Beyond Single Tetrahedron Physics of Breathing Pyrochlore Compound Ba 3Yb 2Zn5O11 Rabindranath Bag1Sachith E. Dissanayake1Han Yan2Zhenzhong Shi1David Graf3Eun Sang Choi3Casey Marjerrison1Franz Lang4Tom Lancaster5Yiming Qiu6

2025-04-24 0 0 4.46MB 16 页 10玖币
侵权投诉
Beyond Single Tetrahedron Physics of Breathing Pyrochlore Compound Ba3Yb2Zn5O11
Rabindranath Bag,1, Sachith E. Dissanayake,1, Han Yan,2Zhenzhong Shi,1David
Graf,3Eun Sang Choi,3Casey Marjerrison,1Franz Lang,4Tom Lancaster,5Yiming Qiu,6
Wangchun Chen,6Stephen J. Blundell,4Andriy H. Nevidomskyy,7and Sara Haravifard1, 8,
1Department of Physics, Duke University, Durham, NC 27708, USA
2Rice Academy of Fellows, Rice University, Houston, TX 77005, USA
3National High Magnetic Field Laboratory and Department of Physics,
Florida State University, Tallahassee, Florida 32310, USA.
4Clarendon Laboratory & Physics Department, University of Oxford,
Parks Road, Oxford OX1 3PU, United Kingdom
5Department of Physics, Centre for Materials Physics,
Durham University, Durham DH1 3LE, United Kingdom
6NIST Center for Neutron Research, National Institute of Standards and Technology, Gaithersburg, Maryland 20899, USA
7Department of Physics and Astronomy, Rice University, Houston, TX 77005, USA
8Department of Materials Sciences and Mechanical Engineering, Duke University, Durham, NC 27708, USA
(Dated: October 14, 2022)
Recently a new class of quantum magnets, the so-called breathing pyrochlore spin systems, have
attracted much attention due to their potential to host exotic emergent phenomena. Here, we present
magnetometry, heat capacity, thermal conductivity, Muon-spin relaxation, and polarized inelastic
neutron scattering measurements performed on high-quality single-crystal samples of breathing py-
rochlore compound Ba3Yb2Zn5O11. We interpret these results using a simplified toy model and
provide a new insight into the low-energy physics of this system beyond the single-tetrahedron
physics proposed previously.
Frustrated quantum magnets provide a fruitful arena
to search for novel quantum phenomena [1,2]. Py-
rochlore lattice magnets, in which magnetic ions form
corner-sharing regular tetrahedra, are one of the most
studied frustrated systems in three-dimension [36]. In
the pyrochlore system the conventional magnetic order-
ing is suppressed by the geometrically frustrated lat-
tice, consequently resulting in emergence of exotic phases
[719]. Recently a new class of systems, the so-called
breathing pyrochlore magnets, have attracted much at-
tention due to their potential to host exotic phenom-
ena and topological phases [2023]. In breathing py-
rochlore compounds the lattice inversion symmetry at
each site is broken due to the different sizes of up-
pointing and down-pointing tetrahedra, thus resulting
in large Dzyaloshinskii-Moriya (DM) interactions on the
two tetrahedra [24] (see Fig. 1(a,b) for the structure
of breathing pyrochlores). On the theory front, recent
works have shown that breathing pyrochlore spin systems
can host novel physics including classical rank-2 U(1)
spin-liquid states [25], quantum fractons [26], competing
quantum spin liquids [27], and hedgehog lattices of mag-
netic monopoles and antimonopoles [23]. Thus, it is of
great interest to synthesize and understand breathing py-
rochlore materials. The majority of the work performed
on the breathing pyrochlore-based compounds have fo-
cused on Cr-based spinels with S= 3/2 [2836], while
the studies performed on quantum systems with S= 1/2
remain limited to Ba3Yb2Zn5O11 in powder form [3742].
Our recent work reported the first comprehensive neutron
scattering studies performed on single-crystal sample of
Ba3Yb2Zn5O11 [43].
