Evolution of structure and magnetism across the metal-insulator transition in the pyrochlore iridate ( Nd 1−x Ca x ) 2 Ir 2 O 7
Zach Porter, Eli Zoghlin, Samuel Britner, Samra Husremovic, Jacob P.C. Ruff, Yongseong Choi, Daniel Haskel, Geneva Laurita, Stephen D. Wilson
EEvolution of structure and magnetism across the metal-insulator transitionin the pyrochlore iridate ( Nd − x Ca x ) Ir O Zach Porter, Eli Zoghlin, Samuel Britner, Samra Husremovic, Jacob P. C. Ruff, Yongseong Choi, Daniel Haskel, Geneva Laurita, and Stephen D. Wilson Materials Department, University of California, Santa Barbara, California 93106, USA Department of Chemistry and Biochemistry, Bates College, Lewiston, Maine 04240, USA CHESS, Cornell University, Ithaca, New York 14853, USA Advanced Photon Source, Argonne National Laboratory, Argonne, Illinois 60439, USA (Dated: August 15, 2019)We report on the evolution of the thermal metal-insulator transition in polycrystalline samples ofNd Ir O upon hole-doping via substitution of Ca for Nd . Ca substitution mediates a filling-controlled Mott-like transition with minimal resolvable structural changes and without altering sitesymmetry. Local structure confirms that Ca substitution does not result in local chemical phaseseparation, and absorption spectroscopy establishes that Ir cations maintain a spin-orbit entangledelectronic configuration. The metal-insulator transition coincides with antiferromagnetic orderingon the Ir sublattice for all measured samples, and both decrease in onset temperature with Cacontent. Weak low-temperature upturns in susceptibility and resistivity for samples with high Cacontent suggest that Nd sublattice antiferromagnetism continues to couple to carriers in the metallicregime. I. INTRODUCTION
The pyrochlore structure A B O is comprised of in-terpenetrating A and B sublattices of corner-sharingtetrahedra. This structure hosts a variety of electronicand magnetic phases owing in part to geometric frus-tration in the presence of antiferromagnetic exchangeinteractions and the diversity of cation species whichcan be accommodated within this structural framework.In particular, setting B to Ir and A to a trivalent lan-thanide ion (or Y ) results in a material with smallenergy gaps between quadratic bands. Upon varyingthe lanthanide site, the A Ir O series exhibits a metal-insulator transition (MIT) where the gap monotonicallydecreases with increasing ionic radius until a metallicstate is reached between A=Nd and A=Pr lanthanideions. Interestingly, the transition from a metal into aninsulator for the pyrochlore iridates coincides with theformation of all-in-all-out (AIAO) antiferromagnetic or-der of the J eff =1/2 moments on the Ir magnetic sublat-tice, while magnetic A site lanthanide cations establishthe same order at lower temperatures.Theoretical studies predict that the suppression ofboth the insulating and magnetic phases results in a novelvariety of antiferromagnetic quantum critical points(QCPs) and the formation of topologically nontrivialelectronic states. Neighboring the quantum criticalregime, the Weyl semimetal phase has been directlyobserved, as have unconventional electronic propertiesnearby. However, fully exploring the experimentalmanifestation of this QCP has been challenging for theA Ir O materials system, primarily because of ambigu-ities regarding the mechanism of the MIT itself. Therole of magnetism in the semimetal phase is unresolved,and chiral spin textures are reported to persist into themetallic regime. Additionally, while q =0 AIAO order coincides with the opening of a charge gap, the MITshows evidence of both Mott and Slater (or mean-field)character. As an alternative to bandwidth control of theMIT with A-site isoelectronic substitution, filling controlvia carrier doping has the potential to access metallicstates in the A Ir O phase diagram with thermodynam-ically distinct magnetic and electronic properties.Previous studies of filling control in the pyrochloreiridates via hole-substitution have shown varied mag-netic responses. In recent work on (Eu − x Ca x ) Ir O ,the MIT and AIAO transition temperatures remaincoincident, and this transition is rapidly suppressedwith increasing x until a metallic ground state is real-ized between x =0.05 and x =0.10. However, other studieson A=(Y,Ca) and A=(Eu,Sr) describe different behaviorentirely: while the MIT temperature decreases rapidlywith x , the magnetic transition is decoupled in temper-ature, either fixed to the x =0 value with another fixedhigh-temperature transition or more slowly decreasingwith x . The discrepancies between these na¨ıvely similarmaterials, and the relation between their synthesis condi-tions and magnetic properties, warrants further investi-gation. Specifically, how the interplay between structureand magnetism evolves upon carrier-doping and how thepresence of magnetism on the A-site affects the evolutionof the filling-controlled MIT remain open questions.Here we study the effects of Ca substitution on theNd-site of (Nd − x Ca x ) Ir O with 0 ≤ x ≤ a r X i v : . [ c ond - m a t . m t r l - s c i ] A ug shows that Ca enters the lattice homogeneously with noresolvable clustering. As Ca substitution levels increase,the MIT is pushed downward in temperature and coin-cides with the onset of magnetic order. For doping levelsgreater than x = 0 .
