This paper was converted on www.awesomepapers.org from LaTeX by an anonymous user.
Want to know more? Visit the Converter page.

Strong magnetic correlations to 900 K in single crystals of the trigonal antiferromagnetic insulators SrMn2As2 and CaMn2As2

N. S. Sangeetha    Abhishek Pandey Present address: Department of Physics and Astronomy, Texas A&M University, College Station, Texas 77840-4242, USA    Zackery A. Benson    D. C. Johnston johnston@ameslab.gov Ames Laboratory and Department of Physics and Astronomy, Iowa State University, Ames, Iowa 50011, USA
(October 2, 2025)
Abstract

Crystallographic, electronic transport, thermal and magnetic properties are reported for SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} single crystals grown using Sn flux. Rietveld refinements of powder x-ray diffraction data show that the two compounds are isostructural and crystallize in the trigonal CaAl2Si2{\rm CaAl_{2}Si_{2}}-type structure (space group P3¯m1P\bar{3}m1), in agreement with the literature. Electrical resistivity ρ\rho versus temperature TT measurements demonstrate insulating ground states for both compounds with activation energies of 85 meV for SrMn2As2{\rm SrMn_{2}As_{2}} and 61 meV for CaMn2As2{\rm CaMn_{2}As_{2}}. In a local-moment picture, the Mn+2 3d53d^{5} ions are expected to have high-spin S=5/2S=5/2 with spectroscopic splitting factor g2g\approx 2. Magnetic susceptibility χ\chi and heat capacity measurements versus TT reveal antiferromagnetic (AFM) transitions at T=N120(2)T{\rm{}_{N}}=120(2) K and 62(3) K for SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}}, respectively. The anisotropic χ(TT)N\chi(T\leq T{\rm{}_{N}}) data indicate that the hexagonal cc axis is the hard axis and hence that the ordered Mn moments are aligned in the abab plane. The χ(T)\chi(T) data for both compounds and the Cp(T)C_{\rm p}(T) for SrMn2As2{\rm SrMn_{2}As_{2}} show strong dynamic short-range AFM correlations from TNT_{\rm N} up to at least 900 K, likely associated with quasi-two-dimensional connectivity of strong AFM exchange interactions between the Mn spins within the corrugated honeycomb Mn layers parallel to the abab plane.

pacs:
75.50.Ee, 74.70.Xa, 75.40.-s, 72.15.Eb

I Introduction

The body-centered tetragonal AM2X2AM_{2}X_{2} ternary compounds (AA = rare or alkaline earth, MM = transition metal, XX = Si, Ge, P, As, Sb) with the ThCr2Si2{\rm ThCr_{2}Si_{2}} structureJust1996 have generated tremendous interest in the scientific community due to their novel electronic and magnetic properties. Prominent among these is the iron-arsenide family of parent compounds AFe2As2A{\rm Fe_{2}As_{2}} (AA = Ca, Sr, Ba, Eu).Johnston2010 ; Stewart2011 ; Scalapino2012 ; Dagotto2013 ; Fernandes2014 ; Hosono2015 ; Dai2015 ; Inosov2016 ; Si2016 These materials are metallic and show nearly contiguous antiferromagnetic (AFM) spin-density wave and structural transitions at temperatures TT up to 200\sim 200 K. The suppression of these transitions by external pressure or chemical doping leads to superconductivity with bulk superconducting transition temperatures TcT_{\rm c} up to 56 K. It is believed that the FeAs layer as the conducting sheet in this structure plays a crucial role in the occurrence of superconductivity.Johnston2010 ; Stewart2011 ; Scalapino2012 ; Dagotto2013 ; Fernandes2014 ; Hosono2015 ; Dai2015 ; Inosov2016 ; Si2016 Hence, it is important to investigate other related materials with similar compositions and structures in the search for new superconductors and other novel phenomena. For example, SrNi2As2 (Tc=0.62T_{\rm c}=0.62 K, Ref. Bauer2008, ) and BaNi2As2 (Tc=0.7T_{\rm c}=0.7 K, Ref. Ronning2008, ) were both found to be superconductors.

On the other hand, SrCo2As2 (Refs. Pandey2013, ; Jayasekara2013, ; Wiecki2015, ) and BaCo2As2 (Refs. Sefat2009, ; Anand2014, ) are correlated metals with no structural, superconducting or long-range magnetic ordering transitions. From inelastic neutron scattering measurements, SrCo2As2 is found to exhibit strong AFM correlations at the same stripe wavevector as do the superconducting iron arsenides, which raises the interesting question of why SrCo2As2 is not a high-TcT_{\rm c} superconductor.Jayasekara2013 The reason has been suggested from NMR measurements to be that SrCo2As2 exhibits strong ferromagnetic (FM) spin correlations/fluctuations in addition to the AFM correlations and these compete with the AFM correlations that are the presumptive glue for superconductivity in these systems. Subsequent NMR studies indicated that the large range of TcT_{\rm c} observed within the FeAs-based systems may also arise from the competition between FM and AFM correlations.Wiecki2015b

Recently significant attention has focussed on Mn arsenides. Our studies of the properties of the parent and doped BaMn2As2 systems were originally motivated by their potential to be ThCr2Si2-type high-TcT_{\rm c} superconductors analogous to the cuprates. The semiconductor BaMn2As2 shows G-type (checkerboard-type) local-moment collinear AFM order below its high Néel temperature TN=625T_{\rm N}=625 K with the ordered moments aligned along the tetragonal cc axis.Singh2009 ; Singh2009b ; Johnston2011 Thus magnetoelastic coupling does not cause a distortion of the crystal structure below TNT_{\rm N}, contrary to the orthorhombic distortion associated with AFM ordering in the AFe2As2A{\rm Fe_{2}As_{2}} compounds due to the collinear Fe ordered moments aligned in the abab plane. An optical gap of 48 meV was inferred for BaMn2As2 from the optical conductivity,Antel2012 consistent with results from the electrical resistivity ρ\rho versus temperature TT measurements in the abab plane.Singh2009 Furthermore, this optical studyAntel2012 found that BaMn2As2 is much more two-dimensional in its electronic properties than are the AFe2As2A{\rm Fe_{2}As_{2}} parent compounds.Singh2009 A neutron scattering study of isostructural BaMn2Bi2{\rm BaMn_{2}Bi_{2}} found the same G-type AFM structure as in BaMn2As2{\rm BaMn_{2}As_{2}} but with a lower TN=387T_{\rm N}=387 K.Calder2014

