Pressure-Stabilized MnSb$_2$ with Complex Incommensurate Magnetic Order

Pressure -Stabilized MnSb 2 with Comp lex Incom mensurat e Magnetic Order Mingyu Xu 1,2 ,3 , Matt Boswell 1,4 , Qing-Ping Ding 2 , Peng Cheng 1 , Aashish Sapkota 2,3 , Qiang Zhang 4 , Danielle Yahne 4 , Sergey. L. Bud’ko 2 , 3 , Yuji Furukawa 2,3 , Paul. C. Canfield 2,3 , Raquel A. Ribeiro 2,3 , Weiwei Xie 1* 1 Department of Chemistry , Michigan State University , East Lansing , Michigan 48824, USA 2 Ames National Laboratory , Io wa State University , Ame s, Iowa 5001 1 , USA 3 Department of Physics and Astronomy , Iowa State University , Ames, Iowa 5001 1, USA 4 Neutron Scattering Division, Oak Ridg e National Laboratory , Oak Ridge, T ennessee 37831, USA Corresponding Authors: W eiwei Xie (xieweiwe@msu.edu) Abstract Marcasite-type compounds have been proposed as promising hosts of exotic magnetic quantum states, yet experime ntal realizations i n s toichiometric, disorder-f ree systems remain limited. Here, we report the high-pressure stabilization and magnetic characterization of MnSb 2 , a marcasite- type compound that is thermodynamically me tastable under ambient pressure. Single crystals were synthesized usin g a cubic multi-anvil press, a nd powder and single-crystal X-ray dif fraction con firm the orthorhombic Pnnm structure. These crystals are stable at ambient pre ssure for a long time up to be tween 450-500 K. Heat-capacity measurements reveal phase transition s at ap proximately 220 K and 118 K. Neutron diffraction uncovers an unconventional ma gnetic g round state below 220 K. Magnetic powder neutron diffraction re finements reveal possible multiple magnetic configurations that provide comparably ac ceptable fits to the experimental data. While most solutions are consistent with a s pin-density-wave (SDW) description, helical models systematically yield i nferior agreement factors. Across a broad range of models, t he Mn ordered moment reaches a ma ximum value of approximately 2 μ B and re mains predominantly collinear , with minimal canting along the c -a xis. At 200 K, the magnetic pr opagation v ector is q = (0 , 0.3975, 0.3783); upon cooling, the b c omponent increases toward 0.5, reflecting a temperature -de pendent evolution of the modulation. The n e ed for modification of the magnetic model betwee n high a nd low temperatures further highlight s the complex and strongly temperature -dependent nature of the magnetic order in this system. These results establish MnSb 2 as a pressure-stabilized ma rcasite ma gnet with a highly tunable, complex magnetic ground state and a compelli ng stoichiometric platform for exp l oring unconventional magnetic behavior , including potential altermagnetism. Introduction When ma gnetism i ntertwines with band topology in quantum materials, it can, in some cases, give rise to a re cently identified class of magnetic systems characterized by collinear spin arrangements with zero net magnetization, ex hibiting time -reversal-symmetry-breaking responses 1– 8 and spin-split electronic band structures 9– 16 . These properties, referred to as altermagnetism, combine ze ro-net-moment magnetic order with symmetry-allowed spin splitting relevant for spintronic applications and have motivated intense theoretical a nd experimental inter est in identifying chemically c lean material platforms that realize this ma gnetic p hase 8,17– 31 . Marcasite -type compounds, i n particular , have emer ged as promising candidates due to their low c rystal symmetry and symmetry-allowed spin s plitting in antiferromagnetic states . FeSb 2 is a narrow-gap, str ongly correlated semic onductor 32 – 36 t hat has been theoretically predicted to host a collinear altermagnetic state upon C o substitution or hole doping 37 . Ex perimentally , however , FeSb 2 exhibits no long-ra nge magnetic order between 1.8 and 300 K. Stabilizing magnetic or der in FeSb 2 therefore requires deliberate tuning of its electronic structure. While chemical substitution a nd external pressure are widely used to modify band filling and band width, substitution often i ntroduces chemical disord er that complicates in terpreta tion, a nd pressure -dependent magnetic me asurements a re experimentally nontrivial, e specially for antiferromagnetism . This all m otivates the search for chemic ally clean routes to stabilize magnetic order in the FeSb 2 structure class. A promising route to this goal is provided by MnSb 2 , a high-pressure marcasite-type compound t hat shares the same orth orhombic structure a s FeSb 2 but incorpo rates Mn with a larger , and often more ro bust, local magnetic moment and intrinsic hole do p ing 38 (relative to FeSb 2 ). Whereas MnSb 2 d oes not form at ambient pre ssure, it can be synthesized under high-pressure conditions a nd quenche d into a metastable form at ambient pressure 39 . Across the marcasite f amily , CrSb 2 exhibits robust a ntiferromagnetic order with a Née l temperature near 273 K 40 , whereas FeSb₂ remains nonmagnetic, s uggesting that ma gnetic order e merges through band filling b y h ole doping . By analogy , MnSb₂ represents a chemically ordered, stoichiometric platform in which unconventional states may be realized. Results and Discu ssion MnSb 2 wa s sy nthesized at 3.3 G Pa pressure a nd 490 °C using a Rockland Re search cubic multi-anvil press. After maintaining this temperature for 24 hours, a high -quality MnSb 2 prod uct, with more th an h alf of the final materia l consisting of sub -mil limeter-sized MnSb 2 single crystals that can be e asily se parated me chanically from the surrounding MnSb 2 polycrystalline ma trix , wa s obtained . (See the Experimental Method section f or further details.) The single-crystal X-ra y dif fra ction ( SC XRD) a nalysis of MnSb 2 , summarize d in T ables S1 and S2 , illustrates the crystal structure of MnSb 2 shown in the ri ght inset of Fig. 1 , which adopts the orthorhombic Pnnm s pace group and features edge -sharing M nSb 6 octahedra. The unit cell contains two crystallographically distinct a tomic sites: Mn atoms occupy the 2 a W yckof f position, while Sb atoms reside a t the 4 g positions. V acancies and site mixing we re considered during the refinement, but no str uctural disorder or clear vacancies were detected. Fig. 1 | Phase identification an d structural characterization of M nSb 2 . Powder X-ray diffraction pattern of MnSb 2 with Rietveld refinement. Red circle s in dicate experimental d a ta, th e black line the calculated pattern, and the blue line the dif ference curve. Th e main phase is indexed to MnSb 2 , with a minor impurity phase identified as eleme ntal Sb (about 1% by weight) . Righ t inset : crystal struc ture of MnSb 2 . Left inset : optical image of representative single cry stals; grid spacing is 1 mm. T o assess th e phase purity of samples synthesized u nder h igh- pressure conditions, p owder X- ray dif fra ction (PXRD) mea surements were performed using a b ulk sample with single crystals i n the polycrystalline matrix, as shown in Fig. 1 . The experimental data (red c ircles) were analyzed by Rietveld refineme nt (b lack line) using the GSA S -II software package 41 , and t he difference betwe en the observed a nd ca lculated patterns is shown by t he blue line. The minor phase of elemental Sb is observed. The PXRD re sults a re consistent with single -crystal X-ray diffraction mea surements and with the previous report. 