We successfully grew single crystal samples of breath-
ing pyrochlore Ba3Yb2Zn5O11 using the modified optical
floating zone technique [43]. Inelastic neutron scatter-
ing studies using our single crystal sample revealed that
the single-tetrahedron model with isolated tetrahedra
can explain the high-temperature and high-energy regime
of the collected data [43]. However, the diffuse neu-
tron scattering performed at low-temperature and low-
energy reveals features which cannot be understood with
this model [43]. Pair distribution function (PDF) anal-
yses performed on high quality powder neutron diffrac-
tion data provided evidence for the absence of chemical
disorder within experimental resolution. Single crystal
X-ray diffraction studies also found no evidence of site
disorder [43]. Diffuse neutron scattering on single crys-
tal samples [43], and the low temperature heat capacity
data collected on powder samples [39,42] suggest that
physics beyond the single tetrahedron, with small but
finite inter-tetrahedron interactions, is essential to cap-
ture the behavior of this quantum breathing pyrochlore
system. This calls for additional experimental probes
at low temperature with higher resolution, so that we
can study the subtle changes in magnetic properties of
Ba3Yb2Zn5O11 with higher accuracy.
In this letter, we report low-temperature heat capac-
ity measurements in applied field, ultra-sensitive mag-
netic susceptibility, thermal conductivity, muon spin re-
laxation (µ+SR), and polarized inelastic neutron scatter-
ing measurements of the ytterbium based breathing py-
rochlore compound Ba3Yb2Zn5O11 in single-crystalline
form, to investigate the intrinsic low temperature mag-
arXiv:2210.06534v1 [cond-mat.str-el] 12 Oct 2022
2
A(t)
t(µs)
λ(µs1)
T(K)
A(t)
t(µs)
(a)
(b)
(c) (d)
1.8 K
250 K
0.08 K
ZF
2 mT
FIG. 1. (a) Structure of the breathing pyrochlore lattice.
Considering each A-tetrahedron (blue) as a single site and
retaining the bonds on the B-tetrahedron (red), we obtain
a face-centered cubic (FCC) lattice shown in (b). The cube
in solid black line marks the repeating unit in the left panel
in the FCC lattice. (c) Zero-field muon asymmetry for two
temperatures. Solid lines represent fits to A(t) = A(0)e(λt)β
with A(0) fixed across all temperatures. (d) The temper-
ature dependence of the fitted relaxation rate λ. The inset
shows ultra-low temperature data, demonstrating the absence
of long range order at 0.08 K and the effect of applying a small
2 mT longitudinal field.
netic properties and provide a first look into the physics
governing the low-energy regime of this system. We pro-
pose a simplified model that captures the field depen-
dence of the heat capacity for lower field region well
and provides a scenario beyond the previously reported
single-tetrahedron physics, with finite inter-tetrahedron
coupling necessary to interpret the experimental results.