Only 1.6% K substitution for Ba transforms BaMn2As2 into a local-moment AFM metal.Pandey2012 ; Yeninas2013 Higher doping levels lead to the onset of FM at 16\approx 16% K-dopingBao2012 and half-metal FM behavior below the Curie temperature TC100T_{\rm C}\sim 100 K at 40% K doping (Refs. Pandey2013b, ; Ueland2015, ) and at 60% Rb doping.Pandey2015 The FM is thus thought to be associated with FM ordering of the itinerant doped-hole spins and coexists with the G-type AFM order of the local Mn moments with TN>300T_{\rm N}>300 K.Ueland2015 ; Lamsal2013

Unlike BaMn2As2 with the tetragonal ThCr2Si2{\rm ThCr_{2}Si_{2}} structure, the compounds SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} both crystallize in the trigonal CaAl2Si2{\rm CaAl_{2}Si_{2}}-type structureMewis1978 ; Brechtel1978 containing a corrugated honeycomb Mn sublattice which can be viewed as a triangular lattice bilayer. The possibility of geometrically-frustrated triangular-lattice exchange connectivity exists and such compounds often show novel physical behaviors associated with the geometric frustration.Ramirez1994 ; Moessner2006 ; Balents2010 Single crystals of SrMn2As2{\rm SrMn_{2}As_{2}} were grown previously using Sn flux.Wang2011 These authors’ in-plane electrical resistivity ρ(T)\rho(T) measurements indicated that the ground state is insulating with activation energies of 0.29–0.64 eV depending on the TT range, and their magnetic susceptibility χ(T)\chi(T) measurements indicated an AFM transition at TN=125T_{\rm N}=125 K.Wang2011

Two neutron powder diffraction studiesRatcliff2009 ; Bridges2009 of the related CaAl2Si2{\rm CaAl_{2}Si_{2}}-type CaMn2Sb2{\rm CaMn_{2}Sb_{2}} revealed AFM ordering below TN=88T_{\rm N}=88 K and 85 K, respectively, with an AFM propagation vector 𝐤=(0,0,0){\bf k}=(0,0,0), i.e., the crystal and AFM unit cells are the same. In the former paper the AFM structure was deduced to be collinear, with the ordered moments aligned in the abab plane with a low-TT ordered moment of 2.8(1) μB\mu_{\rm B}/Mn, where μB\mu_{\rm B} is the Bohr magneton. In the latter paper, a model was favored with the ordered moments canted at ±25\pm 25^{\circ} with respect to the abab plane with an ordered moment of 3.38(6) μB\mu_{\rm B}/Mn.

Herein, we report the growth, crystal structure, ρ(T)\rho(T), magnetization as a function of magnetic field M(H)M(H), χ(T)\chi(T) and heat capacity Cp(T)C_{{\rm p}}(T) measurements of CaMn2As2{\rm CaMn_{2}As_{2}} and SrMn2As2{\rm SrMn_{2}As_{2}} single crystals. These studies were initiated because of the above-noted possibility that the Mn spin lattice might exhibit novel magnetic behaviors associated with the presence of geometric frustration within the triangular-lattice Mn layers. If the strongest AFM interactions are indeed within a triangular lattice layer, this should lead to a noncollinear AFM structure below TNT_{\rm N}. Instead, in a companion neutron diffraction study to the present work, the AFM structure of SrMn2As2{\rm SrMn_{2}As_{2}} was found to be collinear with the ordered Mn moments aligned in the abab plane with magnitude 3.6 μB\mu_{\rm B}/Mn.Das2016 This magnetic structure is the same as one of the two AFM structures proposed for CaMn2Sb2{\rm CaMn_{2}Sb_{2}} (Ref. Ratcliff2009, ) discussed above.

We discovered that the χ(T)\chi(T) of SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} and the Cp(T)C_{\rm p}(T) of SrMn2As2{\rm SrMn_{2}As_{2}} above their respective Néel temperatures TNT_{\rm N} of 120 and 62 K exhibit behaviors characteristic of strong dynamic short-range AFM spin correlations up to at least 900 K, likely arising from quasi-two-dimensional connectivity of strong AFM Mn–Mn exchange interactions within the corrugated honeycomb Mn spin sublattice. This result is interesting because such strong AFM spin correlations up to high temperatures and the suppression of TNT_{\rm N} to much lower temperatures than expected from molecular field theory, due to AFM fluctuations associated with the low dimensionality of the exchange interaction connectivity, may give rise to novel physical properties upon doping the compounds into the metallic state.

II Experimental Details

Single crystals of SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} were grown using Sn flux. High-purity elements Sr (99.95%) from Sigma Aldrich, and Ca (99.95%), Co (99.998%), As (99.9999%) and Sn (99.999%) from Alfa Aesar were taken in the ratio (Sr,Ca):Mn:As:Sn = 1:2:2:20 and placed in an alumina crucible that was subsequently placed in a silica tube that was evacuated, partially refilled with high-purity argon (\approx1/4 atm pressure) and then sealed. After preheating at 600C for 5 h, the assembly was heated to 1150 C at the rate of 50 C/h and held at this temperature for 20 h for homogenization. Then the furnace was slowly cooled at the rate of 5 C/h to 700 C. At this temperature the molten Sn flux was decanted using a centrifuge. Shiny hexagonal-shape single crystals of maximum dimensions 4×3×1mm34\times 3\times 1\ {\rm mm}^{3} were obtained.

Semiquantitative chemical analyses of the single crystals were performed using a JEOL scanning electron microscope (SEM) equipped with an EDX (energy-dispersive x-ray analysis) detector, where a counting time of 120 s was used. A room-temperature powder x-ray diffraction (XRD) pattern was recorded on crushed single crystals using a Rigaku Geigerflex powder diffractometer with Cu Kα\alpha radiation at diffraction angles 2θ\theta from 10 to 110 with a 0.02 step width. The data were analysed by Rietveld refinement using FullProf software.fullprof

M(T)M(T) measurements for 1.8KT300K{\rm 1.8~K\leq}~T~{\rm\leq 300~K} and M(H)M(H) measurements for H5.5H\leq 5.5 T were carried out using a Quantum Design, Inc., Magnetic Properties Measurement System (MPMS). The high-temperature M(T)M(T) for 300KT900K{\rm 300~K\leq T\leq 900~K} was measured using the vibrating sample magnetometer (VSM) option of a Quantum Design, Inc., Physical Properties Measurement System (PPMS). In this paper we exclusively use Gaussian cgs units for MM, χ\chi and HH (see Sec. 3.5.1 of Ref. Johnston2010, ). In this system of units, the Tesla (T) is a unit of convenience for HH defined as 1 T = 10410^{4} Oe, where Oe is the conventional cgs unit for HH.