39 T he left inset of Fig. 1 s hows representati ve as -grown crystals, which display metallic luster and well-defined faceted morphologies, consistent with high crystallinity . Fig. 2 | ( a ) Temperature-dependent magnetiza t ion of MnSb 2 measured unde r field-cooling (FC) and field-warming (F W) across d iffer ent te mp erature ranges: 300 -350 K (black), 300-400 K (blue), 30 0- 450 K ( yellow), and 300-500 K (red). A significant inc rease in magnetization a bove 450 K indicates the onset of decomposition. ( b ) Powder X-ray diffraction p atterns collected before and after the magnetization measurements are sh own i n ( a ) . The emergence of Mn 1.1 Sb reflections and an increase in Sb peak intensity after heating abo ve 500 K confirm the thermal decomposition of MnSb 2 . T o determine the t hermal stability ran ge of MnSb 2 at ambient pressure, temperature- dependent ma gnetization mea surements were performed under b oth field -wa rming (FW) an d fie ld- cooling (FC) conditions, a s shown in Figure 2 a . Below 450 K, the ma gnetization curves exhibit no significant dif ference between the FW and FC protocols, indicating thermal and magnetic stability within this temperature range. However , abov e 450 K, a pronounced increase in magn etic mom ent is observed, suggesting the onset of decomposition a nd the format ion of M n 1.1 Sb, a known ferromagnetic impurity phase. This observation is corroborated by powder X -ray diffraction data collected before and after the ma gnetization mea surements. As shown in Figure 2 b , additional reflections c orresponding t o Mn 1.1 Sb emer ge, a nd the intensity of th e Sb peak increases signifi c antly after heating ab ove 500 K, confirming the partial decomposition of M nSb 2 . These results i ndicate that while MnSb 2 is metastable at a mbient pressure, it remains structu ra lly and magnetically stable at temperatures up to ~450 K, enabling long -term low-temperature studies without p hase transformation. Moreover , within the tempera ture range of 1.8 K to 300 K, magnetiza ti on measurements (see Figure 5 ) show no signatures of fe rromagnetic or other phase transitions in MnSb 2 . The field-dependent ma gnetization displays predomin ant paramagnetic behavior with minor ferromagnetic contributions attributable to trace Mn 1.1 Sb impurities. Collectively , these results confirm that MnSb 2 is suitable for extended investigations at lo w temperatures under ambient conditions, without under going structural or magnetic phase transitions. Fig. 3 | Electri cal trans port , sp ecific he at, and magnetization of M nSb 2 . ( a ) Electrica l resistance of si ngle-crystalline MnSb 2 measured u pon cooling (black) and wa rming (red) between 300 K and 1.8 K, showing negligible thermal hysteresis. Inset : T empera ture derivative d R/ d T with two anomalies. ( b ) T emperature -dependent specific heat showing three anomalies at T 0 , T 1, a nd T 2 , defined by the midpoint between on se t and offset temperatures, shown in the right insets. Data from 200 -280 K are obtained by point- by -point background subtractio n of N -grease measurements. The temperature range is fro m 4 K -2 80 K. The left inset shows a comparison of H -grease and N-grease measurements. T 1, and T 2 anomalies are c learly shown in both me asurements. ( c ) Field-depend ent magnetization ( M - H ) of polycrystalline MnSb 2 me asured at selected temperatures. Inset: representative M - H curves at 300 K. ( d ) T emperature-dependent magneti zation ( M - T ) of polycrystalline MnSb 2 measured und e r appli ed magnetic fields. Inset : t emperature derivative of MT , showing two anomalies between 1.8 K a nd 300 K. T he black arrow indicates the anomalies sh own in the measurement and the criteria of anomalies’ temperature and width. T o pr obe the magnetic and electronic properties of MnSb 2 , temperature-dependent elec trical resistance measurements we re performed on single -crystalline samples upon cooling from 3 00 K to 1.8 K a nd during subsequent warming, as shown in Fig. 3 a . The resistance curves from the c ooling and warming cycles overlap c losely , in dicating neg ligible th e rmal hysteresis. The residual resistance ratio (RRR), defined a s R(300K)/R(1.8K), is approximately 60 . The inset of Fig. 3 a displays the temperature derivative of the resistance (d R /d T ). T here are two clea r breaks in slope visible in the raw data an d two broad steps in d R /d T around 125 K and 220 K, as i ndicated with the arrows in Fig. 3 a and the inset. Complementary th e rmodynamic information is p r ovided by temperature -dependent specific h eat measurem ents on polycrystalline MnSb 2 performed a t zero magn etic field . As shown in Fig. 3 b , three reproducible anomalies are observed at app roximately 1 1 8 K ( T 0 ), 219 K ( T ₁) , an d 2 31 K ( T ₂). These features are independently confirmed using an alternative measurement protocol over a different temperature ra nge . Notably , the a pplication of a magnetic field of 90 kOe produces negligible chang e s in specific heat ( Fig. S3 a ), demonstrating that th e transition is lar gely insensitive to external magnetic fields. Integration of the excess specific heat yields a total ma gnetic entropy change of approximately    ( Fig. S3 b ) using the r eference of FeSb 2 specific heat, wh ich is close to the Mn high-spin state . Curiously , the we ak field dependence of the upper specific heat fe ature might suggest an antiferromagnetic transition t hat couples only we akly t o uniform magnetic fields. T o asse ss the uniform magnetic response of MnSb 2 and the presence of any net magnetic moment, magnetic susceptibility a nd field-dependent ma gnetization measurements we re performed. Fig. 3 a shows the field-dependent magnetization of polycrystalline MnSb 2 me asured up to 70 kOe at selected temperatures. The magnetization exhibits only weak tempera ture depend e nce, remains unsaturated over the en tire field range, and displays a very s mall net moment. A weak ferroma gnetic contribution is observed and attribute d to trace a mounts (<1%) of the Mn 1.1 Sb i mpurity , w hich is below the detection limit of X-ra y diffraction but was d etected by NMR mea surements ( Fig. S5 ). The inset shows represe ntative M ( H ) curves measured at 300 K in five q uadrants, con firmin g the absence of intrinsic ferromagnetic hysteresis. Fig. 3 b pr esents the temperature -dependent magnetization of polycrystalline MnSb 2 measured under a 10 kOe app lied field . There are two anomalies between 2 K and 300 K, T 0 = 122 K and T ’ 1 = 226 K, indicated by the black arrows. This o bservation is consistent with elec tric transport and the heat-capacity measurements . The inset displays the temperature derivative of MT , w hich shows two anomalies. This behavior is consistent with an amplitude-modulated antiferromagnetic ground state and highlights the limited se nsitivity of uniform magnetization measurements to this type of magnetic order as well a s a relatively large contribution to M ( T ) from a fe rromagnetic impurity . T o elucidate the microscopic origin of this magnetic transit ion, we next turn to neutron sca ttering mea surements, which directly probe the magnetic o rdering wave vector and spatial distribu tion of the ordered moments. Fig. 4 | Neutron diffr action and magnetic structure of MnSb 2 . ( a ) Powder neutron diffraction patterns of MnSb 2 collected at 308, 250, 200, 150, 100, and 4 K (logarithmic intensity scale). ( b , c ) Rietveld refinements of the po w der neutron diffraction data collected at 308 K (an aluminum can contribu tes the impurity peaks of Al in Fig . 