In order to search for any trace of magnetic order,
we performed µ+SR measurements using a large powder
sample of Ba3Yb2Zn5O11 on the GPS spectrometer at
the Swiss Muon Source at PSI, and also using co-aligned
single crystal samples of Ba3Yb2Zn5O11 mounted in a
dilution refrigerator at the MuSR spectrometer of the
ISIS Muon Source. In order to understand the origin
of the contributions to the µ+SR signal, we carried out
density functional theory (DFT) calculations to locate
the most probable muon stopping sites, and assess the
degree of perturbation the muon-probe causes in the ma-
terial [44], which we find to be small in this system (for
details see [45]). The measured muon asymmetry A(t)
in zero field (ZF) exhibits a non-oscillating but relaxing
time dependence [see Fig. 1(c)], the relaxation rate of
which decreases on warming above around 50 K. This
relaxation can be fitted to a stretched exponential form
A(t) = A(0)e(λt)βwhere the exponent β0.33, indi-
cating that the spin fluctuations that give rise to the re-
laxation have a range of timescales (an exponent of β= 1
would indicate a single fluctuation time, and exponents
less than one are often found in systems with complex
spin dynamics, see e.g. [46]). The temperature depen-
dence of λis plotted in Fig. 1(d), illustrating the decrease
in λwith increasing temperature at high temperatures,
corresponding to the thermally-induced increase in spin-
fluctuation rate (λν1in the fast-fluctuation limit,
where νis a characteristic fluctuation rate [47]). There
is a maximum in λaround 35 K, the low-temperature
approach to which we attribute to the switching on of
excitations associated with the intra-tetrahedral interac-
tions (note that the energy gap of 0.38 meV [40] to the
next crystal field level is larger than any thermal energy
in this experiment). Our experiments at milliKelvin tem-
peratures were hampered by the very small size of the sin-
gle crystals (leading to a small relaxation amplitude) but
nevertheless demonstrate that the relaxing signal persists
down to 0.08 K [see inset to Fig. 1(d)], with no appear-
ance of an oscillatory signal, ruling out the development
of long range magnetic order. It is noteworthy to add
that the ultra-low temperature persistent slow dynamics
are consistent with what is seen in many other frustrated
systems, and in this case one can argue that they might
be connected with the very weak inter-tetrahedra inter-
actions. We will explore such possibility in more details
later as we discuss the ultra-low temperature thermody-
namics results.
We show in Fig. 2(a-i) the magnetic heat capacity
for a Ba3Yb2Zn5O11 single crystal sample from 54 mK
to 1 K under different applied magnetic fields, with
the phonon contribution subtracted using the results
of measurements made on iso-structural, non-magnetic
Ba3Lu2Zn5O11.The magnetic entropy at zero field is
shown in Fig. 2(j). The heat capacity data were col-
lected on two different Ba3Yb2Zn5O11 single-crystal sam-
ples (grown using different techniques) and are compared
with the reported powder Ba3Yb2Zn5O11 sample [45].
Previous reports discussed the possibility of having de-
fects, such as structural disorder, as an underlying cause
for the peak observed at low temperatures, whereas our
heat capacity data collected on multiple single crystal
samples excludes the existence of measurable defect ef-
fects such as structural disorder. To further elaborate on
this, we show in Fig. ?? of the Supplementary Materials
[45] the results obtained for two single-crystal samples
grown with different techniques (sample 1 and sample
2). The peak position at 110 mK remains the same for
both single-crystal samples and agrees with the reported
powder study by Haku et al. [39]. This is while the
fit to the data is significantly improved using the model
we employed to analyze the data. We explain the de-
tails of this model in the following. Additionally, here we
show the field-dependence of the low-temperature fea-
ture which agrees reasonably with our proposed model,
in particular for the low and high field region.
As shown in Fig. 2(j), there is 1
4Rln 2 entropy release
per Yb ion, corresponding to an effective pseudo-spin-1/2
degree of freedom on each tetrahedron. As discussed by
Rau et al. [42] the experimental specific heat results col-
lected at T < 0.4 K disagree with the single tetrahedron