C(T)pC{\rm{}_{p}}(T) data were obtained using a relaxation method with the heat capacity option of the PPMS. Four-probe ρ(T)\rho(T) data were obtained with an ac current amplitude I=1μAI=1\,\rm{\mu A} at a frequency of 37.7 Hz using the ac transport option of the PPMS. Electrical contacts to a crystal were made by soldering 0.05 mm diameter Pt wire to a crystal using indium solder.

III Experimental Results

III.1 Crystal Structure

Refer to caption
Refer to caption
Figure 1: (Colour online) Powder x-ray diffraction patterns (open circles) of (a) SrMn2As2{\rm SrMn_{2}As_{2}} and (b) CaMn2As2{\rm CaMn_{2}As_{2}} at room temperature. The solid line represents the Rietveld refinement fit calculated for the CaAl2Si2{\rm CaAl_{2}Si_{2}}-type trigonal structure with space group P3¯m1P\bar{3}m1 together with the Sn impurity phase.
Table 1: Crystallographic and Rietveld refinement parameters obtained from powder XRD of SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} crystals. The structures are trigonal CaAl2Si2{\rm CaAl_{2}Si_{2}}-type with space group P3¯m1P\bar{3}m1. The atomic coordinates of SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} are Sr/Ca: 1a1a (0, 0, 0); Mn: 2d2d (1/3, 2/3, zMnz{\rm{}_{Mn}}); and As: 2d2d (1/3, 2/3, zAsz{\rm{}_{As}}). The shortest Mn–Mn interatomic distances in SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} [see Fig. 2(b)] are also listed.
SrMn2As2{\rm SrMn_{2}As_{2}} CaMn2As2{\rm CaMn_{2}As_{2}}
Lattice parameters
aa (Å) 4.2962(1) 4.2376(1)
cc (Å) 7.2997(2) 7.0331(2)
c/ac/a 1.6991(1) 1.6596(1)
Vcell(Å3)V_{\rm cell}~(\rm{\AA }^{3}) 116.682(6) 109.372(6)
Atomic coordinates
zMnz{\rm{}_{Mn}} 0.6231(1) 0.6248(4)
zAsz{\rm{}_{As}} 0.2667(2) 0.2537(3)
Refinement quality
χ2\chi^{2} 3.05 4.03
RpR_{\rm p} (%) 10.3 12.7
RwpR_{\rm wp} (%) 13.6 16.4
Shortest Mn–Mn
distances (Å)
d1d_{1} 3.06306(8) 3.0112(2)
d2d_{2} 4.29620(5) 4.23760(5)
d3d_{3} 5.27633(7) 5.1985(2)
dz1d_{z1} 6.0357(2) 5.8171(4)
dz2d_{z2} 7.2997(2) 7.0331(2)

The crystal symmetry of several SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} crystals was checked by x-ray Laue back scattering which showed trigonal symmetry with well-defined diffraction spots which clearly indicated the good quality of the crystals. In this paper we use the hexagonal setting for the trigonal unit cell. The data also revealed that the SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} platelike crystals grow with the plate surface parallel to the hexagonal abab plane. SEM imaging and EDX analyses were performed to check the chemical composition and surface morphology of the crystals. The average elemental ratio of the samples was in agreement with the expected 1:2:2 stoichiometry of the compounds to within the errors. The amount of Sn incorporated into the crystal structure from the Sn flux is zero to within the experimental error. The present analyses did not show any other elements.

The phase purity of our SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} crystals was confirmed by powder XRD. Their XRD patterns at 300 K along with the results of Rietveld refinements are shown in Figs. 1(a) and 1(b), respectively. One sees the presence of adventitious elemental Sn flux, so two-phase Rietveld refinements were carried out. The refinement results confirm that the crystals have the trigonal CaAl2Si2{\rm CaAl_{2}Si_{2}}-type structure with space group P3¯m1P\bar{3}m1. The refinement and crystal parameters obtained are listed in Table 1. The crystal parameters are in good agreement with previously reported values.Mewis1978 ; Brechtel1978 ; Wang2011

Refer to caption
Refer to caption
Refer to caption
Refer to caption
Figure 2: (Colour online) Trigonal CaAl2Si2{\rm CaAl_{2}Si_{2}}-type crystal structure of SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} in the hexagonal setting. (a) Outline of a unit cell containing one formula unit. (b) Corrugated Mn honeycomb layers as viewed from nearly perpendicular to the cc axis. The smallest Mn–Mn interatomic distances within a corrugated Mn honeycomb layer (d1d_{1}, d2d_{2}, d3d_{3}) and between layers (dz1d_{z1}, dz2d_{z2}) are indicated. (c) Expanded view of the structure from a view nearly perpendicular to the cc axis showing the corrugated [Mn2As]22{}_{2}]^{-2} honeycomb layers separated by Sr or Ca. (d) Projection of the Mn sublattice onto the abab plane with a slight cc axis component illustrating the corrugated Mn honeycomb lattice. The corrugated honeycomb lattice layer can be viewed as a triangular lattice bilayer [compare with panel (b)]. The outline of a unit cell in the abab plane is also shown.

Figure 2(a) shows a unit cell of trigonal SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} in the hexagonal setting. As shown in Figs. 2(b) and 2(d), the structure consists of corrugated honeycomb [Mn2As2]2\left[{\rm Mn_{2}As_{2}}\right]^{-2} layers that are stacked along the cc axis and separated by Sr+2 or Ca+2 cations, respectively. Alternatively, the Mn sublattice can be viewed as triangular double layers of Mn stacked along the cc axis and separated by Ca or Sr atoms. The three smallest Mn–Mn interatomic distances [see Fig. 2(b)] are within the corrugated Mn honeycomb layers and are listed in Table 1. The nearest-neighbor Mn–Mn distance (d1d_{1}) is between the two Mn atoms at different heights (zz values) within a unit cell. The second-nearest-neighbor Mn-Mn distance (d2d_{2}) is between Mn atoms at the same height in adjacent unit cells along the abab plane forming a triangular-lattice layer, and the third-nearest-neighbor Mn–Mn distance (d3d_{3}) is between nearest-neighbor Mn atoms in adjacent unit cells in the abab plane. The nearest- and second-nearest-neighbor distances dz1d_{z1} and dz2d_{z2} between Mn atoms in adjacent layers in different unit cells along the cc axis are also listed in Table 1.