4 b ) , 200 K, and 4 K, illustrating the emergence of magnetic reflections upon cooling. Powder neutron dif frac tion me asurements on MnSb 2 were ca rried out between 308 K and 4 K ( Fig. 4 ). Upon cooling below ~200 K, we ll below the ~220 K upper magnetic transitions , additional Bragg reflections (shown in the red arrows) of magnetic origin e merge ( Fig. 4 a ). Similar magnetic peaks are fou nd at 150 K and new peaks (s hown in the b lac k arrows) at 100 K and 4 K, tem peratures below the l ower magnetic transition . Refinements of d iffraction patterns collected a bove ( Figs. 4 b ) and below ( Fig. 4 c ) the show that orthorhombic MnSb 2 dev elops a complex magnetic ground state that evolves with tem perature. T wo irreducible representations can be utilize d to a nalyze the structure that is related by symmetry . The magnetic powder r efinement allows many magnetic structures to fit reasonably well. Although most stru ctures are aligned with a spin density wave formation , trying a helical magnetic structure overall results in a lo wer quality of fit. The mul tiple m odels that can describe this system make it challenging to determine the direction of the spi n density wave specifically a nd require polarized neutrons to better understand the structure. However , in ma ny models , the Mn mo ment reaches a maximum of 2 μ B and tends to be c ollinear with little orientation along the c- axis. The propagation vector q is (0, 0.3975, 0.3783) at 200 K, and the b component of the propagation vector increases as temperature decreases, and the value ap proaches 0.5. W ith the c hanging temperature , the modulation of the spin wa ve changes and typically requires changing t he model from high to low temperatures, further highligh t ing the complexity of the magnetic order within the system. T he ma gnetic ground state of orthorhombic MnSb 2 is fundamentally distinct from both conventional helical magnets and simple spin-d ensity-wave (SDW) systems. Importantly , the st rictly collinear antiferromagnetic order with zero net magnetization , combined with sublattice -depen dent anisotropy and broken spin degeneracy at fi nite wave vecto r , places MnSb 2 in close pr oximity to the emergin g class of altermagnetic materials. Fig. 5 | Calculated electronic ban d structure of M n Sb 2 . ( a ) Non-magnetic (spin-unpolarized) state. ( b ) Antiferromagnetic state. T o evaluate th e magnetic ground state o f MnSb 2 , first-principles calculations were perfo rmed using the experime ntally determined singl e-crystal structure, considering non-magnetic (NM), ferromagnetic (FM), and a simple collinear antife rromagnetic (AFM) confi gurations with antipara llel Mn moments in the primitive cell. Although neutron dif fraction reveals a considerably mor e complex magnetic struc ture, calculations with a simple c ollinear AFM configuration provide a useful reference for evaluating the magnetic instability of the e lectronic struc ture. In addition, the limited experimental constraints on the f ull magne tic struc ture from unpolar ized neutron di f fraction and the computational cost of large magn etic supercells make su ch simplified ca lcul ations a pr actical first step. The c alculated total ener gies ( T able S3 ) show that the AFM state is the most stable, lying 80.79 meV below the NM configuration, while the FM state is also ener getically favored (69.68 me V below NM), indicating close ener g etic competition among magnetic states. In the AFM configuration, the calculated local magnetic moment on Mn is 2.81 μ B . The c alculated electronic band structures are shown in Fig. 5 . In the NM state ( Fig. 5 a ), a flat b and app ears n ear the F ermi level, producing a pronounced peak in the density o f states and signaling electronic instability . Although Mn Sb 2 shares the same nomi nal valenc e electron count as FeSb 2 , the partial occupation of states at the Fermi level confirms metallic behavior rather than the insulating ground state of FeSb 2 . Upon inclusion of spin polarization in the AFM state ( Fig. 5 b ), the flat band shifts below the Fermi level, and the density of states at the Fermi level is strongly reduc ed, leading to the for mation of a pseudogap and a more stable electronic structure. Importantly , the AFM state b reaks time -reversal symmetry while preserving a net zero magnetization, a sym metry condition that can allow momentum-dependent spin splitting of electronic bands in low -symmetry crystals. This symmetry condition is consistent with the collinear , finite-  antiferromagnetic o rder and orthogonal subl attice anisotropy resolved by n eutron diffraction, although a detailed analysis of altermagnetic band splitt ing is beyond the scope of the present calculations. T o gether , these results indicate that MnSb 2 satisfies the essential ener getic and symmetrical pre requis ites for alte rmagnetic behavior , motivating future momentum-resolved experimental and theoretical studies. In summ ary , alter ma gnets have recently emerged as a distinct class of magnetic materials that combine collinear antiferromagnetic order with spin -split electronic band structures and ze ro net magnetization, offering new opportunities for quantum and s pintronic fun ction alities. In this work, we establish MnSb 2 as a chemic ally c lean, pressure-stabilized marc asite-type c ompound h osting an unconventional magnetic ground sta te. Using high -pressure synthesis, the rmodynamic me asurements, neutron dif fraction, and first -principles calculations, we show th at MnSb 2 does not develop a single, rigid magnetic order below a well-defined T N . Instead, ma gnetic entr opy be gins to be relea se d near ~230 K , accompanied by the appearance of incomm ensurate magnetic B ragg re flections bel ow ~ 200 K. A second thermodynamic anomaly near ~ 1 18 K coincides with the emergence of additional magnetic peaks, indicating a reconstruction of the ordered state. Remarkably , specific heat and neutron refineme nts reveal that t he magnetic structure continues to evolve throughout the ordered regime: the propagation vector shifts systematically with temperature (with the b -component trending toward 0.5). The relating ground state is described a s a complex S DW with the available data d o not uniquely d e termine th e modulation direc tion, a nd polarized neutron experiments will b e required for definitive re solution . These fin dings e stablish MnSb 2 as a rar e stoichiometric platform in which an incommensurate SDW remains h ighly tunable at low temperature, highlighting an unusual interplay between anisotropy , competing exchange interactions, and the underlying electronic structu re. Experi mental Meth ods High-Pressur e Synthesis: M nSb 2 was synthesized using a two-step me thod , whic h is different from the reference 39 (lower pressure and temperature used in this experiment), involving (i) the preparation of a fin ely mixed Mn –Sb precursor and (ii) a subsequent high-pressure, high-temperature reactio n. In the first step, manganese metal (CZ-1-11 0) and anti mony shot (99.999%) were combined in a 1:2 molar ratio and loaded into a 1. 