model, leading us, to propose that this release of entropy
3
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0
0 . 0
0 . 4
0 . 8
1 . 2
1 . 6
Sm( T ) ( J / m o l - Y b - K )
T ( K )
R / 4 l n ( 2 )
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0
0 . 0
0 . 2
0 . 4
Cm ( J / m o l - Y b - K )
T ( K )
1 2 T
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0
0 . 0
0 . 4
0 . 8
1 . 2
1 . 6
Cm ( J / m o l - Y b - K )
T ( K )
1 0 T
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0
0 . 0
0 . 4
0 . 8
1 . 2
1 . 6
2 . 0
Cm ( J / m o l - Y b - K )
T ( K )
8 T
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0
0 . 0
0 . 4
0 . 8
1 . 2
1 . 6
2 . 0
Cm ( J / m o l - Y b - K )
T ( K )
5 T
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0
0 . 0
0 . 2
0 . 4
0 . 6
0 . 8
1 . 0
1 . 2
1 . 4
Cm ( J / m o l - Y b - K )
T ( K )
4 T
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0
0 . 0
0 . 2
0 . 4
0 . 6
0 . 8
1 . 0
1 . 2
Cm ( J / m o l - Y b - K )
T ( K )
3 T
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0
0 . 0
0 . 2
0 . 4
0 . 6
0 . 8
1 . 0
1 . 2
Cm ( J / m o l - Y b - K )
T ( K )
2 T
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0
0 . 0
0 . 2
0 . 4
0 . 6
0 . 8
Cm ( J / m o l - Y b - K )
T ( K )
1 T
( j )( i )( f ) ( g ) ( h )
( e )( d )( c )( b )
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0
0 . 0
0 . 2
0 . 4
0 . 6
Cm ( J / m o l - Y b - K )
T ( K )
0 T
( a )
FIG. 2. (a)-(i) Low temperature magnetic specific heat (Cm) of Ba3Yb2Zn5O11 single crystal samples for different fields from
0 T to 12 T. The phonon contribution was subtracted using the iso-structural, non-magnetic compound Ba3Lu2Zn5O11. Red
lines are fits based on the simplified model outlined in the text [45]. (j) Magnetic entropy of Ba3Yb2Zn5O11 at zero magnetic
field.
is related to the inter-tetrahedron interactions. This is
because in the single tetrahedron theory, the two lowest
states are robustly degenerate, and the third state lies
much higher in energy (at 0.5 meV). Although for a
finite external magnetic field, the degeneracy of the two
lowest states is expected to be lifted, the energy split-
ting is much smaller than 0.01 meV, which cannot
explain the broad peak in heat capacity measurement in
Figs. 2(a,b).
The above observations suggest that the low-energy
properties of Ba3Yb2Zn5O11 cannot be explained by the
single tetrahedron theory, even if tuning the exchange
parameters is allowed. Instead the specific heat data
can be understood quantitatively by introducing inter-
tetrahedron interactions. To this end, we have con-
structed an effective low-energy model, regarding the two
nearly degenerate lowest energy states (out of 16) on the
Atetrahedra as a pseudo-spin 1
2. This is justified by
the fact that the other states lie at much higher energies
(E3>0.3 meV) [42] relative to the range T < 1 K in our
specific heat data.
From exact diagonalization on a single tetrahedron, we
can determine the wave-function of the two lowest states
exactly, which form the two-dimensional (E) irreducible
representation of the Tdpoint group. In the limit of
vanishing intra-tetrahedron DM interaction, these states
span the two-dimensional Hilbert space of two-dimer cov-
erings of the four sites. Note that there are three such
possible dimer coverings classically, but one of them is lin-
early dependent of the other two. The small DM interac-
tion in A-tetrahedra tunes the wave-function away from
the perfect dimer-covering states [38,48], but does not lift
their degeneracy, justifying our treating them as pseudo-
spin 1
2degrees of freedom. We then consider the interac-
tion between these pseudo-spins via the weak bonds on
the B-tetrahedra [shown in red in Fig. 1(a,b)]. Shrinking
every A-tetrahedron to a point connected by these weak
bonds, the effective low-energy model becomes an face-
centered cubic (FCC) lattice of pseudo-spins with nearest
neighbor interactions [c.f. Fig. 1(b)]. The effective inter-
actions are the original interactions between the physical
spins projected onto the pseudo-spin Hilbert space.
We consider the simplest model of the effective inter-
actions between neighboring A-tetrahedra pseudo-spins:
H=Jxy(sx
isx
j+sy
isy
j) + Jzsz
isz
j(1)
We find that choosing ferromagnetic Jz=0.005 meV
and antiferromagnetic Jxy = 0.0125 meV in this XXZ
model can reproduce the zero-field specific heat very well
[Fig 2(a)] (for details, see Supplementary Material [45]).