Since the minimum intralayer Mn–Mn distance d13d_{1}\approx 3 Å is much shorter than the minimum interlayer Mn–Mn distance (dz16d_{z1}\approx 6 Å), CaMn2As2{\rm CaMn_{2}As_{2}} and SrMn2As2{\rm SrMn_{2}As_{2}} likely have a quasi-two-dimensional Mn–Mn exchange interaction connectivity. This large spatial anisotropy in the exchange interactions should be obvious from χ(T>TN)\chi(T>T_{\rm N}) measurements, which is confirmed below. These exchange interactions could arise from direct Mn–Mn interactions and/or from indirect Mn–As–Mn superexchange interactions. The latter would likely occur via two main paths: (i) between first-nearest-neighbor Mn spins with \angle{}Mn-As-Mn (θ1\theta_{1}) = 72 and another between second-neighbor Mn spins with \angle{}Mn-As-Mn (θ2\theta_{2}) = 111 [see Fig. 2(c)]. It will be interesting to see which of these interactions are dominant within the unusual trigonal symmetry of the Mn spin lattice.

To summarize, the SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} trigonal structure is quite different from the body-centered tetragonal ThCr2Si2{\rm ThCr_{2}Si_{2}} structure found for the AFe2As2A{\rm Fe_{2}As_{2}} parent compounds that is composed of metal-arsenide tetrahedra separated by alkaline earth layers. The primary difference between them is the geometry of the transition metal layers. In SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}}, the Mn bilayer is a corrugated Mn honeycomb lattice where each Mn atom is coordinated by three other Mn atoms at 90\sim 90^{\circ} like the corner of the cube as seen in Figs. 2(b) and 2(c), whereas in ThCr2Si2{\rm ThCr_{2}Si_{2}}-type compounds such as BaMn2As2 or BaFe2As2, the Mn or Fe network is a simple square-planar lattice where each Mn or Fe is coordinated by four other Mn or Fe atoms, also at 90 angles between them. On the other hand, the AFM in BaMn2As2{\rm BaMn_{2}As_{2}} is quasi-two-dimensional,Johnston2011 just as we find it to be in SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} from the χ(T)\chi(T) data in Sec. III.3 below.

III.2 In-Plane Electrical Resistivity

Refer to caption
Refer to caption
Figure 3: (Colour online) In-plane electrical resistivity ρ\rho of (a) SrMn2As2{\rm SrMn_{2}As_{2}} and (b) CaMn2As2{\rm CaMn_{2}As_{2}} versus temperature TT from 50\sim 50 to 300 K. The insets show plots of log10ρ\log_{10}\rho versus 1000/TT. The solid straight lines through the data are fits over the temperature interval between 70 and 120 K by the expression log10ρ=A+2.303Δ/T\log_{10}\rho=A+2.303\Delta/T and the resulting fitted activation energies are listed. The dashed lines are extrapolations.

Figures 3(a) and 3(b) show ρ(T)\rho(T) of SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}}, respectively, from 50\sim 50 to 300 K measured in the abab plane. The data show that both compounds are semiconductors with insulating intrinsic ground states. We fitted ρ(T)\rho(T) in the temperature region between 70 and 120 K by the expression log10ρ=A+2.303Δ/kBT\log_{10}\rho=A+2.303\Delta/k_{\rm B}T, where AA is a constant, kBk_{\rm B} is Boltzmann’s constant and Δ\Delta is the activation energy. The fits are shown as the solid straight lines through the data in the insets of Figs. 3(a) and 3(b), which give the activation energies Δ=85\Delta=85 meV for SrMn2As2{\rm SrMn_{2}As_{2}} and Δ=61\Delta=61 meV for CaMn2As2{\rm CaMn_{2}As_{2}}. These activation energies are of the same order as previously obtained from ρ(T)\rho(T) data for BaMn2As2.Singh2009 Our activation energy for SrMn2As2 is significantly smaller than the previously reported values Δ=0.29\Delta=0.29 and 0.64 eV, depending on the TT range, that were also obtained from single-crystal in-plane ρ(T)\rho(T) data.Wang2011

III.3 Magnetization and Magnetic Susceptibility

Refer to caption
Refer to caption
Refer to caption
Figure 4: (Color online) ZFC magnetic susceptibility χ(T)\chi(T) of SrMn2As2{\rm SrMn_{2}As_{2}} from 1.8 to 300 K measured in magnetic fields (a) H=0.1H=0.1 T and (b) H=3H=3 T applied in the abab plane (Hab,χabH\parallel ab,\ \chi_{ab}) and along the cc axis (Hc,χcH\parallel{c},\ \chi_{c}). (c) ZFC susceptibility versus TT for 1.8T9001.8\leq T\leq 900 K measured in H=3H=3 T applied in the abab plane (HcH\perp c). The “intrinsic” values are obtained from M(H)M(H) isotherms using Eq. (1) and the “corrected” ones are obtained from M(T)M(T) data using Eq. (2).
Refer to caption
Refer to caption
Refer to caption
Figure 5: (Color online) Magnetic susceptibility data for CaMn2As2{\rm CaMn_{2}As_{2}}. The plot and symbol designations are the same as for SrMn2As2{\rm SrMn_{2}As_{2}} in Fig. 4.

The zero-field-cooled (ZFC) magnetic susceptibility χM/H\chi\equiv M/H versus TT measured in H=0.1H=0.1 T and H=3H=3 T applied in the abab plane (Hab,χabH\parallel ab,\ \chi_{ab}) and along the cc axis (Hc,χcH\parallel c,\ \chi_{c}) for single crystals of SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} are shown in Figs. 4 and 5, respectively.

Clear AFM transitions are observed in χ(T)\chi(T) at TN120T{\rm{}_{N}}\approx 120 K for SrMn2As2{\rm SrMn_{2}As_{2}} and TN65T{\rm{}_{N}}\approx 65 K for CaMn2As2{\rm CaMn_{2}As_{2}}, as indicated by vertical arrows in Figs. 4(a) and 5(a), respectively. We also performed FC (field-cooled) and ZFC χ(T)\chi(T) measurements at H=0.1H=0.1 T and H=3H=3 T (not shown here). No hysteresis was observed between the ZFC and FC data, which is consistent with long-range AFM ordering of SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} below their respective Néel temperatures. The data in Figs. 4(a) and 5(a) for T>TNT>T_{\rm N} are nearly isotropic, as expected for Mn+2 with spin S=5/2S=5/2 and g2g\approx 2.