05, the ground state switches to ametal with a weak upturn in the low temperature resis-tivity coupled to Nd magnetism. X-ray absorption spec-troscopy shows that the metallic state retains a strongspin-orbit coupled character with branching ratios littlealtered from the undoped material, and magnetic circu-lar dichroism data collected at the Ir L edges reveal ananomalous, weak net moment that survives across theMIT. Our data establish a complex interplay betweenmagnetism and the formation of the metallic state inhole-doped Nd Ir O . II. METHODS
Polycrystalline samples of (Nd − x Ca x ) Ir O weresynthesized by a solid-state reaction. Powders (99.99%,Alfa Aesar) of Nd O , CaCO , and IrO were mixed instoichiometric ratios, ground and heated at 1073 K in analumina crucible in air for 18 h. Next, the mixtures wereground, pressed into pellets at 300 MPa within an iso-static press, placed in alumina crucibles, and heated at1273 K in air for 8 days with one intermediate grinding.This step was repeated at 1323 K − O was reacted by adding4 mol% additional IrO to the powder before sinteringthe pellet at 1373 K in an alumina crucible sealed in aquartz tube under vacuum for 8 days with an interme-diate grinding. For these samples, the final pellet wassintered at 1173 K for 2 days in air.Samples were characterized by synchrotron powder X-ray diffraction (XRD) measurements at Beamline 11-BMof the Advanced Photon Source (APS) at Argonne Na-tional Laboratory, and the patterns were refined usingthe TOPAS software package. Refinements of the dataexhibit the expected pyrochlore phase as well as small( < < µ m par-ticle size were sealed into Kapton tubes using copper wireand epoxy in a He-filled glove-bag to provide a thermalexchange gas. The samples were measured in transmis-sion using an area detector. The 2D data were integratedto 1D diffraction data utilizing the Fit2D software. Cor-rections to obtain I ( Q ) and subsequent Fourier Trans-form with Q max =24 ˚ A to obtain G ( r ) were performed using the program PDFgetX2. Analysis of the totalscattering data was performed using the PDFgui soft-ware suite over the range 1.75 ˚ A − A .Magnetotransport measurements were carried out in aQuantum Design DynaCool Physical Property Measure-ment (PPMS) system. Cut portions of sintered pelletswere mounted with GE varnish in a four wire configura-tion using silver paint to create contacts. Current wasdriven perpendicular to the applied magnetic field, andvoltage was measured with a dc resistance bridge. Mag-netization data were collected using polypropylene cap-sules containing 20 mg of powder, and measured witha vibrating sample magnetometer (VSM) within a Dy-naCool PPMS or a MPMS3 Quantum Design SQUIDmagnetometer.X-ray absorption spectroscopy (XAS) measurementswere performed at Beamline A2 at the Cornell HighEnergy Synchrotron Source (CHESS), and X-ray mag-netic circular dichroism (XMCD) measurements wereperformed at Beamline 4-ID-D at the APS. Sieved pow-ders with ≈ µ m particle size were prepared on layersof tape to achieve a uniform sample thickness corre-sponding to nearly two absorption lengths. Both mea-surements were collected at the Ir L , absorption edges(2 p / , / → d ) in transmission geometry. At A2, theenergy of the incident X-ray beam was selected using adouble-crystal (cid:104) (cid:105) diamond monochromator that wasdetuned to reject higher harmonics, and the absorptionwas detected with ion chambers. At 4-ID-D, the inci-dent energy was selected using a double-crystal (cid:104) (cid:105) Simonochromator and circularly polarized X-rays were gen-erated in helicity-switching mode at 13 Hz using a dia-mond phase retarder. The absorption was detected usinga diode with lock-in amplification. To screen out spu-rious signal, XMCD measurements were repeated under µ H =+5 T and − III. EXPERIMENTAL RESULTSA. Average and local lattice structuremeasurements
Synchrotron X-ray diffraction data were collected fromsamples across the doping range 0 ≤ x ≤ .