The origin of the 110 mK peak in the specific heat is
due to the ferromagnetic ordering of the pseudo-spins,
which spontaneously lifts the two-fold degeneracy of the
single-tetrahedron model. Linear spin-wave theory then
produces pseudo-magnons of the bandwidth Jxy which
propagate on the FCC lattice, as depicted in Fig. ??. It is
important to note that pseudo-magnons are not conven-
tional spin waves, but rather collective excitations of the
dimer-covering states spanned by pseudo-spin degrees of
freedom on A-tetrahedra, and hence may be challenging
to detect by inelastic neutron scattering.
The effective low-energy model in Eq. (1) is expected
4
( c )( b )
0 . 1 1
1 0 - 3
1 0 - 2
1 0 - 1
0 . 1 1
1 0 - 2
1 0 - 1 B a 3Y b 2Z n 5O1 1
kx x (W m - 1 K- 1 )
T( K )
α= 2 . 6 3
k ~ T
µ0
T | | [ 1 1 1 ]
H = 0 T
kx x / T (W m - 1 K- 2 )
T( K )
T | | [ 1 1 1 ]
H = 0 T
0 . 5 1 . 0 1 . 5
- 1 0
- 8
- 6
0 . 2 0 . 4 0 . 6 0 . 8 1 . 0 1 . 2 1 . 4 1 . 6 1 . 8
- 2 0
- 1 5
- 1 0
- 5
a c T D O
c' ( a r b . u n i t s )
0H( T )
1 T
f
0 . 3 2 T
a c
c' ( a r b . u n i t s )
0H( T )
T D O
0 . 0 2 4 K
f
0 . 6 6 0 K
0 . 2 3 3 K
0 . 0 7 4 K
0 . 8 8 0 K
0 . 4 5 2 K
0 . 3 1 7 K
0 . 1 9 0 K
0 . 1 4 3 K
0 . 1 0 6 K
0 . 0 4 1 K
0 . 0 2 4 K
( a )
α
µ0
FIG. 3. Low-Hanomalies beyond the single-tetrahedra model: (a) magnetic ac-susceptibility χ0(left axis) and TDO frequency
(f) (right axis) as a function of external magnetic field (Hk[111]) at different temperatures. Two anomalies are marked
by arrows. χ0is measured with an oscillating frequency of 1616 Hz. Measurements (χ0) with different frequency (87.1 Hz)
reproduce the two anomalies at the same fields. Inset shows zoomed-in plot of χ0(left axis) and TDO f(right axis) at T=
0.024 K. (b) TDO frequency (f) data is shown for different rotation angles between the field and the crystal [111] axis at 0.024
K. Traces are shifted vertically for clarity. (c) Thermal conductivity data (κxx vs T) for Ba3Yb2Zn5O11 at µ0H= 0 T; δT ||
[111]. Solid red lines are power-law fit to κxx data. A clear saturation of κxx at low Tis seen for Ba3Yb2Zn5O11. Inset shows
κxx/T vs Tplot.
to work in a moderately large applied magnetic field,
provided its strength does not exceed the energy gap
(E30.38 meV) to the first excited state beyond the E-
doublet in the single-tetrahedron model [42] . The mag-
netic field splits pseudo-spin degrees of freedom in Eq. (1)
at an energy scale much smaller than the parameters in
the effective model (Fig. S3 in [45]). The main effect of
the field, from the exact diagonalization of a single tetra-
hedron, is to shift the higher-energy states downwards,
which we treat as flat bands. This approximation breaks
down at a critical value of the field Bc4 T when the
lowest excited state E3crosses the ground state doublet,
resulting in a phase transition. Our model (1) does not
apply in this regime or higher fields. In in the limit of
high fields B > 10 T, we are able to obtain a good match
with the experimental specific heat [see Fig. 2(i)] by using
a single tetrahedron theory, which predicts a unique non-
degenerate ground state separated by a large gap from
the higher-lying states. For intermediate field strengths,
one cannot ignore the effect of the excited states, which
result in the ground state level crossing as already noted.