7 mL fritted alumina Canfield Cr ucible Set (CCS, LSP Industrial Ceramics, Inc.). 42 The crucible was se aled in a silica a mpoule under an ar gon atmosphere (~1/ 3 a tm). Silica wool was packed above and below the CCS to stabilize the crucible du r ing centrifugation and to contain any escaped liquid. The a mpoule wa s h eated in a b ox f urnace to 760 °C o ver two hours, held at this temperature for 10 hours, and then rapid ly centrifuged to s eparate the molten and non - molten phases. The solid residue retained on the frit disc a ppeared as a black solid , likely a minor oxide crust, w hile the solidified, decanted melt was collected, ground i nto a fine powder , and pressed into a pellet using a die press. In the se cond step, the pellet was loaded into a boron nitride (BN) crucible (7 mm length, 5.7 mm inner diameter), with a ny remaining void space filled with BN po wd er . The assembly was su bjected to 3.3 GPa pressu re at room temperature using a Rockland Research cubic multi-anvil press (~20mm anvil size) and then heated to 490 °C. This temperature was selected as the maximum before the Mn- Sb mix begins to melt, determined u sing the sample current (power) as a function of tem perature curve, as shown in Fig. S7 c . The reason to prevent the mixture f rom melting during the expe riment is to avoid a l arge a mount of Mn 1.1 Sb impurity from incongruent melting, as shown in Fig s. S7 b and d . After maintaining this temperature for 24 hou rs , the sample was quenched to r oom t emperature before slowly releasing the pressure. This two -step p roce ss yields a h igh-quality M nSb 2 produ ct, with more than ha lf of the final ma terial con sisting of sub-millimeter- sized single crystals that can be easily separated fro m the surrounding polycrystalline matrix. Single Crystal X -ray diffraction: T o determine th e crystal structure o f the obtained single cr ystal, the sample with dimensions 0.218 × 0.158 × 0.138 mm ³ was picked u p, mounted on a nylon loop with Paratone oil, and measured using an XtalLAB Syner gy , Dualflex, Hypix single crystal X -ray dif fra ctometer with an Oxford Cryosystems 800 low -temperature devic e. Da ta were collected using ω scans with Mo K α ra diation ( λ = 0.71073 Å) . The total number of r uns and images was based on the strategy c alculation from the program CrysAlisPro 1.171.43.92a (Rigaku OD, 2023). Data reduction was performed with correction for L orentz polarization. The i ntegration of the data u sing an orthorhombic unit c ell yielded a total of 3540 re flections t o a maximum θ an gle of 40.13° (0.55 Å resolution), of whi c h 4 78 were independent (average redundancy 7.406, completeness = 9 8.8%, Rint = 5.61%). A numerical absorption correction was applied based on Gaussian integration o ver a multifaceted c rystal model 43 . Empirical absorption correc tion used spherical harmonics, implemented in the SCALE3 ABSP ACK scaling algorithm. 44 The structure was solved and refined using the SHELXTL Softwa re Package 45,46 . Powder X-ray d iffraction: T o examine the phase information, the powder X -ray diffraction (PXRD) analysis was p erformed subsequent to t he synthe sis process. The crystals we re ground using an agate mortar and pestle to achieve a homogeneous powder . This powdered sample wa s then unif o rmly distributed on a single c rystalline silicon sample holder , designed for ze ro background measurements, with a minimal application of v a cuum grease to secure the powder in place. The PXRD data acquisition spanned a 2 θ range from 15° to 80°, utilizing increme ntal steps of 0.01° and a fixed d well time of 3 seconds p e r step. These measurements were conducted using a Rigaku Mi niFlex II powder dif fra ctometer , employing Bragg-Bre ntano geometry c oupled with Cu K α radiation (λ = 1.5406 Å). The refinement of the powder X-ray data was executed using the GS AS-II software suite 47 . Physical Properties Measurements: T emperature-, magnetic-field-dependent DC ma gnetization data and re sistance measurements were collected using Quantum Design (QD), Magnetic Property Measurement Systems (MPMS and MPMS3), and Physical Pr operty Measurement S ystems (PPMS). During the measureme nts of single crystals, the field is along a ran dom direction of the crystals due to the poorly defined facets and the small size of the crystal. Th e samples are placed between two collapsed plastic straws , with the t hird uncollapsed straw providing supp ort as a sh e ath on t he outside or a q uartz sample holder . Samples were fi xed on the straw or quartz sa mple holder with GE -7031- varnish. AC electrical resistance measurements were performed in a standard four-c ontact geometry using the ACT option of the PPM S, with a 3-mA current a nd a frequ ency of 17 Hz. 50µm diame ter Pt wires were bonded t o the samples with silver p a int (DuPont 4 929N) with con tac t resistance valu es of about 2-3 Ohms. T em perature-dependent specific he at m easurements on the MnSb 2 in a mass of a bout 6 mg polycrystalline sample were ca r ried out using a Quantum Design, Physical Property Measurement System (PPMS DynaCool) in the temperature range of 200 -280 K ( H grease), 150 K- 280 K (N gre as e with addenda measu rement points that are the same as the sample measure ment), 107 K-127 K( N g rease), and 4 K- 200 K (N grease). Neutron Powd er Diffraction: T o determine whether magnetic ordering exists, neutron powder dif fra ction (NPD) measurements were performed using a time - of -flight p owder d iffractometer , POWGEN, at the Spalla tion Neutron Sou rce at Oa k Ridge Nation a l Laboratory . Approximately 2.6 g of samples were prepared a fter grinding se veral single crystals to a fine powder and passing them through a 45µm mesh sieve. Diffraction patterns were collected a t 30 8 K for 2 hours 36 minutes using a neutron beam with a center wavelength of 1.5. Rietveld refine ments using GSAS-II softwa re 47 and Fullprof suite 48 were performed to determine the crystal and ma gnetic structures, respectively . Determination of the magn etic structure was performed on HB-2A POWDER beamline at the High- Flux Isotope Reactor (HFIR) located at Oak Ridge National Lab (ORN L). Roughly 1 g of MnSb 2 was loaded into a n aluminum can backfilled with He . Diffraction patterns were measured with a vertically focused Ge mon oc hromator to select the 2.41 Å wavelength while a collimation of open- open-12 was used. Magnetic stru c ture a nalysis was performed with SARAh Representation Ana lysis and SARAh Refine and refined with the Fullprof suit e software 49 . Nuclear Magnetic R esonance: Nuclear magnetic resonance (NMR) me asurements of 55 Mn ( nuclear spin I = 5/2, gyromagnetic ra tio  N /2  = 10.50 MHz/T), 121 Sb ( I = 5/2,  N /2  = 10.189 MHz/T ), and 123 Sb ( I = 7/2,  N /2  =5.517 MHz/T) nuclei were conducted using a laboratory- built phase-coherent spin-echo pulse spectrometer on polycrystalline powder sa mples. The NMR spectrum was obtained under magnetic fields by swe eping the magnetic field a t a fixed N MR frequency of 67 MHz, whi le the z ero magneti c field NMR spectrum was me asured by plotting spin -echo intensity as a function of NMR frequency . DFT Calculation: Density Functional Theory (DFT) c alculations were carried out using ver sion 7.3.1 of the Quantum ESPRESSO package to evaluate the total energies of MnSb 2 in non-magnetic, ferromagnetic, a nd antiferromagnetic configurati ons. 50 T he calcu lations employ e d ultrasoft pseudopotentials and the Perdew -Burke-Ernzerhof (PBE) exchange-c orrelation functional withi n the generalized gradient approximatio n (GG A). 