Structural and magnetic properties of Mg doped YbMnO3

Bhumireddi Sattibabu a, Anil K. Bhatnagar a,b,n, Sudhindra Rayaprol c, Dasari Mohan a,Dibakar Das a, Mahadevan Sundararaman a, Vasudeva Siruguri c

a School of Engineering Sciences and Technology, University of Hyderabad, Hyderabad 500046, Indiab School of Physics, University of Hyderabad, Hyderabad 500046, Indiac UGC-DAE CSR, Mumbai Centre, R-5 Shed, BARC, Mumbai 400085, India

a r t i c l e i n f o

Keywords:Hexagonal manganitesMagnetic orderingMg-dopingStructural studies

a b s t r a c t

We have studied the effect of Mg doping on structure and magnetism of multiferroic YbMnO3. Roomtemperature neutron diffraction studies were carried out on polycrystalline Yb1�xMgxMnO3 (x¼0.00and 0.05) samples to determine phase formation as well as cation distribution and structural propertiessuch as bond length and bond angles. The structural analysis shows that with Mg substitution, there isa marginal change in a and c parameters of the hexagonal unit cell, c/a ratio remains constant for x¼0and 0.05 samples. Due to changes in bond angle and bond lengths on substituting Mg, there is a slightdecrease in the distortion of MnO5 polyhedra. Magnetic measurements show that the Néel temperature(TN) increases marginally from 85 K for x¼0.00 to 89 K for x¼0.05 sample.

& 2014 Elsevier B.V. All rights reserved.

1. Introduction

Materials in which magnetic ordering can be controlled byelectric field and vice-versa are called multiferroic. Such com-pounds have tremendous potential in practical applications suchas spintronics. Rare-earth and yttrium manganites of generalformula RMnO3 crystallize in hexagonal structure for R withsmaller ionic radius (R¼Ho, Er, Tm, Yb, Lu, and Y) [1] or inorthorhombic structure for larger ionic radius (R¼La, Ce, Pr, Nd,Sm, Eu, Gd, Tb, and Dy) [2]. Hexagonal RMnO3 compounds areknown to show simultaneously ferroelectric as well as magneticordering in the ordered phase. The crystal structure of hexagonalmanganites consists of MnO5 polyhedra in which Mn3þ ion issurrounded by three oxygen ions in plane and two apical oxygenions. Mn ions with in Mn–O plane form a triangular lattice and arecoupled in terms of spins through the antiferromagnetic (AFM)super exchange interaction. Due to an incomplete AFM couplingbetween neighboring Mn ions in the triangular lattice, the systemforms a geometrically frustrated magnetic state [3–5].

Among rare earth manganites YbMnO3 has been scarcelystudied. Magnetization measurements of YbMnO3 showed anAFM transition around 82 K [6]. In the crystallographic structure

of YbMnO3, Yb occupies two crystallographic sites (in Wyckoffnotations), 2a and 4b of space group P63cm. There have been manystudies on the effect of substitution at Mn site of YbMnO3, whereaswe have concentrated on Yb site substitutions. The aim of thisstudy is to understand the effects of partial replacement of Yb3þ

by Mg2þ on the structural and magnetic properties of YbMnO3.We have synthesized Mg-doped YbMnO3 and focused on twocompounds YbMnO3 (x¼0.00) and Yb0.95Mg0.05MnO3 (x¼0.05) forcarrying out the comparative study.

2. Experimental

Polycrystalline samples of Yb1�xMgxMnO3 (x¼0.00 and 0.05)were prepared by a standard solid-state reaction method. Stoi-chiometric amounts of high-purity (purity greater than 99.9%)Yb2O3, MnCO3, and MgO powders were thoroughly mixed andsubsequently calcined in air at 1200 1C for 24 h, with an inter-mediate grinding for homogenization. The calcined mixture wascold pressed into pellets, sintered at 1450 1C in air for 24 h. Finally,all the samples were slowly cooled to room temperature forsufficient oxygenation.

The phase purity of samples was checked by powder X-raydiffraction (XRD) on a Brüker D8 Advance X-ray powder diffract-ometer using Cu Kα radiation. Neutron diffraction experimentswere carried out at room temperature at Dhruva, BARC on theUGC-DAE CSR beam line (TT1015) using a neutron beam ofwavelength 1.48 Å. All the magnetization measurements wereperformed on a vibrating sample magnetometer (PPMS-VSM) in

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/physb

Physica B

http://dx.doi.org/10.1016/j.physb.2014.03.0120921-4526/& 2014 Elsevier B.V. All rights reserved.

n Corresponding author at: School of Engineering Sciences and Technology &School of Physics, University of Hyderabad, Hyderabad 500046, India.Tel.: þ91 40 23134301/23013200; fax: þ91 40 23010227.