08 andpowder patterns were indexed to the cubic space group
F d m . Bragg peaks were slightly asymmetric, indica-tive of a distribution of strains in the bulk which iscommonly observed for pyrochlore samples. In the py-rochlore structure there are four sites A B O O (cid:48) , whichare located at the 16 d , 16 c , 48 f , and 8 b Wyckoff posi-tions respectively. The cubic lattice constant a decreaseswith Ca substitution (Table I) in accordance with Veg-ard’s law, as expected since 8-coordinate Ca ionic radiiare nearly 1% smaller than Nd . Scattering power fromCa and O sites was sufficiently weak that direct refine-ment of Ca and O occupancies was unreliable. Therefore, | | || | | | || | || | || | || ||| ||| ||||| |||| || |||| |||||| ||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| | || | | | || | || | || | || ||| ||| ||||| |||| || |||| |||||| ||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| | || | | | || | || | || | || ||| ||| ||||| |||| || |||| |||||| ||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| FIG. 1. Synchrotron XRD patterns, all taken at 300 K. Cal-culated curves include the pyrochlore phase and the impurityphases Ir, IrO , and Nd O . Lines above the residual curvesindex only pyrochlore peaks. Ca concentrations were fixed at the WDXRF-measuredvalues and fixed on the A-sites. This A-site preferencefor Ca is consistent with the small increase in atomicdisplacement parameters on the A-site with Ca content.Refinement attempts to place Ca on the B-sites resultedin inferior fits.Occupancies refined for the A- and B-sites reflect aslight deviation from ideal stoichiometry. The XRD re-finements indicate less than 3% ‘stuffing’ through anti-site defects of excess Nd cations on the B-sites.The free coordinate u for O 48 f sites decreasesmarginally with Ca content. For the IrO octahedra,this trend signifies a reduction in trigonal compressiontoward octahedral symmetry ( u =0 . ◦ in the x =0 sample to 131.1(2) ◦ inthe x =0.08 sample. This remains less than the ≈ ◦ x .
02 0 . x WDXRF . . a (6 K) [˚ A ] 10 . . . a (25 K) [˚ A ] 10 . . . a (45 K) [˚ A ] 10 . . a (300 K) [˚ A ] 10 . . .
100 K u (O f ) 0 . . . (cid:54) Ir-O-Ir [ ◦ ] 129 . . . (cid:54) O-Ir-O [ ◦ ] 82 . . . . . . . . . . . . . . . U iso [˚ A ] 0 . . . U iso [˚ A ] 0 . . . B O [%] 96 .
48 95 .
59 98 . R wp [%] 10 .
54 12 .
83 9 . χ .
32 3 .
00 2 .
300 K u (O f ) 0 . . . (cid:54) Ir-O-Ir [ ◦ ] 130 . . . (cid:54) O-Ir-O [ ◦ ] 82 . . . . . . . . . . . . . . . U iso [˚ A ] 0 . . . U iso [˚ A ] 0 . . . B O [%] 96 .
68 96 .
04 99 . R wp [%] 10 .
01 12 .
12 8 . χ .
20 2 .
88 2 . x and sample x WDXRF from quantitative WDXRF analysis.Second, cubic lattice parameters a at several temperatures.Next, refined values at 100 K and 300 K: u for the O 48 f site; nearest-neighbor Ir-O-Ir bond angles; intra-tetrahedronO-Ir-O bond angles; occupancies of Nd and Ir on the A andB sites; isotropic atomic displacement parameters U iso forA- and B-sites; pyrochlore phase fractions less Ir, IrO , andNd O ; and Rietveld goodness-of-fit parameters R wp and χ .Note oxygen occupancies and U iso were fixed to 1 and 0.001,respectively. bond angle in the metallic A=Pr system, which isthe value associated with the onset of metallicity in thebandwidth-driven global MIT of the A Ir O series. Sowhile the increase in bond angle observed in the x =0.08system implies a sterically-driven increase in bandwidthof the valence band due to enhanced Ir-O orbital over-lap, the small bond angle change alone may not ac-count for metallicity in the substitution-driven MIT for(Nd − x Ca x ) Ir O .The evolution of the local structure was also analyzedvia X-ray PDF experiments (Fig. 2). No new atomic cor-relations were observed as Ca was introduced into the lat-tice or as the lattice was cooled through the MIT. Short-range order of the A- and B-sites is unchanged withinexperimental resolution for the measured samples for 6 FIG. 2. X-ray PDF refinement of the A B O local structure:(a) 300 K PDF data in circles, offset vertically, with fits inred solid lines. (b) Metal-metal distances between A and Bsites. (c) A- and (d) B-site isotropic atomic displacementparameters U iso . Dashed lines are guides to the eye. K ≤ T ≤
300 K. All PDF measurements fit well to thesite symmetry of the parent pyrochlore structure, withgoodness of fit R w values between 9% and 15%. This isconsistent with a previous synchrotron XRD study thatreported no symmetry change for Nd Ir O upon coolingto 4 K. Notably, the A- and B-sites’ isotropic displacementparameters, as determined from PDF refinements (Fig.2(c,d)), are also unchanged under varying Ca contentwithin uncertainty. This precludes Ca clustering effects or nanoscale chemical phase separation. Furthermore,metal-metal distances track well with lattice constants,indicating that minimal site disorder is introduced withCa substitution.