51 – 53 A plane-wave kinetic ener gy cutof f of 300 R y was used, with a charge density cutoff set to 3600 R y ( 12 times t he wavefunction cutoff). Brillouin zone integration was performed usi ng a 7×7×13 Monkh orst-Pac k k-point grid. 54 T otal energy convergence was ensured with a thr eshold of les s than 1 me V per atom. The elec tronic self -consistency wa s achieved using t he Davidson diagonalization algorithm, wit h a convergence criterion of 10 -9 R y . 55 High-symmetrical k -point paths for band struc ture calculations were generated using the Spglib library . 56,57 Acknowled gment s The work at MSU was supported by the Department of Energy under DE - SC -0023648. Th e work at Ames National Laboratory was supported by the U.S. Department of E nergy, Office of Science, Basic Energy Sciences, Materials Sciences, and Engineering Division. Ames National Laboratory is operated for the U.S. Department of Energy by Io wa State University under c ontract No. DE-AC02-07CH11358. C.P. in Weiwei Xi e ’ s group is supported by NSF-DMR-2422361. A portion of th is research used resource s at the Spallation Neu tron Source, a DOE Office of Science User Facil ity operated by the Oak Ridge National L aboratory. This m aterial is based upon work supported by t he U.S. Department of Energy, Office of Sci ence, Office o f Workforce Development for Teachers and Scientists, Office of Science G raduate Student R esearch (SCGSR) p rogram. The SCGSR progra m is administe r ed by t he Oak Ridge In stitute for Science and Education (ORISE) for the DOE. ORI S E is managed b y ORA U unde r contract number DESC0014664. All opinions expressed in this paper are t he author’s and do not necessarily r eflect the policies and views of DOE, ORAU, or ORISE. No animals were harmed in the course of this research. Reference (1) Šmejkal, L.; Sinov a, J.; Jungwirth, T . Be yond Conv enonal Ferromag n esm and Anf erromagnesm: A Phase with Nonr elavisc Spin and Crys tal Rot aon Symmetry . Phys. Rev . X 2022 , 12 (3), 031042. h ps://doi.org /10.1103/Ph ysRevX.12.031042. (2) Ma, H. - Y .; Hu, M.; Li, N.; L iu , J.; Y ao, W .; Jia, J. - F .; Liu, J. Mulfunconal An f erromagnec Mat erials with Giant Piezomagne sm and Noncollinear Spin Curr ent. Nat. Commun. 2021 , 12 (1), 2846. hps://doi.or g /10.1038/ s41467 - 021 -23127- 7. (3) Shao, D . -F .; Zhang, S. -H.; Li, M.; Eom, C. -B.; T symb al, E. Y . Spin-Neutral Current s f or Spintr on ics. Nat. Commun. 2021 , 12 (1), 7061. hps://doi.or g /10.1038/s41467-021-26915- 3. (4) Šmejkal, L.; Hellenes, A. B.; Gonzále z -Hernández, R.; Sinova, J.; Jungwirth, T . Giant and T unneling Magnetoresis tance in Unc onvenonal Collinear Anf erromag n ets with Nonrela visc Spin - Momentum Coupling. Ph ys. Rev . X 2022 , 12 (1), 011028. hps://doi.or g /10.1103/Ph ysRevX.12.011028. (5) Karube, S.; T anaka, T .; Sug awar a, D.; K ad oguchi, N.; K ohda, M.; Nia, J . Observaon of S pin - Splier T orque in Colli n ear Anf erromagnec RuO 2 . Phys. R ev . Le. 2022 , 129 (13), 137201. hps://doi.or g /10.1103/Ph ysRevLe .129.137 201. (6) Naka, M.; Ha yami, S.; K usunose, H.; Y anagi, Y .; Motome, Y .; Seo , H. Spin Curren t Gener aon in Or ganic Anf erromag nets. Nat. Commun. 2019 , 10 (1), 4305. hps://doi.or g /10.1038/s4 1467- 019 -12229- y. (7) Šmejkal, L.; Gonz ález-Hernández, R.; Jungwirth, T .; Sinova, J. Crys tal Time -Rev ersal Symmet ry Breaking and Spont aneous Hall E ect in Collinear Anf erromagnets. Sci. Adv . 2020 , 6 (23). hps://doi.or g /10.1126/ sciadv .aaz 8809. (8) Šmejkal, L.; MacDonald, A. H.; Sinova, J.; Nakat suji, S.; Jungwirth, T . Anomalous Hall Anf erromagnets. Nat. R ev . Mater . 2022 , 7 (6), 482–496. hps://doi.or g /10.1038/ s41578 - 022 - 00430- 3. (9) Ahn, K. - H.; Hariki, A.; Lee, K. -W .; K u neš, J. An f err om agnesm in RuO 2 as d-W av e Pomer anchuk Inst ab ility . Phys. R ev . B 2019 , 99 (18), 184432. hps://doi.or g /10.1103/Ph ysRe vB.99.184432. (10) Ha yami, S.; Y anagi, Y .; K usunose , H. Momentum-Dependent Spin Spling by Collinear Anf erromagnec Or dering. J. Ph y sical Soc. Japan 2019 , 88 (12). hps://doi.or g /10.7566/ JPSJ.88.123702. (11) Y u an, L. -D.; W ang, Z.; Luo, J. -W .; Rashba, E. I.; Zung er , A. Giant Momentum-Dependent Spin Spling in Centr osymmetric Lo w-Z Anf erromagne ts. Phys. Rev . B 2020 , 102 (1), 014422. hps://doi.or g /10.1103/Ph ysRevB .10 2.014422. (12) Ha yami, S.; Y anagi, Y .; K usunose , H. Bo om-up Design of Spin-Split and Res hap ed Electr onic Band Structures in Anf erromagnets wit hout Spin -Orbit Coupling: Procedu re on the Basis of Augmente d Mulpoles. Phys. R ev . B 2020 , 102 (14), 144441. hps://doi.or g /10.1103/Ph ysRevB .10 2.144441. (13) Y u an, L. -D.; W ang, Z.; Luo, J. -W .; Zunger , A. Predicon of Low-Z Collinear an d Noncollinear Anf erromagnec Compounds Ha ving Momentum -Dependent Spin Spling E ven without Spin- Orbit Coupling. Phys. Re v . Mater . 2021 , 5 (1), 014409. hps://doi.or g /10.1103/Ph ysRevM aterials.5.014409. (14) Egor ov , S. A.; E varesto v , R. A . Colossal Spin Spling in the Monolay er of the Collinear Anf erromagnet MnF₂. J. Ph ys. Chem. Le. 2021 , 12 (9), 2363–2369. hps://doi.or g /10.1021/ acs.jpcle.1c00282. (15) Liu, P .; Li, J.; Han, J.; W an, X.; Liu, Q. Spin-Group S ymmetry in Magnec Mat erials with Negligible Spin-Orbit Coupling. Phys. R ev . X 2022 , 12 (2), 021016. hps://doi.or g /10.1103/Ph ysRevX.12.021016. (16) Y ang, J.; Liu, Z. -X.; F an g, C. Symme try Invariant s and Classes of Quasiparcles in Magnecally Or dered Sy st ems Having W eak Spin- Orbit Coupling. Nat. Commun. 2024 , 15 (1), 10203. hps://doi.or g /10.1038/ s41467 -024-53862- 6. (17) Šmejkal, L.; Sinov a, J.; Jungwirth, T . Emerging R esear ch Landscape of Alter magn esm. Phys. R ev . X 2022 , 12 (4), 1–27. h ps:// doi.org/ 10.1103/Ph ysRevX.12.040501. (18) Šmejkal, L.; Sinov a, J.; Jungwirth, T . Beyond Con venonal F erromag n esm and Anf erromagnesm: A Phase with Nonr elavisc Spin and Crys tal Rot aon Symmetry . Phys. Rev . X 2022 , 12 (3), 1–16. h ps:// doi.org/ 10.1103/Ph ysRevX.12.031042. (19) Gonzále z -Hernández, R.; Šmejk al, L.; Výbor ný , K.; Y ah agi, Y .; Sinova, J .; Jungwirth, T .; Železn ý , J. E cient Electrical Spin Splier Based on Nonrela visc Collinear Anf erromagnesm. Phys. R ev . Le. 2021 , 126 (12), 1–6. hp s://doi.org /10.1103/Phy sRevLe .126.127701. (20) Kr emp aský , J.; Šmejkal, L.; D’Souz a, S. W .; Hajlaoui, M.; Springholz, G.; Uhlíř ová, K.; Alar ab, F .; Const annou, P . C.; Str ocov , V .; Usanov , D.; Pudelk o, W . R.; González-Hernández, R.; Birk Hellenes, A.; Jansa, Z.; Re ichlová, H.; Šobáň, Z.; Gonzalez Bet a n court, R. D .; W ad ley , P .; Sinova, J .; Kriegner , D.; Minár , J.; Dil, J . H.; Jungwirth, T . Altermagnec Liing of K ramer s Spin D egener acy . Nature 2024 , 626 (7999), 517–522. hps://doi.or g /10.1038/ s41586 - 023 -06907- 7. (21) Feng, Z.; Zhou, X.; Šmejk al, L.; W u, L.; Zhu, Z.; Guo, H.; Gonz ález -Hernández, R.; W an g, X.; Y an, H.; Qin, P .; Zhang, X.; W u, H.; Chen, H.; Meng , Z.; Liu, L.; Xia, Z.; Sinov a, J.; Jungwirth, T .; Liu, Z. An Anomalous Hall E ect in Altermagnec Ruthenium Dioxid e. Nat. Electron. 