From Figs. 4(a) and 5(a), the anisotropy in χ\chi at T<TNT<T{\rm{}_{N}} indicates that the hard axis is the cc axis and the abab plane is the easy plane for both compounds. Furthermore, the nonzero limits of χab(T0)\chi_{ab}(T\to 0) suggest that the AFM structure could be either a collinear AFM with multiple domains aligned within the abab plane or an intrinsic noncollinear structure with moments again aligned in the abab plane.Johnston2012 ; Johnston2015 ; Anand2015 ; Ryan2015 For collinear ordering, magnetic dipole interactions between the Mn moments favor abab-plane moment alignment over cc-axis alignment.Johnston2016 In SrMn2As2{\rm SrMn_{2}As_{2}} [see Fig. 4(b)], the anisotropy in χ\chi for T<TNT<T{\rm{}_{N}} is eliminated by a field of 3 T, which indicates a relatively small magnetocrystalline anisotropy compared to that in CaMn2As2{\rm CaMn_{2}As_{2}} judging from Fig. 5(b). The small upturns in χ\chi in Figs. 4 and 5 below 20\sim 20 K are believed due to trace amounts of paramagnetic local-moment impurities.

Refer to caption
Refer to caption
Figure 6: (Color online) Magnetization MM versus magnetic field HH at various temperatures TT with (a) HH in the abab plane (HabH\parallel ab) and with (b) HH along the cc axis (HcH\parallel c) for SrMn2As2{\rm SrMn_{2}As_{2}}.
Refer to caption
Refer to caption
Figure 7: (Color online) Magnetization MM versus magnetic field HH at various temperatures TT with (a) HH in the abab plane (HabH\parallel ab) and with (b) HH along the cc axis (HcH\parallel c) for CaMn2As2{\rm CaMn_{2}As_{2}}.

A small jump in χ(T)\chi(T) is observed for SrMn2As2{\rm SrMn_{2}As_{2}} in Fig. 4(b) on cooling below about 320 K. This is believed due to FM MnAs impurities with this Curie temperature that are present on the crystal surface and/or as an inclusion in the crystal, as previously observed for BaMn2As2{\rm BaMn_{2}As_{2}} crystals.Singh2009 To extract the intrinsic magnetic behavior of SrMn2As2{\rm SrMn_{2}As_{2}}, we carried out M(H)M(H) isotherm measurements at various temperatures. Figures 6 and 7 show isothermal M(H)M(H) data for SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} crystals, respectively, at different TT for HabH\parallel{ab} plane (MabM_{ab}) and HcH\parallel{c} axis (McM_{c}). The M(H)M(H) curves are almost linear at high fields, but for SrMn2As2{\rm SrMn_{2}As_{2}} one sees nonlinearities at low fields (H<1H<1–2 T) for T<200T<200 K, confirming the presence of saturable FM impurities.

To extract the intrinsic χ\chi (χint\chi_{\rm int}) we fitted the M(H)M(H) data in the high-field range H=3.5H=3.5–5.5 T by the linear relation

M(H,T)=Ms(T)+χint(T)H,M(H,T)=M_{\rm s}(T)+\chi_{\rm int}(T)H, (1)

where Ms(T)M_{\rm s}(T) is the saturation magnetization due to the FM impurities. The T0T\to 0 value of MsM_{\rm s} for SrMn2As2{\rm SrMn_{2}As_{2}} is 7Gcm3/mol=0.0013μB{\rm 7~G~cm^{3}/mol}=0.0013~\mu_{\rm B}/f.u., which corresponds to 0.04 mol% of MnAs impurities using the saturation moment of 3.5μB\approx 3.5~\mu_{\rm B}/f.u. (Refs. Haneda1977, ; Saparov2012, ) for MnAs. The χ(T)M(T)/H\chi(T)\equiv M(T)/H data in Figs. 4(b) and 5(b) were measured with H=3H=3 T. Therefore, we obtained the intrinsic χ\chi from the isotherm data according to

χint(T)=M(T)measuredM(T)s3T.\chi_{\rm int}(T)=\frac{M{\rm{}_{measured}}(T)-M{\rm{}_{s}}(T)}{3~\rm{T}}. (2)

The χint(T)\chi_{\rm int}(T) data are shown by the filled green triangles in Figs. 4(b) and 5(b). It is seen that the χint(T)\chi_{\rm int}(T) data for CaMn2As2{\rm CaMn_{2}As_{2}} match very well for both field directions with the χM/H\chi\equiv M/H data in Fig. 5(b), indicating a clean crystal without any detectable FM impurities.

In order to further clarify the magnetism in these systems we measured χ(T)M(T)/H\chi(T)\equiv M(T)/H in the extended temperature range up to 900 K for SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} as shown in Figs. 4(c) and 5(c), respectively. One sees that χ\chi exhibits very broad maxima at 400\sim 400 K for both compounds. This feature is a signature of a low-dimensional local-moment AFM system.Johnston2011 Thus SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} undergo a phase transition to a long-range ordered state below TNT{\rm{}_{N}}, preceeded by strong short-range AFM order at higher temperatures. Indeed, the Curie-Weiss temperature region of χ\chi is not reached even at 900 K, indicating that strong AFM correlations survive to significantly higher temperatures.

From the Mn–Mn interatomic distances discussed in Sec. III.1, we inferred that the Mn–Mn exchange coupling along the cc axis between the corrugated honeycomb Mn layers in the abab plane is much smaller than within the layers. We confirm this here and in addition infer that the Mn–Mn exchange coupling within the corrugated honeycomb layers is dominantly antiferromagnetic.

III.4 Heat Capacity

Refer to caption
Refer to caption
Figure 8: (Color online) Heat capacity CpC{\rm{}_{p}} versus temperature TT for (a) SrMn2As2{\rm SrMn_{2}As_{2}} and (b) CaMn2As2{\rm CaMn_{2}As_{2}}. The insets show C(T)p/TC{\rm{}_{p}}(T)/T versus T2T^{2} for T<5T<5 K, where the straight lines though the respective data are fits by Eq. (4).

Figures 8(a) and 8(b) show zero-field C(T)pC{\rm{}_{p}}(T) data for SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}}, respectively. The sharp anomalies in Cp(T)C_{\rm p}(T) at 120(2) K in SrMn2As2{\rm SrMn_{2}As_{2}} and at 62(3) K in CaMn2As2{\rm CaMn_{2}As_{2}} are the respective Néel temperatures of the two compounds, which are in good agreement with TNT{\rm{}_{N}} values found above from the respective χ(T)\chi(T) data.

