ssc.2004.10.023

158
J. Blanusa et al. / Solid State Communications 133 (2005) 157–161
Table 1
The refined structure parameters of ErxY1KxF3 solid solutions: a, b, c are the cell parameters; x, y, z are the ion fractional coordinates; B is the
overall temperature factor; D is the mean Er–F distance and RB is the integral Bragg factor [12]
x
0.1
0.2
0.7
0.9
1
˚
a (A)
˚
b (A)
˚
c (A)
6.3546(3)
6.8531(3)
4.3957(2)
0.3669(3)
0.0585(3)
0.525(1)
0.595(2)
0.164(1)
0.0627(7)
0.374(1)
0.42(4)
6.3531(3)
6.8523(3)
4.3942(2)
0.3669(2)
0.0598(3)
0.523(1)
0.597(2)
0.161(1)
0.0646(6)
0.374(1)
0.48(4)
6.3492(3)
6.8479(3)
4.3909(2)
0.3677(2)
0.0594(3)
0.527(1)
0.594(2)
0.161(1)
0.0631(7)
0.372(1)
0.78(3)
6.3463(3)
6.8465(3)
4.3903(2)
0.3683(2)
0.0589(3)
0.527(1)
0.596(2)
0.161(1)
0.0646(6)
0.371(1)
0.93(4)
6.3452(3)
6.8457(3)
4.3895(2)
0.3685(2)
0.0590(3)
0.526(1)
0.595(2)
0.163(1)
0.0649(6)
0.371(1)
0.98(3)
Er (4c)
F (4c)
F (8d)
x
z
x
z
x
y
z
2
˚
B (A )
RB (%)
5.7
5.2
3.7
4.5
4.1
˚
D (A)
2.321(2)
2.319(2)
2.317(2)
2.316(2)
2.316(2)
The ionic radii of Y3C and Er3C ions are very similar, so
within the set of ErxY1KxF3 solid solutions one should not
expect a significant change of structural parameters and
consequently neither the change of the crystal field
influence. This fact can be used to separate the crystal
field influence from the exchange interactions one, which
should simplify the interpretation of the magnetic measure-
ments and make the crystal field analysis more precise and
reliable. On the other hand, the competition between long
range (ferromagnetic) dipol–dipol interaction and the short
range (antiferromagnetic) exchange interaction is expected
to be highly sensitive on the erbium concentration.
Measurements of magnetization vs. field where done after
zero field cooling from the room temperature to 2 K and the
subsequent cycles at higher temperatures were carried out in
the same way. To prevent the grains reorientation in the
magnetic field the sample powder were fixed using epoxy
resin.
3. Results and discussion
Both ErF3 and YF3 crystallize in orthorhombic structure
(space group Pnma) in which cations are placed into 4c
Wychoff position (x, 1/4, z), while the fluorine ions are in 4c
(x, 1/4, z) and 8f general position (x, y, z). The X-ray powder
patterns of the solid solutions were refined by the full profile
Rietveld method [11], using the Fullprof program [12] and
by taking the starting structural parameter values from [7].
The refinement results indicate expected orthorhombic
crystal structure of YF3 type for all samples, without
noticeable presence of oxides, oxy-fluorides or any other
phase.
2. Experimental
The samples were prepared starting from the Er2O3 and
Y2O3 (99.9%) oxides, which were mixed in appropriate
stoichiometric ratios and dissolved in the boiling chloric
acid. The concentrated HF acid (50%) was added to the clear
solutions. The fluoride precipitates were washed in
concentrated HCl, evaporated almost to dryness and
subsequently dried and heated in vacuum. The heating
temperature was first slowly raised to 100 8C within the 5 h
period to eliminate the remains of water and acid and than
elevated to 600 8C where it was kept constant during the
next 6 h. The so-obtained samples should be kept in dry
boxes, since, they appear to be highly hygroscopic.
The X-ray powder diffraction measurements were
performed on a Philips 1050 X-ray powder diffractometer
using Ni-filtered Cu Ka radiation and Bragg–Brentano
focusing geometry. The patterns were taken in the 20–
1108 2q range with the step of 0.028 and exposure time of
15 s per step.
The refined values of some most important structural
parameters are summarized in Table 1. As should be
expected, there is no considerable difference between the
sample parameters regarding the erbium concentration.
Although, there seems to be a small tendency of cell
parameters reduction with increasing x, this change lies
within the range of estimated errors (the values in
parenthesis given in Table 1. are actually standard
deviations, the errors being obtained by multiplying these
deviations by approximately three [12]). The relative
change of cell volume is only about 0.4% and the mean
cation–anion distance D shows almost no variation,
accordingly.
Magnetic measurements were performed using the
quantum design MPMS-XL-5 squid magnetometer. The
magnetic susceptibility was measured in 2–300 K tempera-
ture range and at applied magnetic field of 100 Oe.