2022 , 5 (11), 735–743. hps://doi.or g /10.1038/ s41928 -022-00866- z. (22) Bai, H.; Han, L.; Feng, X. Y .; Zhou, Y . J.; Su, R. X.; W ang, Q.; Liao, L. Y .; Zhu, W . X.; Chen, X. Z.; Pan, F .; F an , X. L.; Song, C. Observao n of Spin Spling T orque in a Collinear Anf erromagnet RuO 2 . Phys. R ev . Le. 2022 , 128 (19), 1–6. hps://doi.or g /10.1103/Phy sRevLe .128.197202. (23) Naka, M.; Mot om e, Y .; Seo , H. Pero vskite as a Spin Current Genera tor . Phys. R ev . B 2021 , 103 (12), 1–6. hps://doi.or g /10.1103/Ph ysRevB .103.125114. (24) Chang, P . -H.; Mazin, I. I. The Myst erious Magnec Gr ound State of Ba 14 MnBi 11 Is Lik ely Self- Doped and Altermagnec . npj Spintronics 2024 , 2 (1), 52 . hp s://doi.org /10.1038/s44306- 024 - 00056- x. (25) Hodt, E. W .; Sukhachov , P .; Linder , J. Interf ace- Induced Magnez aon in Altermagnets and Anf erromagnets. Ph ys. Re v . B 2024 , 110 (5), 054446. hps://doi.or g /10.1103/Ph ysRevB .11 0.054446. (26) Banerjee, S.; Scheurer , M. S. Altermagnec Superc onducng Diode E ect. Phys. Re v . B 2024 , 110 (2), 024503. hps://doi.or g /10.1103/Phy sRevB.110.024503. (27) Sødequist, J.; Olse n, T . T wo-Dimen sional Alt ermagnets fr om High Thro u ghput Computa  onal Screening: Symme try Requirement s, Chir al Magnons, and Spin-Orbit E ects. Appl. Ph ys. Le. 2024 , 124 (18). hps://doi.org/ 10.1063/5.0198285. (28) Kazmin, D . Y u.; Esin, V . D .; Bar ash, Y u. S.; Timonina, A. V .; Kolesnik ov , N . N.; Devia tov , E. V . Andreev R eecon f or MnT e Altermagnet Candidat e. Physica B Condens. Maer 2025 , 696 , 416602. hps://doi.or g /10.1016/j.ph ysb.2024.416602. (29) Garc ia-Gaitan, F .; Ke fa ya, A.; Xiao, J . Q.; Nik olić, B. K. Magnon Spectrum of Alt ermagnets bey ond Linear Spin W av e Theory: Magnon-Magnon Intera con s via Time-Dependent Ma trix Product Sta tes v ersus A tomisc Spin Dynamics. Phys. Rev . B 2025 , 111 (2), L020407. hps://doi.or g /10.1103/Ph ysRevB .11 1.L020407. (30) Naka, M.; Mot om e, Y .; Seo , H. Altermagnec P erovskit es. npj Spintronics 2025 , 3 (1), 1. hps://doi.or g /10.1038/ s44306 -024-00066- 9. (31) W ei, C. - C.; Lawrence, E.; T ran, A.; Ji, H. Cry stal Chemist ry a nd Design Principles of Altermagnet s. ACS Organic & Inorganic Au 2024 , 4 (6), 604–619. hps://doi.or g /10.1021/ acsorginor gau.4c00064. (32) Herz og , A.; Marutzky , M.; Sichelschmidt, J.; Steglic h , F .; Kimura, S.; J ohnsen, S.; Iver sen, B. B. Str ong Electron Corr elaons in FeSb 2 : An Opcal In vesg aon and Comparison with RuSb 2 . Phys. Re v . B 2010 , 82 (24), 245205. hps://doi.or g /10.1 10 3/Phy sRevB.82.245205. (33) T omczak, J. M .; Haule, K.; Miyak e, T .; Georges, A.; K otliar , G. Th ermopower of Correlat ed Semiconductor s: Applicaon to F eAs 2 and FeSb 2 . Phys. R ev . B 2010 , 82 (8), 085104. hps://doi.or g /10.1103/Ph ysRevB .82 .085104. (34) Diakhate, M. S.; Herma nn, R. P .; Möchel, A.; Ser gueev , I.; Sønderg aard, M.; Christ ensen, M.; V erstr aete, M. J. Thermodynamic, Thermoelectric, and Magne  c Pro peres of FeS b 2 : A Combined Firs t -Principles and Experimental Study . Phys. R ev . B 2011 , 84 (12), 125210. hps://doi.or g /10.1103/Ph ysRevB .84 .125210. (35) Pokhar el, M.; Huaizhou Zhao; Modic, K. A.; Zhif eng Ren; Opeil, C. Magnec Properes of Hot- Pressed F eS b 2 . IEEE T rans. Magn. 2014 , 50 (5), 1–4. hps://doi.or g /10.1109/TMA G.2013 .2292607. (36) Pe trovic, C.; Kim, J. W .; Bud’k o, S. L.; Goldman, A. I.; Caneld, P . C.; Choe , W .; Miller , G. J. Anisotropy and Larg e Magnetor esistance in the Narr ow-Gap Semiconductor F eSb 2 . Phys. Rev . B 2003 , 67 (15), 155205. hp s:/ /doi.org /10.1103/Phy sRevB.67.155205. (37) Mazin, I. I.; Koeper nik, K.; Johannes, M. D .; González-Hernández, R.; Šmejk al, L. P redicon of Uncon venonal Magne sm in Doped F eSb 2 . Proc. Natl. Aca d . Sci. U. S. A. 2021 , 118 (42), 1–8. hps://doi.or g /10.1073/pnas.2108924118. (38) Holseth, H.; Kjekshus, A.; Harnung, S. E.; Lundstr öm, H.; Borch, G.; Cr aig, J. C. Compounds with the Marc asite T ype Crys tal Structure. IV . The Crys tal Structure of FeSb₂. Acta Ch em. Scand. 1969 , 23 , 3043–3050. hps://d oi.org /10.3891/act a.chem.scand.23 -3043. (39) T akizaw a, H.; Sh imada, M.; Sat o, Y .; Endo , T . High-Pressure S ynthesis of MnSb 2 with the Mar casite-T ype Structure. Mat er . Le. 1993 , 18 (1 –2), 11–13. hps://doi.or g /10.1016/0167- 577X(93)90047- 2. (40) Holseth, H.; Kjekshus, A.; Andr esen, A. F .; Schäf er , L.; Shimizu, A. Compounds with the Marc asite T ype Crys tal Structure. VI. Neutr on Diraco n Studies of CrSb₂ and FeSb₂. Acta Chem. Scan d . 1970 , 24 , 3309–3316. hp s://doi.org /10.3891/act a.chem.scand.24 -3309. (41) T oby , B. H.; V on Dreele, R. B. GSAS- II : The Genesis of a Modern Open-Source All Purpose Crys tallograph y So ware Pack age. J. Appl. Crysta llogr . 2013 , 46 (2), 544–549. hps://doi.or g /10.1107/S0021889813003531. (42) Caneld, P . C.; K ong, T .; Kaluar achchi, U . S.; Jo, N. H. Use of Frit-Disc Crucibles f or Roune and Explora tory Soluon Gr owth of Sin gle Cry st allin e Samples. Philosophical M agazine 2016 , 96 (1), 84 –92. hps://doi.or g /1 0.1080/14786435.2015.1122248. (43) Parkin, S.; Moezzi, B.; Hope, H. XABS 2: An Empiri cal Absorpon Correc on Progra m. J . Appl. Crysta llogr . 1995 , 28 (1), 53–56. hps://doi.or g /10.1107/S00218 89894009428. (44) W alk er , N.; Stuart, D. An Empirical Met h od for Co rrecng Diracto meter Dat a f or Absorpon E ects. Act a Cr ysta llogr . A 1983 , 39 (1), 158–166. hps://doi.or g /10.1107/S0108767 383000252. (45) Sheldrick, G. M. Crys tal Structur e Re n ement with SHELXL. Acta Cryst allogr . C Struct. Chem. 2015 , 71 (1), 3–8. hps://doi.org/10.1107/S2053229614024218. (46) Sheldrick, G. M. SHELXT – Integr ated Space-Gro up and Cryst al-Structure Det er minaon. Acta Crysta llogr . A Found. Adv . 2015 , 71 (1), 3–8. hps: // doi.org/ 10.1107/S2053273314026370. (47) T oby , B. H.; V on Dreele, R. B. GS AS-II : The Genesis of a Modern Open- Sourc e All Purpose Crys tallograph y So ware Pack age. J. Appl. Crysta llogr . 2013 , 46 (2), 544–549. hps://doi.or g /10.1107/S0021889813003531. (48) R odríguez-Carvajal, J. Rec ent Advances in M agnec Structur e Determinaon by Neutr on Pow d er Diracon. Ph ysica B Condens. Ma er 1993 , 192 (1 –2), 55 –69. hp s://doi.or g /10 .1016/0921- 4526(93)90108- I. (49) Wills, A. S. A New Prot ocol for the Determina on of Magnec Structures Using Simulat ed Annealing and Re present aonal Analysis (SARAh). Physic a B Condens. Maer 2000 , 276–278 , 680 –681. hp s://doi.org/10 .1016/S0921-4526(99)01722- 6. (50) Giannozzi, P .; Bar oni, S.; Bonini, N.; Calandr a, M.; Car , R.; Cavazz oni, C.; Ceresoli, D .; Chiaro, G. L.; Cococcioni, M.; Dabo , I.; Dal Corso , A.; de Gironc oli, S.; F ab ris, S.; Fra tesi, G.; Gebauer , R.; Gers tmann, U.; Go u goussis, C.; K okalj, A.; Lazzeri, M.; Marn-Samos, L.; Marzari, N .; Mauri, F .; Mazzar ello, R.; Paoli n i, S.; Pas q uarello , A.; Paula o, L.; Sbr accia, C.; Scandolo , S.; Sclauzer o, G.; Seitsonen, A. P .; Smogunov , A.; Umari, P .; W entzc ovitch, R. M. QU AN TUM ESPRESSO: A Modular and Open-Source Sowa re Project f or Quantum Simulaons of Ma terials. Journal of Physics: Condensed Maer 2009 , 21 (39), 395502. h ps://doi.org /10.1088/0953-8984/21/39/395502. (51) Garrity , K. F .; Benne, J . W .; Rabe, K. M.; V anderbilt, D. P seud opotena ls f or High -Throughput DFT Calculaons. Comput. Mat er . Sci. 2014 , 81 , 446–452. hps://doi.or g /10.1016/j.c ommatsci.2013.08.053. (52) Prandini, G.; Marrazz o, A.; Cas telli, I. E.; Mounet, N.; Marzari, N. Pr ecision and Eciency i n Solid - Sta te P seu dopotenal Ca lculaons. NP J Comput. Ma ter . 