III.4.1 Low-Temperature Behaviors

The insets of Figs. 8(a) and 8(b) show C(T)p/TC{\rm{}_{p}}(T)/T versus T2T^{2} between 1.8 and 5 K. At low temperatures we model the Cp(T)C_{\rm p}(T) data byKittel2005

Cp=γT+βT3,C_{\rm p}=\gamma T+\beta T^{3}, (3)

where the coefficient γ\gamma is usually due to the electronic contribution (Sommerfeld coefficient) and β\beta is the coefficient of the Debye T3T^{3} lattice contribution in the absence of three-dimensional AFM spin-wave contributions. The data were therefore fitted by the expression

CpT=γ+βT2,\frac{C{\rm{}_{p}}}{T}=\gamma+\beta T^{2}, (4)

From the fits of Eq. (6) to the data in the insets of Figs. 8(a) and 8(b) we obtain γ=0.0(1)\gamma=0.0(1) mJ/(mol K2) for SrMn2As2{\rm SrMn_{2}As_{2}} and 0.05(7) mJ/(mol K2) for CaMn2As2{\rm CaMn_{2}As_{2}}. The null values of γ\gamma are consistent with the insulating ground states found from the ρ(T)\rho(T) measurements in Sec. III.2.

The fitted values for β\beta are

β=0.35(1)mJmolK4forSrMn2As2\beta=0.35(1)\,{\rm\frac{mJ}{mol\,K^{4}}}\ {\rm for\ SrMn_{2}As_{2}} (5)

and 0.37(1) mJ/(mol K)4{}^{4}) for CaMn2As2{\rm CaMn_{2}As_{2}}. We estimate the Debye temperatures ΘD\Theta_{\rm D} for the two compounds from the Debye theory according toKittel2005

Θ=D(12π4Rn5β)1/3\Theta{\rm{}_{D}}=\left(\frac{12\pi^{4}Rn}{5\beta}\right)^{1/3} (6)

where RR is the molar gas constant and nn is the number of atoms per formula unit [n=5n=5 for SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}}]. We obtain

ΘD\displaystyle\Theta{\rm{}_{D}} =\displaystyle= 303(3)K(SrMn2As2),\displaystyle 303(3)~{\rm K}\quad({\rm SrMn_{2}As_{2}}), (7a)
=\displaystyle= 297(3)K(CaMn2As2).\displaystyle 297(3)~{\rm K}\quad({\rm CaMn_{2}As_{2}}). (7b)

In the absence of anisotropy gaps in the AFM spin-wave spectrum, contributions to β\beta could arise from excitations of three-dimensional AFM spin waves at the low temperatures at which the β\beta values were extracted; hence the quoted values of ΘD\Theta_{\rm D} are lower limits.

III.4.2 Magnetic Contributions to the Heat Capacity and Entropy of SrMn2As2{\rm SrMn_{2}As_{2}}

Refer to caption
Refer to caption
Refer to caption
Figure 9: (Color online) (a) Heat capacity CpC{\rm{}_{p}} versus TT for SrMn2As2{\rm SrMn_{2}As_{2}} and a fit of CVDebye(T)C_{\rm V\,Debye}(T) in Eqs. (8) to the data for T=1.8T=1.8 K to 60 K. (b) Magnetic heat capacity Cmag(T)C_{\rm mag}(T). (c) Magnetic entropy Smag(T)S_{\rm mag}(T) obtained using Eq. (9). The dashed lines are Smag(T)S_{\rm mag}(T\to\infty) for S=5/2S=5/2 and S=3/2S=3/2.

Here we extract estimates of the magnetic contribution Cmag(T)C_{\rm mag}(T) to the measured heat capacity of SrMn2As2{\rm SrMn_{2}As_{2}} and from that we obtain the magnetic contribution Smag(T)S_{\rm mag}(T) to the entropy over the temperature range from 1.8  to 200 K of the Cp(T)C_{\rm p}(T) measurements. To accomplish this goal we first obtain an estimate of the lattice contribution Clatt(T)C_{\rm latt}(T) to the measured Cp(T)C_{\rm p}(T). We use the expression

Clatt=nCVDebye,C_{\rm latt}=nC_{\rm V\,Debye}, (8a)
where CVDebyeC_{\rm V\,Debye} is the Debye lattice heat capacity per mole of atoms given byKittel2005
CVDebye=9R(TΘD)30ΘD/Tx4ex(ex1)2𝑑x.C_{\rm V\,Debye}=9R\left(\frac{T}{\Theta_{\rm D}}\right)^{3}\int_{0}^{\Theta_{\rm D}/T}\frac{x^{4}e^{x}}{\left(e^{x}-1\right)^{2}}dx. (8b)

The representation of CVDebye(T/ΘD)C_{\rm V\,Debye}(T/\Theta_{\rm D}) used here is an accurate analytic Padé approximant function of T/ΘDT/\Theta_{\rm D} obtained by fitting numerical solutions of Eq. (8b) for a list of T/ΘDT/\Theta_{\rm D} values.Goetsch2012 The fit of Cp(T)C_{\rm p}(T) for SrMn2As2{\rm SrMn_{2}As_{2}} in Fig. 9(a) by Eqs. (8) over the temperature range from 1.8 to 60 K and its extrapolation is shown by the blue curve in Fig. 9(a) using the fitted Debye temperature ΘD=265.7\Theta_{\rm D}=265.7 K. This value of ΘD\Theta_{\rm D} is comparable with the value of 303 K obtained from the fit to the CpC_{\rm p} data for SrMn2As2{\rm SrMn_{2}As_{2}} at low TT in Eq. (7a), especially considering that ΘD\Theta_{\rm D} for a compound typically varies by ±20\pm 20% on cooling from 300 K to 2 K.Smart1966

The Cmag(T)C_{\rm mag}(T) is calculated as the difference between the measured Cp(T)C_{\rm p}(T) and the fitted Clatt(T)C_{\rm latt}(T) in Fig. 9(a). The result in shown in Fig. 9(b), where a sharp peak at TN=120T_{\rm N}=120 K is seen. The CmagC_{\rm mag} at T>TNT>T_{\rm N} shows that there is strong dynamic short-range AFM order above TNT_{\rm N}. The magnetic entropy Smag(T)S_{\rm mag}(T) is calculated from Cmag(T)C_{\rm mag}(T) using