As can be seen from Fig. 1, the temperature dependence
of the inverse molar susceptibility deviates from Curie–
Weiss law below 30 K for all samples. The values of erbium
effective magnetic moment meffZ3CMkB/(NAm2Bx) obtained

J. Blanusa et al. / Solid State Communications 133 (2005) 157–161
159
(1)
KWTi
P
7
NAg2Jm2B
iZ0 Mi2eff
e
KWTi
cM
Z
P
7
kT
e
iZ0
where NA, gJ, k and mB have their usual meaning and Wi is
the energy gap between the ground and i-th crystal field
level. Mieff denotes linear combinations of the free ion
magnetic quantum numbers MJ:
X
2
Mieff
Z
jaiM j MJ
(2)
J
MJ
where aM i are coefficients from the linear expansion of the
i-th level wave function over the free ion states in the
magnetic field:
J
J
X
jii Z
aiM jMJi
(3)
Fig. 1. Inverse molar susceptibilities of ErxY1KxF3 vs. temperature.
The lines represent the fit of Eq. (1) to the experimental data
(*Values for xZ1 sample are multiplied by 0.5 for better visibility).
J
KJ
By fitting Eq. (1) to the experimental data we have
determined the positions and the magnetic effective
numbers of four lowest crystal field levels of erbium.
These values are listed in Table 3. Since there is a good
agreement between so-obtained values (within the estimated
errors) for different x, their mean values are also given in the
last column of Table 3. According to Eq. (1) the mean value
M20Z10.7 gives the magnetic moment of Er3C ions at the
ground crystal field level to be 6.7(1) mB, which is in very
good agreement with the value of 6.67(9) mB obtained from
the neutron diffraction experiments [7].
from the high temperature region (100–300 K) are close to
the free ion value of 9.6 mB (Table 2). The susceptibility
behavior at low temperatures (in the region 30–2 K) can be
seen in Fig. 2. The values of Curie–Weiss temperatures q are
negative and relatively high, considering that ErF3 has very
low ordering temperature of about 1 K [7]. Thus, it should
be treated as a behavior originating from the crystal field
influence, rather than the one from the weak exchange
interactions.
Measured magnetization vs. magnetic field data for all
samples are depicted in Fig. 3, where both concentration and
temperature dependency of the magnetization curves can be
easily noticed. For undiluted fluoride ErF3 typical para-
magnetic shape of M(H) curve was obtained, Fig. 3(a). The
same behavior was also found for highly diluted samples
xZ0.2 and 0.1, Fig. 3(d). However, samples with xZ0.9
In order to fit the experimental data within the whole
measured temperature range it appeared necessary to
account for the crystal field influence. In crystal fields
with symmetry lower than cubic, the Er3C ion ground term
4I15/2 splits into the eight Kramer’s doublets. In such a case
the magnetic susceptibility can be expressed as [1]:
Table 2
The values of CM, q and meff obtained from the fit to the Curie–Weiss law between 100 and 300 K
x
0.1
0.2
0.7
0.9
1
CM (emuK/mol)
q (K)
1.2022(2)
K16.7(2)
9.48(1)
2.522(2)
K12.1(1)
9.46(1)
7.879(5)
K11.9(1)
9.49(2)
10.363(3)
K14.4(2)
9.59(1)
11.241(2)
K11.5(1)
9.48(1)
meff (mB)
Table 3
Positions of the four lowest crystal field levels and their effective magnetic numbers for ErxY1KxF3
x
0.1
0.2
0.7
0.9
1
Mean value
M2
M2
M2
M2
10.9(1)
19.3(2)
36.2(2)
18.6(5)
20(1)
10.8(1)
20.8(3)
38(1)
10.5(2)
20.3(4)
37.5(6)
17(1)
11.2(3)
20.8(4)
40(3)
9.9(6)
21.1(3)
36.3(4)
17.7(7)
19(2)
10.7(3)
20.5(3)
38(1)
0
1
2
3
14.7(5)
22(1)
13(2)
16(1)
W1 (K)
W2 (K)
W3 (K)
18(2)
20(2)
20(2)
93(2)
108(4)
190(6)
91(4)
103(5)
180(8)
95(4)
98(4)
203(7)
183(7)
210(8)
193(7)

160
J. Blanusa et al. / Solid State Communications 133 (2005) 157–161
Fig. 2. Inverse molar susceptibilities of ErxY1KxF3 below 30 K. The full lines represent the fit of Eq. (1) to the experimental data (*Values for
xZ1 sample are multiplied by 0.5 for better visibility).
and 0.7 show irreversible behavior of the initial magnetiza-
tion curve at sufficiently low temperatures, Fig. 3(b) and (c).