E-mail addresses: bsb.satti@gmail.com (B. Sattibabu), anilb42@gmail.com,akbsp@uohyd.ernet.in (A.K. Bhatnagar).

Please cite this article as: B. Sattibabu, et al., Physica B (2014), http://dx.doi.org/10.1016/j.physb.2014.03.012i

Physica B ∎ (∎∎∎∎) ∎∎∎–∎∎∎

the temperature range from 2 to 300 K in 100 Oe applied dcmagnetic field, in both zero field cooled (ZFC) and field cooled (FC)states of the samples.

3. Results and discussion

In Fig. 1 room temperature XRD patterns of Yb1�xMgxMnO3

(x¼0.00 and 0.05) samples are shown. The samples studied here aresingle phase and the measured patterns can be indexed according tothe hexagonal structure with P63cm space group (JCPDS No. 38-1246).In Fig. 2 the Rietveld refinement patterns of neutron powder diffrac-tion (NPD) data are shown for both x¼0.00 and 0.05 samples. Thegood agreement between observed (black scattered points) andcalculated profiles (red continuous line) is indicated by the differenceline (blue color). Refined values of lattice parameters and discrepancyfactors for Yb1�xMgxMnO3 (x¼0.00 and 0.05) are shown in Table 1.The values obtained for x¼0.00 sample are in good agreement withthose reported in the literature [7].

From the values given in Table 1, it is clearly seen that due toMg substitution, the cell parameter a decreases marginally, whilethere is a slight increase in c. Overall the cell volume decreases forx¼0.05 sample when compared to that of x¼0.00 sample. Thisdecrease in cell volume can be ascribed to ionic size effects, sincesmaller Mg2þ (Shannon ionic radii �0.890 Å) ion replaces theslightly bigger Yb3þ (Shannon ionic radii �0.925 Å) ion. Theobserved decrease in cell volume shows that Mg2þ does indeedreplace Yb3þ in the doped x¼0.05 sample. As the average A-site(i.e. Yb site) radius changes from x¼0.00 to 0.05, it is expected thatthe tolerance factor will also change. It is well known formanganites that the average A-site radius plays a crucial role inmaintaining the valence state(s) of Mn ions at B-site. Therefore, asthe divalent Mg is substituted at Yb3þ site, it will not only inducechemical pressure effect but also change the valence of Mn ionsfrom Mn3þ towards a mixed-valence state. Presence of Mn4þ ionin MnO5 polyhedra results in Jahn–Teller distortions and this alsocontributes to variations in lattice parameters; similar explanationwas given by Jeuvreya et al. for YMn1�xCuxO3 compounds [8].

Some selected bond distances of Yb–O and Mn–O for x¼0.00and 0.05 samples are listed in Table 2. The value of cell parametera decreases in Mg doped sample when compared to that inpristine sample. This change is ascribed to the decrease in theaverage ab-plane i.e. Mn–O3 and Mn–O4 bond lengths. Mn–O1and Mn–O2 bond lengths are along the c axis, and exhibit minorchange on doping. The average Mn–O distances in MnO5 units aresignificantly shorter in doped sample compared to those of pureYbMnO3.

It is expected that MnO5 polyhedra will play an important rolein the magnetic properties observed in both the samples. Mn–Obond lengths are in agreement with the sum of ionic radii [9].Moreover, in YbO7 polyhedron, Yb–O distances are larger inx¼0.00 sample than those in x¼0.05 sample, as expected due to

Fig. 1. Room temperature XRD pattern of Yb1�xMgxMnO3 (x¼0.00 and 0.05)samples indexed in space group P63cm.

Fig. 2. Rietveld refinement of room temperature neutron diffraction data ofYb1�xMgxMnO3 (x¼0.00 and 0.05) samples. The raw data is plotted along withthe calculated profile, the difference between observed and calculated data. Thevertical tick marks indicate the expected Bragg peak positions. (For interpretationof the references to color in this figure, the reader is referred to the web version ofthis article.)

Table 1Structural parameter after the Rietveld refinement of NPD for Yb1�xMgxMnO3

(x¼0.00 and 0.05) samples at room temperature.