Smag(T)=0TCmag(T)T𝑑T,S_{\rm mag}(T)=\int_{0}^{T}\frac{C_{\rm mag}(T)}{T}\,dT, (9)

and the result is shown in Fig. 9(c). The entropy of completely disordered spins SS per mole of SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} is S(T)=2Rln(2S+1)S(T\to\infty)=2R\ln(2S+1), which gives

Smag(T)\displaystyle S_{\rm mag}(T\to\infty) =\displaystyle= 23.1JmolK(S=3/2)\displaystyle 23.1\,{\rm\frac{J}{mol\,K}}\quad(S=3/2) (10a)
=\displaystyle= 29.8JmolK(S=5/2),\displaystyle 29.8\,{\rm\frac{J}{mol\,K}}\quad(S=5/2), (10b)

as shown by the horizontal dashed black lines in Fig. 9(c). This range of spin values encompasses the known variations in the ordered moments of Mn spins in various materials similar to ours which can arise from quantum fluctuation and/or hybridization effects (see, e.g., Ref. Johnston2011, ). We find that Smag(200K)10J/molKS_{\rm mag}(200~{\rm K})\approx 10~{\rm J/mol\,K} in Fig. 9(c) is only 33\approx 33% of the value for S=5/2S=5/2 in Eqs. (10) and is still only 43\approx 43% of the value for S=3/2S=3/2. Thus the strong short-range AFM order revealed in the Cmag(T)C_{\rm mag}(T) and Smag(T)S_{\rm mag}(T) data above TNT_{\rm N} is consistent with the above conclusion from the χ(T)\chi(T) data that strong short-range AFM order survives from TNT_{\rm N} up to at least 900 K.

IV Summary

We have shown that SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}} are AFM insulators with Néel temperatures TN=120T_{\rm N}=120 K and 62 K, respectively. The microscopic origin of this large difference in magnitude of the Néel temperatures together with the reason why the Sr compound has a higher TNT_{\rm N} than the Ca one, in spite of the smaller unit cell of the latter, remain to be explained.

The χ(T)\chi(T) data at TTNT\leq T_{\rm N} indicate that the hexagonal cc axis is a hard axis, with the ordered Mn spin-5/2 moments lying within the abab plane. Since a collinear AFM structure within the abab plane is inferred for SrMn2As2{\rm SrMn_{2}As_{2}} from neutron diffraction measurements that were carried out in a companion study,Das2016 the nonzero limits of χab(T<TN)\chi_{ab}(T<T_{\rm N}) observed for this compound must arise from the three collinear AFM domains with their axes at 6060^{\circ} to each other within the abab plane. If the populations of the three domains are equal, within molecular field theory one obtains χab(T0)=χ(TN)/2\chi_{ab}(T\to 0)=\chi(T_{\rm N})/2, in approximate agreement with the χab(T)\chi_{ab}(T) data in Fig. 4(a). The prediction of the easy axis arising from the Mn–Mn magnetic dipole interactions in a collinear magnetic structure of SrMn2As2{\rm SrMn_{2}As_{2}} obtained using the formalism of Ref. Johnston2016, and the experimental crystal structure is that the ordered moments should lie in the abab plane as inferred here from the χ(T)\chi(T) data and also observedDas2016 in the neutron diffraction experiments.

Thus the potential geometric frustration for AFM ordering within the triangular-lattice bilayers parallel to the abab plane that originally motivated this work is apparently not important in SrMn2As2{\rm SrMn_{2}As_{2}} and CaMn2As2{\rm CaMn_{2}As_{2}}. In particular, if AFM Mn–Mn exchange interactions within a triangular sublattice layer were dominant, a noncollinear AFM structure would have resulted instead of the observedDas2016 collinear AFM structure.

Strong dynamic AFM short-range correlations up to at least 900 K as observed in our χ(T)\chi(T) measurements, consistent with our Smag(T)S_{\rm mag}(T) data up to 200 K, are likely due to quasi-two-dimensional connectivity of strong AFM Mn–Mn exchange interactions within the corrugated honeycomb Mn layers. This in turn offers the possibility of novel electronic ground states arising upon doping these materials into the metallic state.

Acknowledgements.
We thank P. Das, A. Kreyssig and A. I. Goldman for helpful discussions. This research was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering. Ames Laboratory is operated for the U.S. Department of Energy by Iowa State University under Contract No. DE-AC02-07CH11358.