As can be seen, a continuous increase up to the saturation
value of the initial curve (denoted with 1) occurs, followed
by the hysteretic curve (denoted with 2) during the
isothermal demagnetizing. The hysteresis curve is of the
field induced type with the zero coercitivity value at HZ0.
The magnetization curves remain indistinguishable at
negative H cycles, Fig. 3(c). The obtained hysteresis
strongly decreases with increasing temperature and dis-
appeares at TO5 K.
The observed M(H) behavior for xZ0.9 and 0.7 samples
Fig. 3. Magnetization vs. field data for ErxY1KxF3 samples.

J. Blanusa et al. / Solid State Communications 133 (2005) 157–161
161
can be correlated with the proposed description of the
exchange mechanism in rare-earth fluorides. Namely,
according to [9] magnetic properties of rare-earth fluorides
strongly depend on the balance between competing short-
range antiferromagnetic superexchange interaction and
long-range ferromagnetic dipol–dipol interaction. In case
of undiluted ErF3 compound these balance is preserved and
above magnetic critical temperature TO1 K it behaves like
common paramagnet. Magnetic dilution affects to the
greater extent the short-range interactions and consequently
long-range ferromagnetic interactions prevail. This enables
emergence of the field-induced ferromagnetic ordering in
the applied magnetic field at sufficiently low temperatures.
However, for large magnetic dilution, i.e. for xZ0.2 and 0.1
samples, long-range correlations become also ineffective
and system displays paramagnetic behavior again.
leads to the field-induced magnetic behavior of the diluted
samples with high erbium concentration, xZ0.9 and 0.7.
This behavior is in compliance with the findings made on
the HoF3 compound [9].
Acknowledgements
This work is an outcome of the joint project ‘Synthesis
and Physical Properties of Nanosize and Polycrystalline
Magnetic Materials’ financed by the Serbian Ministry of
Science and Environment and Slovenian Ministry of
Education, Science and Sport.
References
4. Conclusion
[1] M. Mitric, B. Antic, M. Balanda, D. Rodic, M. Napijalo,
J. Phys: Condens. Matter 9 (1997) 4103.
[2] J. Blanusa, M. Mitric, I. Felner, N. Jovic, I. Bradaric, J. Magn.
Magn. Matter 263 (2003) 295.
All the synthesized polycrystalline solid solutions of
ErxY1KxF3 (xZ0.1, 0.2, 0.7, 0.9, 1) crystallize in the same
orthorhombic space group Pnma, with the negligible change
of crystal structure parameters with concentration. This
significantly reduces contributions caused by the changes of
the crystal parameters and consequently simplifies the
interpretation of magnetic properties.
[3] J. Blanusa, M. Mitric, D. Rodic, A. Szytula, M. Slaski,
J. Magn. Magn. Matter 213 (2000) 75.
[4] A. Zalkin, D.H. Templeton, J. Am. Chem. Soc. 75 (1953)
2453.
[5] P.E. Hansen, R. Nevald, G. Guggenheim, Phys. Rev. B 17
(1978) 2866.
Magnetic susceptibility data show paramagnetic beha-
vior in the 2–300 K temperature range with the crystal field
influence visible in the low-temperature region below 30 K.
This enables determination of the four lowest crystal field
levels and their effective magnetic numbers by treating the
Er3C as Kramer’s ion in the low-symmetry crystal field. The
calculated value of the magnetic moment of 6.7(1) mB at the
ground crystal field level is in excellent agreement with the
value of 6.67(9) mB obtained from the neutron diffraction
data [7].
[6] M. Piotrowski, A. Murasik, Phys. Status Solidi A 89 (1985)
571.
¨
¨
[7] K. Kramer, H. Romstedt, H.U. Gudel, Eur. J. Solid State Inorg.
Chem. 33 (1996) 273.
[8] B. Bleaney, J. Gregg, R.W. Hill, M. Lazzouni, M. Leask,
M.R. Wells, J. Phys. C: Solid State Phys. 21 (1988) 2721.
[9] P.J. Brown, J.B. Forsyth, P.C. Hansen, M. Leask, R.C. Ward,
R. Wells, J. Phys.: Condens. Matter 2 (1990) 4471.
[10] D.J. Craik (Ed.), Magnetic Oxides, Part 1, Wiley, New York,
1975, p. 29.
The low-temperature M(H) measurements point to the
presence of both short-range antiferromagnetic and long-
range ferromagnetic interactions. Magnetic dilution
increases the role of the ferromagnetic interactions that
[11] H. Rietveld, J. Appl. Crystallogr. 2 (1969) 65.
[12] J. Carvajal, T. Roisnel, Collected Abstracts of the 8th
European Powder Diffraction Conference, Uppsala, Sweden
2002 (p. 189).