B. Magnetotransport measurements
Resistivity measurements shown in Fig. 3(a)reveal high temperature metallic behavior in all(Nd − x Ca x ) Ir O samples with a low temperature tran-sition into an insulating state (defined via the slope δρδT ).The temperature of the resulting MIT monotonically de-creases as a function of Ca content; but the “insulat-ing” state for samples with x ≥ T MIT =34 K transition in the parent( x =0) sample weakens and broadens with the additionof carriers. For samples with x ≥ ρ ( T ) upon cooling likely arises from carriers couplingto the residual Nd moments in the samples, and the up-turn temperatures in the high-Ca x ≥ µ SR) features linked to orderingor freezing of the Nd magnetic sublattice.
In contrast, the MIT in the low-Ca samples x< This isconfirmed by the (Eu − x Ca x ) Ir O system, a nonmag-netic A sublattice analogue, wherein the MIT and Ir sub-lattice antiferromagnetism are fully suppressed below 2 Kat a similar value of 0 .
3T the hysteresis in MR ascribed to a field-induced mag-netic phase transition remains resolvable up to x = 0 . C. Magnetization measurements
Magnetization measurements shown in the inset to Fig.5(a) reveal a weak irreversibility, hereby defined as the
FIG. 3. Magnetotransport measurements. (a) Relative resis-tivity from 2 K to 300 K, all at 0 T on zero field cooling. Ar-rows indicate the onset of δρδT >
0. Resistances are normalizeddue to variation in pellet densities; typical values are ρ (300K) ≈
10 mΩ · cm. (b) Relative magnetoresistance at 2 K onzero-field cooling, swept from µ H =0 → +9 →− → +9 T, witharrows to indicate field sweep direction. Note the hystereticsplitting, which is largest for the x =0 sample. (c) Hystereticdifferences of the magnetoresistance in b, for sweeps after theinitial 0 → +9 T sweeps (virgin curves). FIG. 4. Temperature-concentration phase diagram based onresistivity measurements. Inset: field-concentration phase di-agram, indicating magnetism on the Nd sublattice. All linesare guides to the eye. difference between field cooling (FC) and zero-field cool-ing (ZFC) sweeps, that persists across the doping range.This is conventionally ascribed to either domain wall for-mation or spin canting within the all-in-all-out networksof Ir and Nd spins. Irreversibility in magnetization dataappears at the same temperature where the low temper-ature resistivity changes slope. This connects the on-set of magnetic correlations or freezing with the onsetof the MIT for the low doping regime ( x ≤ .
02) as wellas with the low temperature increase in resistivity forthe high doping regime ( x ≥ . not sintered under vacuum, an addi-tional weak splitting between FC and ZFC data occursnear 120 K (Fig. 7), as was previously reported in theparent A=Nd system. This is a synthesis-dependent ef-fect with no resolvable influence on structural propertiesor T MIT or qualitative features in electron transport. Wediscuss this further in the Appendix.Field-dependent magnetization data for the(Nd − x Ca x ) Ir O series are plotted in Fig. 5(a).At T = 2 K, magnetization data are expected to bedominated by the Nd sublattice, and hystereticdifferences appear between sweeps of increasing anddecreasing fields. The splitting is illustrated by plots of δMδ ( µ H ) data shown in Fig. 5(b). In the ordered state ofthe parent system, applying a magnetic field polarizesthe Ising-like domains of both sublattices toward eitherAIAO or all-out-all-in (AOAI). Upon a substantialincrease in magnetic field, a second hysteretic featureappears, consistent with a spin-flop transition likely intothe Nd 3-in-1-out (3I1O) state reported for the parent
FIG. 5. Magnetization measurements: (a) Field dependenceof the isothermal dc magnetization, swept as in Fig. 3(b),with just H> T MIT values are in-dicated with arrows. (b) Solid lines are numerical derivativesof magnetization with respect to field, highlighting the hys-teretic splitting. The initial 0 → +9 T sweeps (virgin curves)are not shown. Dashed lines are the absolute differences be-tween sweep directions, magnified for clarity. system. Upon Ca substitution, this higher field spin-flop fea-ture decreases in onset field and vanishes for x =0 . FIG. 6. X-ray spectroscopic measurements: Ir L -edge XAS isindicated with solid lines, and XMCD with dots. XAS linesare offset by 0.3 for clarity. XMCD was measured at 5 K and ± x n h XAS n h BR (cid:104) L · S (cid:105) m tot L z /S z [ q e ] [ q e ] [¯ h ] [ µ B / Ir]0.00 5.00 5 5.7(2) 2.8(1) 0.0042(9) 3.0(4)0.02 5.02 5.06(6) 5.7(2) 2.8(1) − − − x Ca x ) Ir O from Ir L -edgeXAS and XMCD, as described in the text: stoichiometric andXAS-calculated number of holes n h ; branching ratios BR ;spin-orbit expectation values (cid:104) L · S (cid:105) ; total moments m tot at 5K and 5 T; and L z /S z = 2 m l /m s . fields extracted from each measurement (marked by *)are equal within measurement uncertainty and suggestthat the isothermal magnetoresistance data in Fig. 3(b)is governed by domain scattering effects at low dopingvalues and low temperatures. D. X-ray absorption spectroscopy and X-raymagnetic circular dichroism measurements