References

  • (1) G. Just and P. Paufler, J. Alloys Compd. 232, 1 (1996).
  • (2) D. C. Johnston, Adv. Phys. 59, 803 (2010).
  • (3) G. R. Stewart, Rev. Mod. Phys. 83, 1589 (2011).
  • (4) D. J. Scalapino, Rev. Mod. Phys. 84, 1383 (2012).
  • (5) E. Dagotto, Rev. Mod. Phys. 85, 849 (1913).
  • (6) R. M. Fernandes, A. V. Chubukov, and J. Schmalian, Nat. Phys. 10, 97 (2014).
  • (7) H. Hosono and K. Kuroki, Physica C 514, 399 (2015).
  • (8) P. Dai, Rev. Mod. Phys. 87, 855 (2015).
  • (9) D. S. Inosov, Compt. Rend. Phys. 17, 60 (2016).
  • (10) Q. Si, R. Yu, and E. Abrahams, Nat. Rev. Mater. 1, 1 (2016).
  • (11) E. D. Bauer, F. Ronning, B. L. Scott, and J. D. Thompson, Phys. Rev. B 78, 172504 (2008).
  • (12) F. Ronning, N. Kurita, E. D. Bauer, B. L. Scott, T. Park, T. Klimczuk, R. Movshovich, and J. D. Thompson, J. Phys.: Condens. Matter 20, 342203 (2008).
  • (13) A. Pandey, D. G. Quirinale, W. Jayasekara, A. Sapkota, M. G. Kim, R. S. Dhaka, Y. Lee, T. W. Heitmann, P. W. Stephens, V. Ogloblichev, A. Kreyssig, R. J. McQueeney, A. I. Goldman, A. Kaminski, B. N. Harmon, Y. Furukawa, and D. C. Johnston, Phys. Rev. B 88, 014526 (2013).
  • (14) W. Jayasekara, Y. Lee, A. Pandey, G. S. Tucker, A. Sapkota, J. Lamsal, S. Calder, D. L. Abernathy, J. L. Niedziela, B. N. Harmon, A. Kreyssig, D. Vaknin, D. C. Johnston, A. I. Goldman, and R. J. McQueeney, Phys. Rev. Lett. 111, 157001 (2013).
  • (15) P. Wiecki, V. Ogloblichev, A. Pandey, D. C. Johnston, and Y. Furukawa, Phys. Rev. B 91, 220406(R) (2015).
  • (16) A. S. Sefat, D. J. Singh, R. Jin, M. A. McGuire, B. C. Sales, and D. Mandrus, Phys. Rev. B 79, 024512 (2009).
  • (17) V. K. Anand, D. G. Quirinale, Y. Lee, B. N. Harmon, Y. Furukawa, V. V. Ogloblichev, A. Huq, D. L. Abernathy, P. W. Stephens, R. J. McQueeney, A. Kreyssig, A. I. Goldman, and D. C. Johnston, Phys. Rev. B 90, 064517 (2014).
  • (18) P. Wiecki, B. Roy, D. C. Johnston, S. L. Bud’ko, P. C. Canfield, and Y. Furukawa, Phys. Rev. Lett. 115, 137001 (2015).
  • (19) Y. Singh, A. Ellern, and D. C. Johnston, Phys. Rev. B 79, 094519 (2009).
  • (20) Y. Singh, M. A. Green, Q. Huang, A. Kreyssig, R. J. McQueeney, D. C. Johnston, and A. I. Goldman, Phys. Rev. B 80, 100403(R) (2009).
  • (21) D. C. Johnston, R. J. McQueeney, B. Lake, A. Honecker, M. E. Zhitomirsky, R. Nath, Y. Furukawa, V. P. Antropov, and Y. Singh, Phys. Rev. B 84, 094445 (2011).
  • (22) A. Antal, T. Knoblauch, Y. Singh, P. Gegenwart, D. Wu, and M. Dressel, Phys. Rev. B 86, 014506 (2012).
  • (23) S Calder, B. Saparov, H. B. Cao, J. L. Niedziela, M. D. Lumsden, A. S. Sefat, and A. D. Christianson, Phys. Rev. B 89, 064417 (2014).
  • (24) A. Pandey, R. S. Dhaka, J. Lamsal, Y. Lee, V. K. Anand, A. Kreyssig, T. W. Heitmann, R. J. McQueeney, A. I. Goldman, B. N. Harmon, A. Kaminski, and D. C. Johnston, Phys. Rev. Lett. 108, 087005 (2012).
  • (25) S. Yeninas, A. Pandey, V. Ogloblichev, K. Mikhalev, D. C. Johnston, and Y. Furukawa, Phys. Rev. B 88, 241111(R) (2013).
  • (26) J.-K. Bao, H. Jiang, Y.-L. Sun, W.-H. Jiao, C.-Y. Shen, H.-J. Guo, Y. Chen, C.-M. Feng, H.-Q. Yuan, Z.-A. Xu, G.-H. Cao, R. Sasaki, T. Tanaka, K. Matsubayashi, and Y. Uwatoko, Phys. Rev. B 85, 144523 (2012).
  • (27) A. Pandey, B. G. Ueland, S. Yeninas, A. Kreyssig, A. Sapkota, Yang Zhao, J. S. Helton, J.W. Lynn, R. J. McQueeney, Y. Furukawa, A. I. Goldman, and D. C. Johnston, Phys. Rev. Lett. 111, 047001 (2013).
  • (28) B. G. Ueland, A. Pandey, Y. Lee, A. Sapkota, Y. Choi, D. Haskel, R. A. Rosenberg, J. C. Lang, B. N. Harmon, D. C. Johnston, A. Kreyssig, and A. I. Goldman, Phys. Rev. Lett. 114, 217001 (2015).
  • (29) A. Pandey and D. C. Johnston, Phys. Rev. B 92, 174401 (2015).
  • (30) J. Lamsal, G. S. Tucker, T. W. Heitmann, A. Kreyssig, A. Jesche, A. Pandey, W. Tian, R. J. McQueeney, D. C. Johnston, and A. I. Goldman, Phys. Rev. B 87, 144418 (2013).
  • (31) A. Mewis, Z. Naturforsch. 33b, 606 (1978).
  • (32) E. Brechtel, G. Cordier, and H. Schäfer, Z. Naturforsch. 33b, 820 (1978).
  • (33) A. P. Ramirez, Annu. Rev. Mater. Sci. 24, 453 (1994).
  • (34) R. Moessner and A. P. Ramirez, Physics Today 59, 24 (2006).
  • (35) L. Balents, Nature 464, 199 (2010).
  • (36) Z.W. Wang, H.X. Yang, H.F. Tian, H.L. Shi, J.B. Lu, Y.B. Qin, Z. Wang, J.Q. Li , J. Phys. Chem. Solids 72, 457 (2011).
  • (37) W. Ratcliff II, A. L. Lima Sharma, A. M. Gomes, J. L. Gonzalez, Q. Huang, and J. Singleton, J. Magn. Magn. Mater. 321, 2612 (2009).
  • (38) C. A. Bridges, V. V. Krishnamurthy, S. Poulton, M. P. Paranthaman, B. C. Sales, C. Myers, and S. Bobev, J. Magn. Magn. Mater. 321, 2653 (2009).
  • (39) P. Das, N. S. Sangeetha, A. Pandey, Z. A. Benson, T. W. Heitmann, D. C. Johnston, A. I. Goldman, and A. Kreyssig, arXiv:1605:02004.
  • (40) J. Rodríguez-Carvajal, Phys. B: Condens. Matter 192, 55 (1993).
  • (41) D. C. Johnston, Phys. Rev. Lett. 109, 077201 (2012).
  • (42) D. C. Johnston, Phys. Rev. B 91, 064427 (2015).
  • (43) V. K. Anand and D. C. Johnston, Phys. Rev. B 91, 184403 (2015).
  • (44) D. H. Ryan, J. M. Cadogan, V. K. Anand, D. C. Johnston, and R. Flacau, J. Phys.: Condens. Matter 27, 206002 (2015).
  • (45) D. C. Johnston, Phys. Rev. B 93, 014421 (2016).
  • (46) S. Haneda, N. Kazama, Y. Yamaguchi, and H. Watanabe, J. Phys. Soc. Jpn. 42, 1201 (1977).
  • (47) B. Saparov, J. E. Mitchell, and A. S. Sefat, Supercond. Sci. Technol. 25, 084016 (2012).
  • (48) C. Kittel, Introduction to Solid State Physics (Wiley, Hoboken, NJ, 2005).
  • (49) R. J. Goetsch, V. K. Anand, A. Pandey, and D. C. Johnston, Phys. Rev. B 85, 054517 (2012).
  • (50) J. S. Smart, Effective Field Theories of Magnetism (Saunders, Philadelphia, 1966).