Yb1�xMgxMnO3 x¼0.00 x¼0.05

a (Å) 6.070(5) 6.064(4)c (Å) 11.345(4) 11.352(4)V (Å3) 362.05(2) 361.87(2)χ 2 3.32 4.03RP 5.51 9.21Rwp 7.25 12.6RB 4.45 4.73t (Å) 0.831 0.833

‘t’ is the tolerance factor, given as t ¼ rYbþroð Þ=ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2 rMnþroð Þ

p� �.

B. Sattibabu et al. / Physica B ∎ (∎∎∎∎) ∎∎∎–∎∎∎2

Please cite this article as: B. Sattibabu, et al., Physica B (2014), http://dx.doi.org/10.1016/j.physb.2014.03.012i

the larger ionic radius of Yb3þ when compared to that of Mg2þ .Also, the average Yb–O distances are close to the sum of ionic radii.

The effect of Mg doping in YbMnO3 has been observed inmagnetization measurements also. Fig. 3 shows temperature depen-dence of dc magnetic susceptibility (χ¼M/H) of Yb1�xMgxMnO3

(x¼0.00 and 0.05) samples in zero field cooled (ZFC) and fieldcooled (FC) states of the sample, and in an applied field of 100 Oe.With decreasing temperature, χ increases gradually with T, as iffollowing the Curie–Weiss law; however, around 200 K there isa deviation from the Curie–Weiss behavior. At low temperatures,around 10 K, there is a sudden rise in χ, as if the samples areundergoing ferromagnetic ordering. The ferromagnetic orderinghas been attributed to the ordering in Yb3þ sub-lattice [7]. On further

decrease in temperature, χZFCðTÞ for both x¼0.00 and 0.05 samplesexhibits another peak around 3 K and bifurcation between ZFC and FCcurves. This bifurcation is believed to arise due to the competitionbetween ferromagnetic and anti-ferromagnetic sub-lattices in thepresence of field. It is difficult to figure out any long-range magneticorder in the plot of χ Tð Þ. The magnetization data of our polycrystal-line sample agrees well with the single-crystal data of YbMnO3

measured along the ab-plane [10]. Therefore, we use the firstderivative of χ Tð Þ curves (see Fig. 4) for both samples to find outthe temperature at which antiferromagnetic ordering sets in.As shown in the figure, anomaly is observed around 85 K inx¼0.00 sample, while in the case of x¼0.05, TN increases to89 K. The ordering in both compounds may be due to a cantedspin ordering of Mn3þ ions [11]. The increase in TN for x¼0.05sample can be explained on the basis of smaller cell volume, whichmay lead to strong exchange interactions and therefore higherordering temperatures [9].

The plot of inverse magnetic susceptibility is shown in Fig. 5.The inverse susceptibility data exhibits linear behavior above200 K, and can be fitted by using the Curie–Weiss lawχ ¼ C=T�θp. The paramagnetic Curie temperature (θp) and effec-tive magnetic moment (μeff) are calculated from the Curie–Weissfit. The values of θp obtained for x¼0.00 and 0.05 samples are�219 K and �182 K, respectively. The value of θp for x¼0.00sample is in good agreement with the reported values [7]. Thenegative θp implies predominant antiferromagnetic interactions.The frustration factor f and |θp|/TN were calculated for bothcompounds. The values of f for x¼0.00 and 0.05 sample are 2.58and 2.04, respectively. f decides the frustration of Mn spins intriangular lattice and it is slightly decreased for x¼0.05 sample.The effective moment meff is 5.91mB (mB is Bohr magneton) forYbMnO3 which is close to the value 6.1mB reported by Fabregeset al. [7]. In the case of x¼0.05 sample, the observed value of μeff is5.66mB for Yb0.95Mg0.05MnO3 obtained from the equationmeff¼(7.99C)0.5, where C is Curie–Weiss constant.

Table 2Selected bond distances and bond angles for Yb1�xMgxMnO3 (x¼0.00 and 0.05).

Parameter x¼0.00 x¼0.05

Mn–O1 1.858 1.889Mn–O2 1.882 1.874Mn–O3 1.964 1.981Mn–O4 2.075 2.067oMn–O4 1.945 1.953Yb1–O1 2.272 2.243Yb1–O2 2.303 2.219Yb1–O3 2.370 2.411Yb2–O1 2.247 2.279Yb2–O2 2.258 2.264Yb2–O4 2.371 2.369oYb–O4 2.322 2.319Mn–O3–Mn (1) 119.34(2) 119.89(3)Mn–O4–Mn (1) 118.25(2) 117.66(3)Δ(10�4) 19.08 15.77

Δ is the distortion of MnO5 polyhedra.

Fig. 3. Temperature dependence of dc susceptibility of samples (a) YbMnO3 and(b) Yb0.95Mg0.05MnO3 with ZFC/FC mode. The inset shows low temperaturebehaviors of samples with ZFC and FC modes.