2018 , 4 (1), 72. hps://doi.or g /10.1038/ s41524 -018-0127- 2. (53) Per dew , J. P .; Burk e, K.; Ernz erhof , M. Generaliz ed Gr adient Appro ximaon Made Simple. Phys. Re v . Le. 1996 , 77 (18), 3 86 5–3868. hps://doi.or g /10.1103/Ph ysR evLe.77.3865. (54) Monkhors t, H. J.; Pack, J . D . Special P oints for Brillouin-Zone Int egrao n s. Phys. R ev . B 1976 , 13 (12), 5188–5192. hps://doi.or g /10.1103/Ph ysRevB .13 .5188. (55) Davidson, E. R. The It erave Calcula on of a Few of the Low est Eigen values and Corres p onding Eigen vector s of Large R eal-Symmetric Ma trices. J. Comput. Ph ys. 1975 , 17 (1), 87–94. hps://doi.or g /10.1016/0021-9991(75)90065- 0. (56) Sety awan, W .; Cu rtar olo, S. High-Throughput Elect ronic Band Structur e Calculaons: Challenges and T ools. Comput. Mater . Sci. 2010 , 49 (2), 299–312. hps://doi.or g /10.1016/j.c ommatsci.2010.05.010. (57) T ogo, A.; Shinohar a, K.; T anaka, I. Spglib: A Sowa re Library f or Cryst al Symmetry Sear ch. Scie nce and T ec hnology of Advanced Mat eri als: Methods 2024 , 4 (1). hps://doi.or g /10.1080/27660400.2024.2384822. Support Informa tion Pressure -Stabilized MnSb 2 with Co mplex I ncomm ensurat e Magnet ic Order Mingyu Xu 1,2 ,3 , Matt Boswell 1,4 , Qing-Ping Ding 2 , Peng Cheng 1 , Aashish Sapkota 2,3 , Qiang Zhang 4 , Danielle Yahne 4 , Sergey. L. Bud’ko 2, 3 , Yuji Furukawa 2,3 , Paul. C. Canfield 2,3 , Raquel A. Ribeiro 2,3 , Weiwei Xie 1* 1 Department of Chemistry , Michigan State University , East Lansing , Michigan 48824, USA 2 Ames National Laboratory , Iowa State University , Am es, Iowa 5001 1, USA 3 Department of Physics and As tronomy , Iowa State University , Ames, Iowa 5001 1, USA 4 Neutron Scattering Division, Oak Ridg e National Laboratory , Oak Ridge, T ennessee 37831, USA T able of Content s T abl e S1 The crystal structure and refinement of MnSb 2 ……………….….…...........…………..S2 T abl e S2 Atomic coordinates of MnSb 2 ……………….….….….….… .. .….…...........…………..S2 T abl e S3 Calculated the total ener gy of different magnetic models …… ... ………….….….….……..S2 Fig. S1 . Powder XRD results from the dif ferent batches…………………….…...........…………..S2 Fig. S2. M ( H ) data of single crystals and polycrystals at 300K………………...........…………….S3 Fig. S3. Heat capacity and entropy analysis and M ( T ) measurements ………............…………….S3 Fig. S4. PXRD results for the same batches……………………...…………..............…………….S5 Fig. S5. NMR spectrum of polycrystalline samples at 4.2K…...…….………............…………….S6 Fig. S6. Rietveld refinement of the 300 K powder neutron dif fraction data for MnSb 2 …………...S7 Fig. S7. The single crystal X-ray dif fraction sample picture and the synthesis design of MnSb 2 ... . S7 T abl e S4 Refined atomic coordinates of MnSb 2 using neutron dif fraction at 300 K…………….. .S8 T abl e S5 Basis vectors for the space group Pnnm at 300 K…….….….….…........ .. . ..……..……..S9 T abl e S7 Basis vectors for the space group P   at 200 K……………………….….….…………..S9 T abl e S1. The crystal st ructure a nd refin e ment of M nSb 2 at room temp erature. V alues in parentheses are estimated standard deviation fr om refinement. Chemical Formula MnSb 2 Formula W eight 298.44 g/mol Space Group Pnnm lUnit Cell dimensions a = 6.0227(3) Å b = 6.8893(3) Å c = 3.3240(2) Å V o lume 137.918(12) Å 3 Density (calculated) 7.186 g/cm 3 Absorption coef ficie nt 23.579 mm −1 F (000) 254 2θ range 9.00 to 80.26° Reflections collected 3540 Independent reflection s 478 [R int = 0.0561] Refinement method Full-matrix least-squares on F 2 Data/restraints/parameters 478/0/1 1 Final R indices R 1 (I>2 σ( I)) = 0.0282; wR 2 (I > 2 σ( I)) = 0.0795 R 1 (all) = 0.0289; wR 2 (all) = 0.0799 Larg est dif f. peak and hole +2.072 e - /Å 3 and -1.637 e - /Å 3 R. M. S. deviation from mean 0.428 e - /Å 3 Goodness- of -fit on F 2 1.229 T abl e S 2. Atomic coo r dinates and equivalent isotropic atomic displacement parame ters (Å 2 ) MnSb 2 . ( U eq is defined a s one-third of the trace of the orthogonalized U ij tensor .) V a lues in parentheses are estimated standard deviation s from refinement. Atoms W yck . x y z Occ . U eq Sb 4 g 0.17921(5) 0.36252(4) 0 1 0.01228(1 1) Mn 2 a 0 0 0 1 0.01 1 82(17) T able S3 . Calculated the total ener gy of di ffere nt magnetic models. NM FM AFM Magnetic Mode l T otal Ener gy 0 meV -69.68 meV -80.79 meV 20 40 60 80 0 1000 2000 3000 4000 Intensity 2  Batch 1 Batch 2 Batch 3 Batch 4 MnSb 2 Sb Figure S1 , Powder XRD results from the dif fer ent batches. T he sample for XRD is a part of the chunk directly from synthesis without th e separation of single crystals. Figure S1 shows the powder XRD results from the different b a tches. The sample for XRD is a part of the chunk directly from synthesis without th e separation of single crystals. To complete the reaction of Mn, a small excess of Sb (~1 wt% ) was added. Figure S2 , M(H) data of single crystals and polycrystals at 300 K. The mole ratio of Mn 1.1 Sb is estimated by assuming all the ferromagnetic singl e crystals come from Mn 1.1 Sb. Figure S2 s hows M ( H ) d ata of single crystals and polycrystals at 300 K. The mo le ra tio of Mn1.1Sb is estimated by assuming all the ferromagnetic sing le c rystals come from MnSb. Figure S3 , T empera ture-dependent s pecific heat was measured in th e temperature range from 150 K to 280 K and 2 K to 200 K . The measureme nt is taken on the N -grease with addenda taken with the same data points as the sa mple measurements in the temperature ra nge from 150 K to 280 K. There is no clear dif fe rence between the field at 0 Oe or 9 0 kOe. Magnetic entropy is calculated with reference to FeSb 2 . Figure S4. Powder XRD results from the same batch. The interval of the time is around six months. The highest intensity peak normalizes the intensity. T o assess the ambient stability of MnSb 2 over time, Figure S4 sho ws the n ormalize d powder X- ray diffraction (XRD) patterns of t he same sa mple batch measured immediately after synthesis (blac k) and after six months of air expo sure at room temperature (red). De spite MnSb 2 being a me tastable phase at ambient pressure, no discernible chan ges in the dif fra ction pattern are observed, indicating that the phase remains structurally stable under ambient cond itions for 6 months. This result sugge sts that MnSb 2 is stable at or below room temperature, with no evidence of decomposition or transformation over six month s. 0 2 4 6 8 50 100 150 200 250 300 (b) T = 4.2 K 121 Sb NMR signal s from Sb metal (im purity ) Spin-ech o Intensity (arb. un its)  0 H (T) f = 67 MHz 55 Mn from MnSb (i mpurity ) Spin e cho i ntens ity (a rb. uni t) f (M Hz) T = 4.2 K 123 Sb from MnSb (i mpurity ) (a) Figure S5 . (a) NMR spectrum me asured at T = 4.2 K by sweeping the ma gnetic field on polycrystalline powder samples. (b) Zero -field NMR spectrum at T = 4.2 K on polycrystalline po wder samples. W e attempted to get information about the magnetic ground state of MnSb 2 from 55 Mn, 121 Sb, and 123 Sb NMR measure ments at a low temperature of T = 4.2 K usi ng the same sample from n e utron dif fra ction measurements. Figure S 5 a shows the NMR spectrum, where a very broad spectrum from the hi ghest ma gnetic fie ld o f our ma gnet ( 8.25 T) to zero field wa s ob served. The