XAS data were collected at the Ir L , edges (Fig. 6,Table II) in order to probe the electronic structure of the5 d valence states. Qualitatively, the spectra for all sam-ples are quite similar, with each showing nearly identicalfine structure. This is consistent with the small struc-tural changes and low carrier concentrations associatedwith the steric and valence modifications arising from x< . t g manifold.XMCD measurements on all samples exhibit weaklocal Ir moments with resolvable signal just abovethe instrumental detection limit. Our applica-tion of the sum rules assumes a sizable magneticdipole term (cid:104) T z (cid:105)≈ . (cid:104) S z (cid:105) from Configuration Interactioncalculations on several other IrO systems BaIrO and Sr IrO . From the inclusion of (cid:104) T z (cid:105) , which slightlydecreases the effective spin moment, we calculate the Irtotal net moment m tot =0.004 µ B / Ir at 5 K and 5 T,which is unchanged within uncertainty for the x =0 and x =0.08 samples. This magnitude is similar to the values0.008 and 0.011 µ B / Ir reported from bulk dc magnetom-etry on the nonmagnetic A sublattice analogue systemsA=Y and A=Lu. Both XAS and XMCD data indicate that the samplesare near a J eff =1 / L and L whiteline intensities are defined as I L , = (cid:82) [ µ ( E ) − Θ( E )] dE ,where µ ( E ) is the XAS signal, Θ( E ) is a broadenedstep function centered on the inflection energy as ex-pected for isolated ions, and the integration range is overthe white line feature. The measured branching ratios BR = I L /I L are much higher than the statistical valuefor free ions of 2, which from the selection rules ∆ j =0 , ± d states are primarily 5 d / rather than 5 d / . Thus, BR is a relative measurement ofspin-orbit coupling for similarly prepared samples. Theratio L z /S z ≈ , coupledwith high BR ≈
6, provides a strong indication of an Ir J eff =1 / Now we comment more on hole doping and bandwidthchanges from the spectroscopic studies. For the mea-sured samples, the Ir L XAS inflection and peak energies(of 11216.2 eV and 11218.8 eV respectively) only changewithin monochromator repeatability ≈ . I L + I L relative to the parent value in Table II.Additionally, the XMCD peak energy is 0.4(2) eV higherfor x =0 .
08 compared to x =0, corresponding to a lower10 Dq value for the photoexcited core-hole state. Whilethis is not a proper measurement of 10 Dq for the groundstate electronic configuration, which is typically ≈ it suggests a small decreasein the e g − t g splitting with Ca substitution. IV. DISCUSSION
Substitution of Nd for Ca suppresses the MIT in(Nd − x Ca x ) Ir O primarily via hole doping with rel-atively minor structurally induced changes to the band-width. Once a doping level of x =0.08 is reached, signa-tures of Ir magnetic order vanish and transport behavesas a metal with weak disorder. We note that this valueof x for the paramagnetic metal phase is somewhat lessthan some theoretical predictions. Under hole dopingalone, the collapse of AIAO order in the Y Ir O is pre-dicted by one study to occur for n h =5.2 ( x =0.20). Wereconcile this difference by considering the much weakereffective correlations
U/t in Nd Ir O . A similar suppression of the MIT is observed upon dop-ing holes directly onto the Ir-sites via B-site alloying. Instudies of polycrystalline Nd (Ir − x Rh x ) O , Rh substi-tution was shown to strongly suppress T MIT = T N until aMIT was reached between x =0.05 and x =0.10. Sim-ilar to the case of 8% Ca-substitution in our study, for10% Rh-substitution both transport and magnetizationreveal a weak upturn and irreversibility respectively near5 K—likely arising from residual Nd moments freezing.The striking similarity between Rh and Ca substitutionsuggests comparable levels of hole-doping from each sub-stituted cation. From a simplistic consideration of chem-ical potentials, Ir sites are expected to transfer elec-trons to the lower energy J eff =1 / states, formingthe Ir /Rh valence states observed in the IrO -basedmaterial Sr Ir − x Rh x O . In comparison to carrier doping, external or “chem-ical” pressure (via introducing a larger A-site cation)modulating the bandwidth has a relatively gradual ef-fect on the MIT. Both hydrostatic pressure and substi-tution of Nd for Pr yield a decrease in T MIT = T N with the suppression of the transition from chemicalpressure proceeding gradually. A future detailed struc-tural study of Nd Ir O under pressure, and in partic-ular measurements of Ir-O bond lengths and angles in(Nd − x Pr x ) Ir O , would be insightful for resolving theprecise role of Ca steric perturbations in assisting anMIT.In the low temperature isothermal magnetization (Fig.5), the low field hysteresis for samples with x ≤ .