Fig. 4. First derivative of the magnetic susceptibility (χ, emu/mol) measured in zerofield cooled state of the sample (ZFC) is plotted as a function of temperature forYb1�xMgxMnO3 (x¼0.00 and 0.05) samples.

B. Sattibabu et al. / Physica B ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

Please cite this article as: B. Sattibabu, et al., Physica B (2014), http://dx.doi.org/10.1016/j.physb.2014.03.012i

4. Conclusions

We have studied structural and magnetic properties of singlephase hexagonal YbMnO3 and Yb0.95Mg0.05MnO3 samples prepared bythe standard solid state reaction method. We find subtle effect ofMg doping on the structure (bond length, bond angles, unit cellvolume, tolerance factor and distortion of MnO5 polyhedra) andmagnetism (change in TN and reduction in frustration parameter f)of YbMnO3. With partial Mg doping at Yb site, the A-site average

radius is affected, which results in equivalent amount of conversionof Mn3þ to Mn4þ , thus affecting the magnetic ordering (tempera-ture) arising due to Mn sub-lattice. However, as Yb has two crystal-lographic positions in YbMnO3, Mg at 5% doping level is too less toinfluence the low temperature magnetic ordering arising due to Yb.The magnetic behavior of polycrystalline YbMnO3 and dopedYbMnO3 can be explained on the basis of similar observations madein single crystal study [10]. Detailed neutron diffraction studies onpristine and samples with varying Mg concentrations are plannedand will be helpful in understanding the role of Mg in modifying thestructural and physical properties of YbMnO3.

Acknowledgment

This work has been supported by UGC-DAE Consortium forScientific Research, Mumbai Centre, India in the form of acollaborative research scheme (CRS) project. BSB acknowledgesUGC-DAE CSR, Mumbai Centre for financial support in the form ofproject fellowship. AKB is thankful to the National Academy ofSciences, India for their support through Senior Scientist PlatinumJubilee Fellowship scheme.

References

[1] M.A. Gilleo, Acta Crystallogr. 10 (1957) 161.[2] H. Yakel, W.C. Koehler, E.F. Bertaut, F. Forrat, Acta Crystallogr. 16 (1963) 957.[3] A. Munoz, J.A. Alonso, M.J. Martinez-Lope, M.T. Casais, J.L. Martinez,

M.T. Fernandez-Diaz, Phys. Rev. B 62 (2000) 9498.[4] K. Katsufuji, M. Masaki, A. Machida, M. Moritomo, K. Kato, E. Nishibori,

M. Takata, M. Sakata, K. Ohoyama, K. Kitazawa, H. Takagi, Phys. Rev. B 66(2002) 134434.

[5] M.C. Sekhar, S. Lee, G. Choi, C. Lee, J.G. Park, Phys. Rev. B 72 (2005) 014402.[6] T. Katsufuji, S. Mori, M. Masaki, Y. Moritomo, N. Yamamoto, H. Takagi, Phys.

Rev. B 64 (2001) 104419.[7] X. Fabreges, I. Mirebeau, P. Bonville, S. Petit, G. Lebras-Jasmin, A. Forget,

G. Andre, S. Pailhes, Phys. Rev. B 78 (2008) 214422.[8] L. Jeuvreya, O. Pen, A. Moure, C. Moure, J. Magn. Magn. Mater. 324 (2012) 717.[9] A. Munoz, J.A. Alonso, M.J. Martınez-Lope, M.T. Casais, J.L. Martinez,

M.T. Fernández-Diaz, Phys. Rev. B 62 (2000) 950.[10] Josep Fontcuberta, Marin Gospodinov, Vassil Skumryev, J. Appl. Phys. 103

(2008) 07B722.[11] Yun-Hui Huang, Marrit Karppinen, Naoki Imamura, Hisao Yamauchi, John

B. Goodenough, Phys. Rev. B 76 (2007) 17440.

Fig. 5. Inverse magnetic susceptibility plots of Yb1�xMgxMnO3 (x¼0.00 and 0.05)samples. The straight line passing through the data points is the fit to the Curie–Weiss law. The values of paramagnetic Curie temperature (θp) and effective Bohrmagneton (μeff) number for each sample are shown in their respective panels.

B. Sattibabu et al. / Physica B ∎ (∎∎∎∎) ∎∎∎–∎∎∎4

Please cite this article as: B. Sattibabu, et al., Physica B (2014), http://dx.doi.org/10.1016/j.physb.2014.03.012i