relatively s harp lines detected from 5.5 T to 7.5 T (indicated by the red a rrows) were assigned to a typical quadru polar split 121 Sb NMR p ower lines from the impurity of Sb metal 1 . The intrinsic broad spectrum indicates that the hyperfine fields at the Mn a nd Sb sites are widely distributed in the ma gnetically ordered state, c onsistent with the inhomogeneous mag netic s tate as determined by the neutron diffraction measurements. Such an inhomogeneous ma gnetic sta te is also suggested b y the very br oad zero - field NMR spectrum shown in Figure S 5 b . The signals around 250-260 MHz and 200- 230 MHz a re from 55 Mn and 123 Sb, respectively , i n the im purity o f the fe rromagnetic Mn 1.1 Sb 2 . Although the broad spectrum is not fully resolved at present, we c onsider that the signals below 200 MHz are due to 55 Mn and 121 Sb N MR signals in antiferromagnetically or dered MnSb 2 . This again suggests the large distributions of hyperfine fields a t the Mn and Sb sites, consistent wit h the NMR spectrum, indicating the inhomogeneou s magnetic state. T o investigate the possibility that high -pressure synthesized MnSb 2 exhibits magnetic ordering above room t emperature (T n > 300 K), powder neutron diffraction measurements were conducted to detect either additional ma gnetic Bragg reflections or intensity enhancements of the nuclear Bragg peaks c onsistent with the P nmm crystal symme try . In general, antiferromagnetic (AFM) ordering gives rise to new magnetic reflections, while ferromagnetic ordering manifests as enhanced intensity at the positions of n uclear Bragg peaks. Figure S 6 a presents the neutron dif frac tion pattern collected at 300 K, al ong with the c orresponding Rietveld refinement. All observed peaks are consistent wit h the P nmm symmetry of MnSb 2 , and no a dditional ma gnetic peaks are d etected. The refined lattice parameters from the n eutron diffraction data are a = 6.0162(2) Å, b = 6.8824(3) Å, and c = 3.3226(1) Å, which a re in good agreement with the values obtain ed from th e X -ra y d iffractio n refinement. Minor impurity phases were identified as MnSb ( 2.7 wt %), MnO (0.3 wt %) , Sb ( 3.6 wt%), and boron nitride (BN, 0.07 wt%), which account for the weak reflec tions not indexed to the primary phase. Statistical parameters from the refinement, including all the phases in t he model, are R wp = 6. 0% and GOF= 3.2, indicating a reasonable refinement. The site occupancy , atomic coordinates, and iso tropic displacement parameters obtained fro m t he refinement are summarized in T abl e S4 . Figure S 6. ( a ) Rietveld refin ement of the 300 K powder neutron dif fraction data for MnSb 2 , yiel ding a weighted profile R-factor (R wp ) of 6.0% and a goodness of fit (GOF) of 3.2. The blue do ts repre sent the experimental data, while the orange line corresponds to the ca lculated pattern. The gree n line indicates the difference between the observed and calculated intensities. V ertical tick marks represent the allowed Bragg refl ection positions for the identified phases: MnSb 2 (red), Sb (gray), MnSb (black), MnO (cyan), and bo ron nitride (BN, magenta). ( b ) Enlarged view of the low-Q re gion from panel ( a ) , emphasizing the absence of additional magnetic Bragg peaks, indicating n o long -range magnetic ordering at room temperature. T abl e S4. Refined a tomic coordinates and t he equivalent isotropic displacement parameters of MnSb 2 at 300 K. Atoms W yck . x y z Occ . U eq Sb 4 g 0.1799(1) 0.3624(1) 0 1 0.0080(2) Mn 2 a 0 0 0 1 0.0060(3) T abl e S5. Basis vectors for the s pace group Pnnm with k = (0, 0, 0). The decomposition of the magnetic representation for the Mn site (0, 0, 0) is                   . Th e atoms of the nonprimitive basis are defined according to 1: (0, 0, 0), 2: (1/2, 1/2, 1/2). IR BV Atom BV co mponents m  a m  b m  c im  a im  b im  c Γ 1 ѱ 1 1 0 0 4 0 0 0 2 0 0 -4 0 0 0 Γ 3 ѱ 2 1 4 0 0 0 0 0 2 4 0 0 0 0 0 ѱ 3 1 0 4 0 0 0 0 2 0 -4 0 0 0 0 Γ 5 ѱ 4 1 4 0 0 0 0 0 2 -4 0 0 0 0 0 ѱ 5 1 0 4 0 0 0 0 2 0 4 0 0 0 0 Γ 7 ѱ 6 1 0 0 4 0 0 0 2 0 0 4 0 0 0 T abl e S6. Symmetry-allowed magnetic basis vectors obtained from representation analysis for the Mn1 and Mn2 sublattices, o btaine d from Rietvel d refinements o f the p owder neutron d iffraction data collected at 20 0 K. For each su blattice, the basis vectors ѱ 1, ѱ 2, an d ѱ 3 correspond to ma gnetic moment components along the crystallographic a , b , and c axes, r e specti vel y . Mn1 a b c ѱ 1 1 0 0 ѱ 2 0 1 0 ѱ 3 0 0 1 Mn2 ѱ 1 1 0 0 ѱ 2 0 1 0 ѱ 3 0 0 1 T o further p robe potential magnetic contributions in Mn Sb 2 , Figure S6 b highlights th e l ow- Q region of the neutron dif frac tion pattern, where magnetic scattering is typically enh a nced due to the Q -dependence of the magnetic form factor . The absence of additional Bragg reflections in this region pr ovides compelling evidence against the presence of long -ra nge ma gnetic ordering at r oom temperature. In particular , the lac k of ma gnetic peaks strongly su ggests that antiferromagnetic (AFM) ordering with a non-zero propagation vector ( k ≠ (0, 0, 0)) is a bsent in t his compound. Nonetheles s, the possibility of ferroma gnetic ordering corresponding to k = (0, 0, 0) cannot be excluded solely based on t he absence of additional reflections. In such a case , magnetic ordering would manifest as an enhancement of nuclear Bragg peak intensities. However , our Rie tveld refinement shows no observable i ntensity i ncre ase at the nuclear pea k positions, and no s ystematic deviation betwe en observed and calculated intensities wa s detected. T o definitively asse ss the presence of ferromagne tic ordering and to estimate an upper bo und for the ordered M n moment, magn etic structure refinements were performed using the FullProf suite with ma gnetic sub gr oups of the Pnmm space group. Symmetry-allowed magnetic configurations were derived from a group -su bgroup analysis of the propagation vector k = (0, 0 , 0), implemented via SARAh-Repr esentational Analysis. Four irreducible representations (IRs) and their corre sponding basis vectors (BVs) were identified, as listed in T able S5 . Among them, IRs Γ 3 (with BV ѱ₂) and Γ 5 (with BV ѱ₅), which describe ferromagnetic alignment of Mn moments within the ab -plane, produced the best magnetic refinements, with magnetic R -factors ( R mag ) below 5. Despite these fits, the refined ordered mome nt was consistently small, with a total magnetic moment o f μ tot that should be much smaller than 0.2 μ B /Mn, indicating the absence of sig nificant static magnetic o rdering in MnSb 2 at room temperature. Figure S 7. ( a ) The sample used to determ ine the crystallographic direction of MnSb 2 is shown. ( b ) Th e PXRD data are shown for solid-state reactions a t temperatures above 700 °C. ( c ) The sample current a s a function of temperature is shown for different MnSb 2 reactions. ( d ) The phase diag ram is shown. A point is the eutectic point, and the b point is the peritectic point. The red curve indicates the liquidus lines. Refer enc e (1) Hewitt, R. R.; W illiams, B. F . Nuclear Quadrupole Interaction of Sb121 and Sb123 in Antimony Metal. Phys Rev 1963, 129, 1 188. https://doi.org/10.1 103/PhysRev .129.1 188 (2) Nagarajan, V .; V ijayar aghavan. Nuclear M agnetic Resonance in Ferromagnetic MnSb and Cr x Mn 1-x Sb. J Phys Soc Jpn 1972, 33, 88-91. https://doi.org/10. 1 143/JPSJ.33.88