02 likelyarises from domain polarization of all-in-all-out and all-out-all-in magnetic domains, while at higher fields, thehysteresis observable only for x ≤ .
02 is indicative of afield-induced spin-flop from Nd AIAO to 3I1O order.This high field feature is consistent with the field iden-tified in polycrystal and single crystal studies on theparent Nd Ir O compound. The disappearance ofthis feature for x ≥ .
05 (i.e. for samples with a metallicground state) suggests that long-range Ir correlations arequenched and the only remaining magnetization arisesfrom short-range freezing of Nd/Ir moments coincidentwith the low temperature upturn in ρ ( T ). In this pic-ture, we assign the low field hysteresis for highly Ca-substituted samples to remnant short-range correlationsprimarily among Nd moments, as illustrated in the insetto Fig. 4.Kondo coupling has been predicted to allow the al-leged 3I1O Nd phase transition to be smooth. Thetemperature dependence of the resistivity in the insu-lating regime ( δρδT <
0) reported here does not fit wellto the empirical Hamann’s expression of the Kondo ef-fect ρ ∝ ln ( T K /T ), unlike in another study of Pr Ir O single crystals. This is likely attributable to the com-paratively weaker Kondo coupling J K ≤ which can be dominated by grainboundary-related scattering channels in polycrystallinesamples. The quadratic low-field negative magnetoresis-tance is consistent with Kondo physics in doped sam-ples; however future measurements on single crystals arerequired to fully explore this.An open question remains regarding the origin of theweak, but finite, Ir ferromagnetism present in both theinsulating parent x =0 and metallic x =0.08 samples. Thesmall net moment extracted from XMCD at the Ir L edges from both samples is identical within error andsuggests that the weak local Ir moment is not triviallytied to the zero field long-range ordered state. Rather,the relative strength of Ir spins that couple to the ap-plied field in both the insulating and metallic regimes isseemingly identical. We note here that the Ir XMCDsignal may arise from either/both net ferromagnetismor a reversible response to the field. These cannot bedistinguished without further measurements of the rem-nant magnetization obtained via hysteresis loops. Yet, ifsome of this signal is ferromagnetic, our measurementswould be consistent with a picture of the weak ferromag-netism arising from antiferromagnetic domain walls inthe parent insulating phase, wherein these domains per-sist locally in an electronically phase separated state wellinto the metallic regime. The fact that the polarized Irmoments at 5 T nearly match those observed in mag-netization measurements of Y Ir O and Lu Ir O withnonmagnetic A-sites supports this notion and suggests acommon origin to the weak ferromagnetism. V. SUMMARY
In summary, we report the suppression of both T MIT and the AIAO T N on the Ir sublattice via hole-dopingwithin (Nd − x Ca x ) Ir O . From a combined analysis ofdiffraction, PDF, and XAS data, we present evidence ofCa incorporation without clustering or phase separationon both local and average length scales. Calcium ionsonly weakly perturb the underlying structure with min-imal changes inferred to the corresponding bandwidth,and hole carriers associated with replacing Nd withCa cations instead drive the suppression of the lowtemperature MIT. For x > .