The minimal model then becomes rather complicated,
with a vast range of unknown parameters, whose deter-
mination lies beyond the scope of the present work.
Further insight into the effect of weak magnetic fields
can be gleaned from the magnetic ac-susceptibility, which
we measured on two separate Ba3Yb2Zn5O11 crystals.
The results are shown in Fig. 3(a), which show two
anomalies at µ0Hc1= 0.32 T and µ0Hc2= 1.0 T upon
cooling at low temperatures T'0.3 K. The two anoma-
lies are also seen in the tunnel diode oscillator (TDO)
measurements on the same crystals and do not shift ap-
preciably with field when the crystal is rotated away from
[111] orientation [see Fig. 3(b)]. These anomalies are in-
dependent of two different oscillation frequencies (1616
Hz and 87.1 Hz) of the ac-susceptibility measurement,
suggesting the signal is unrelated to spin freezing and
consistent with the µ+SR study showing absence thereof.
The most likely explanation for the anomalies is the level
crossing at the corresponding fields µ0Hc1and µ0Hc2.
Given the high frustration of the FCC lattice, it is possi-
ble for the system to go through several different phases.
The exact phases and phase transitions cannot be de-
termined by current experimental data, and await fu-
ture effort. Importantly, the explanation must involve
inter-tetrahedron couplings, because the single tetrahe-
dron model would predict a nonmagnetic Seff = 0 ground
state at low temperatures (0.5 K) and thus a feature-
less magnetic susceptibility, contrary to what we see in
our measurements.
In order to further understand the nature of the low-
lying states observed in the heat capacity and magnetom-
etry data, low temperature thermal conductivity mea-
surements were carried out on single crystal sample of
Ba3Yb2Zn5O11 [45]. A power law fit (κxx Tα) is per-
formed on the collected data at the high-Tregion and
the value of the exponent (α) is found to be 2.63. For
a nonmagnetic insulator, κxx at very low temperature is
only due to the contribution from phonons [49,50]. How-
ever, the exponent value obtained for Ba3Yb2Zn5O11 re-
veals additional contributions coming from various quasi-
particles such as phonons, spinons and magnons, as well
as different scattering channels for the heat current [50
53]. Interestingly, at low T, a clear saturation of κxx
in Ba3Yb2Zn5O11 is seen. Such saturation is not ex-
pected in conventional magnets where both phonons
and magnons freeze out, but suggest itinerant fermionic
(spinon) excitations expected in gapless spin liquids can-
摘要:

BeyondSingleTetrahedronPhysicsofBreathingPyrochloreCompoundBa3Yb2Zn5O11RabindranathBag,1,SachithE.Dissanayake,1,HanYan,2ZhenzhongShi,1DavidGraf,3EunSangChoi,3CaseyMarjerrison,1FranzLang,4TomLancaster,5YimingQiu,6WangchunChen,6StephenJ.Blundell,4AndriyH.Nevidomskyy,7andSaraHaravifard1,8,y1Departmen...

展开>> 收起<<
Beyond Single Tetrahedron Physics of Breathing Pyrochlore Compound Ba 3Yb 2Zn5O11 Rabindranath Bag1Sachith E. Dissanayake1Han Yan2Zhenzhong Shi1David Graf3Eun Sang Choi3Casey Marjerrison1Franz Lang4Tom Lancaster5Yiming Qiu6.pdf

共16页,预览4页

还剩页未读, 继续阅读

声明:本站为文档C2C交易模式,即用户上传的文档直接被用户下载,本站只是中间服务平台,本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私,请立即通知玖贝云文库,我们立即给予删除!
分类:图书资源 价格:10玖币 属性:16 页 大小:4.46MB 格式:PDF 时间:2025-04-24

开通VIP享超值会员特权

  • 多端同步记录
  • 高速下载文档
  • 免费文档工具
  • 分享文档赚钱
  • 每日登录抽奖
  • 优质衍生服务
/ 16
客服
关注