02, as the system entersa metallic ground state, both the charge transport andmagnetism remain influenced by fluctuations and disor-der on the Nd magnetic sublattice. Our results point to-ward the coincident suppression of long-range magneticorder and the charge gap in Nd Ir O as the parent spin-orbit Mott state is suppressed via carrier doping. Appendix: Effect of vacuum annealing on samplemagnetization
We now discuss the observation of magnetic irre-versibility near 120 K for samples of (Nd − x Ca x ) Ir O that were not annealed in vacuum. We compare sam-ples before and after vacuum annealing in Fig. 7, andattribute the additional feature to a secondary electronicphase within the bulk. We note that there is not a corre- FIG. 7. Irreversibility of the dc susceptibility for samplesbefore (circles, with features at 120 K) and after vacuum an-nealing (triangles, denoted ‘V’). sponding onset temperature in the Ir XMCD for samplesthat were not vacuum annealed.A similar magnetic feature may be present in other py-rochlore iridates produced by solid state synthesis. In onestudy of (Y − x Ca x ) Ir O samples there is one fixedtransition temperature near T MIT ( x =0)=160 K and an-other that appears near 190 K for x>
0. Since Y sitesare nominally nonmagnetic, the weak magnetism likelyarises due to the Ir sublattice. The origin of this vari-ety of magnetic feature is not resolved by our study, butit may be related to clustering of nonmagnetic Ir orIr sites. This hypothesis is supported by the measur-able signal that is not attributable to Ir in the X-rayphotoemission (XPS) spectrum for the Y Ir O samplein the aforementioned study. The likely source of Ir impurities is O (cid:48) vacancies, a known issue in the defect-accommodating pyrochlore structure type which mayscale with hole-doping. ACKNOWLEDGMENTS J. S. Gardner, M. J. Gingras, and J. E. Greedan, Reviewsof Modern Physics , 53 (2010). K. Matsuhira, M. Wakeshima, R. Nakanishi, T. Yamada,A. Nakamura, W. Kawano, S. Takagi, and Y. Hinatsu,Journal of the Physical Society of Japan , 43706 (2007). K. Matsuhira, M. Wakeshima, Y. Hinatsu, and S. Takagi,Journal of the Physical Society of Japan , 094701 (2011). K. Ueda, J. Fujioka, C. Terakura, and Y. Tokura, PhysicalReview B , 1 (2015). X. Wan, A. M. Turner, A. Vishwanath, and S. Y.Savrasov, Physical Review B , 205101 (2011). L. Savary and L. Balents, Physical Review Letters ,037202 (2012). W. Witczak-Krempa and Y. B. Kim, Physical Review B , 45124 (2012). L. Savary, E. G. Moon, and L. Balents, Physical ReviewX , 041027 (2014). I. Kimchi and A. Vishwanath, Physical Review B , 14414(2014). K. Ueda, R. Kaneko, H. Ishizuka, J. Fujioka, N. Nagaosa,and Y. Tokura, Nature Communications , 3032 (2018). A. B. Sushkov, J. B. Hofmann, G. S. Jenkins, J. Ishikawa,S. Nakatsuji, S. Das Sarma, and H. D. Drew, PhysicalReview B , 241108 (2015). T. Fujita, Y. Kozuka, M. Uchida, A. Tsukazaki, T.-h.Arima, and M. Kawasaki, Scientific Reports , 9711(2015). K. Ueda, J. Fujioka, and Y. Tokura, Physical Review B , 245120 (2016). Y. Machida, S. Nakatsuji, S. Onoda, T. Tayama, andT. Sakakibara, Nature , 210 (2010). W. Witczak-Krempa, G. Chen, Y. B. Kim, and L. Balents,Annual Review of Condensed Matter Physics , 57 (2013). M. Nakayama, T. Kondo, Z. Tian, J. J. Ishikawa,M. Halim, C. Bareille, W. Malaeb, K. Kuroda, T. Tomita,S. Ideta, K. Tanaka, M. Matsunami, S. Kimura, N. Inami,K. Ono, H. Kumigashira, L. Balents, S. Nakatsuji, andS. Shin, Physical Review Letters , 056403 (2016). R. Kaneko, M.-T. Huebsch, S. Sakai, R. Arita, H. Shi-naoka, K. Ueda, Y. Tokura, and J. Fujioka, Physical Re-view B , 161104 (2019). W. Zhu, M. Wang, B. Seradjeh, F. Yang, and S. Zhang,Physical Review B , 054419 (2014). A. Banerjee, J. Sannigrahi, S. Giri, and S. Majumdar,Physical Review B , 224426 (2017). A. Hammersley, S. Svensson, M. Hanfland, A. Fitch, andD. Hausermann, International Journal of High PressureResearch , 235 (1996). X. Qiu, J. W. Thompson, and S. J. Billinge, Journal ofApplied Crystallography , 678 (2004). C. Farrow, P. Juhas, J. Liu, D. Bryndin, E. Boˇzin, J. Bloch,T. Proffen, and S. Billinge, Journal of Physics: CondensedMatter , 335219 (2007). D. Haskel, Y. Tseng, J. Lang, and S. Sinogeikin, Reviewof Scientific Instruments , 083904 (2007). P. Telang, K. Mishra, A. K. Sood, and S. Singh, PhysicalReview B , 235118 (2018). J. N. Millican, R. T. Macaluso, S. Nakatsuji, Y. Machida,Y. Maeno, and J. Y. Chan, Materials Research Bulletin , 928 (2007). H. Takatsu, K. Watanabe, K. Goto, and H. Kadowaki,Physical Review B , 235110 (2014). S. M. Disseler, C. Dhital, T. C. Hogan, A. Amato, S. R.Giblin, C. De La Cruz, A. Daoud-Aladine, S. D. Wilson,and M. J. Graf, Physical Review B , 174441 (2012). H. Guo, K. Matsuhira, I. Kawasaki, M. Wakeshima, Y. Hi-natsu, I. Watanabe, and Z.-A. Xu, Physical Review B ,060411 (2013). Z. Tian, Y. Kohama, T. Tomita, H. Ishizuka, T. H. Hsieh,J. J. Ishikawa, K. Kindo, L. Balents, and S. Nakatsuji,Nature Physics , 134 (2016). E. Y. Ma, Y.-T. Cui, K. Ueda, S. Tang, K. Chen,N. Tamura, P. M. Wu, J. Fujioka, Y. Tokura, and Z.-X.Shen, Science , 538 (2015). S. M. Disseler, S. R. Giblin, C. Dhital, K. C. Lukas, S. D.Wilson, and M. J. Graf, Physical Review B , 060403(R)(2013). K. Ueda, J. Fujioka, B. J. Yang, J. Shiogai, A. Tsukazaki,S. Nakamura, S. Awaji, N. Nagaosa, and Y. Tokura, Phys-ical Review Letters , 056402 (2015). B. Thole, P. Carra, F. Sette, and G. van der Laan, PhysicalReview Letters , 1943 (1992). This (cid:104) T z (cid:105) value differs from studies of nearly-isolated[IrCl ] − systems in the next reference (Pedersen et al.2016), where (cid:104) T z (cid:105)≈ . (cid:104) S z (cid:105) , but note that the ligand andthe bonding environment are different. K. S. Pedersen, J. Bendix, A. Tressaud, E. Durand,H. Weihe, Z. Salman, T. J. Morsing, D. N. Woodruff,Y. Lan, W. Wernsdorfer, C. Mathoni`ere, S. Piligkos, S. I.Klokishner, S. Ostrovsky, K. Ollefs, F. Wilhelm, A. Ro-galev, and R. Cl´erac, Nature Communications , 12195(2016). M. A. Laguna-Marco, D. Haskel, N. Souza-Neto, J. Lang,V. Krishnamurthy, S. Chikara, G. Cao, and M. van Vee-nendaal, Physical Review Letters , 216407 (2010). D. Haskel, G. Fabbris, M. Zhernenkov, P. Kong, C. Jin,G. Cao, and M. van Veenendaal, Physical Review Letters , 027204 (2012). W. C. Yang, W. K. Zhu, H. D. Zhou, L. Ling, E. S. Choi,M. Lee, Y. Losovyj, C.-K. Lu, and S. X. Zhang, PhysicalReview B , 94437 (2017). J. Clancy, N. Chen, C. Kim, W. Chen, K. Plumb, B. Jeon,T. Noh, and Y.-J. Kim, Physical Review B , 195131(2012). B. Kim, H. Jin, S. Moon, J.-Y. Kim, B.-G. Park, C. Leem,J. Yu, T. Noh, C. Kim, S.-J. Oh, J.-H. Park, V. Durairaj,G. Cao, and E. Rotenberg, Physical Review Letters ,076402 (2008). L. Hozoi, H. Gretarsson, J. Clancy, B.-G. Jeon, B. Lee,K. Kim, V. Yushankhai, P. Fulde, D. Casa, T. Gog, et al. ,Physical Review B , 115111 (2014). H. Shinaoka, S. Hoshino, M. Troyer, and P. Werner, Phys-ical Review Letters , 156401 (2015). K. Ueda, J. Fujioka, Y. Takahashi, T. Suzuki, S. Ishiwata, Y. Taguchi, and Y. Tokura, Physical Review Letters ,136402 (2012). K. Ueda, J. Fujioka, Y. Takahashi, T. Suzuki, S. Ishiwata,Y. Taguchi, M. Kawasaki, and Y. Tokura, Physical ReviewB , 75127 (2014). J. Clancy, A. Lupascu, H. Gretarsson, Z. Islam, Y. Hu,D. Casa, C. Nelson, S. LaMarra, G. Cao, and Y.-J. Kim,Physical Review B , 054409 (2014). K. Ueda, J. Fujioka, C. Terakura, and Y. Tokura, PhysicalReview B , 121110 (2015). S. Nakatsuji, Y. Machida, Y. Maeno, T. Tayama, T. Sakak-ibara, J. van Duijn, L. Balicas, J. Millican, R. Macaluso,and J. Y. Chan, Physical Review Letters , 087204(2006). H.-J. Koo, M.-H. Whangbo, and B. J. Kennedy, Journalof Solid State Chemistry , 269 (1998). G. Giampaoli, J. Li, A. P. Ramirez, A. W. Sleight, andM. Subramanian, Inorganic Chemistry56