Print version ISSN 1516-1439
Mat. Res. vol.15 no.6 São Carlos Nov./Dec. 2012 Epub Oct 11, 2012
Microstructure and metal-insulator transition in single crystalline KMo4O6
Margareth Andrade; Mariana Lanzoni Maffei; Leandro Marcos Salgado Alves; Carlos Alberto Moreira dos Santos*; Bento Ferreira; Antonio Fernando Sartori
Departamento de Engenharia de Materiais, Escola de Engenharia de Lorena, Universidade de São Paulo - USP, CEP 12602-810, Lorena, SP, Brazil
High quality KMo4O6 single crystals with tetragonal structure (space group P4/mbm) have been prepared by fused salt electrolysis. The crystals were studied by scanning electron microscopy (SEM), X-ray diffractometry, electrical resistivity, and magnetization measurements. X-ray powder diffraction patterns and SEM have given some information on the growth of single crystals. Electrical resistivity as a function of temperature shows that the KMo4O6 compound is a bad metal with resistivity change of approximately 30% in the temperature range from 2 to 300K. A metal-insulator transition (MIT), observed at approximately 110K, has been also confirmed for this material. Magnetization as a function of temperature agrees with previous report, however a magnetic ordering has been observed in M(H) curves in the whole temperature range.
Keywords: KMo4O6, single crystal, low-dimensional system
During the last years great attention has been given to the low-dimensional (D) conductors1-3. There are several reasons for that. Three of them are: i) low-D conductors are good models for comparison with theory of electrical conductivity such as Pierls transition4 and Luttinger liquid (LL) theory5, ii) they allow the study of several anisotropic properties, and iii) they are closed related to the existence of superconductivity.
In particular, molybdenum oxides form one of the largest families with high anisotropic behavior. One good example is the Li0.9Mo6O17 purple bronze compound which is known to be the best example for the 1D LL theory3,6. Furthermore, it shows a crossover from metal to insulator at low temperatures and becomes an anisotropic superconductor below 1.9 K7. Recently, our group has found anisotropic properties in a new phase with stoichiometry KxMoO2-δ. This compound has an anomalous metallic behavior at low temperatures which is related to a weak magnetic ordering. Also, some samples display superconducting behavior with surprisingly critical temperature (TC) as high as 10 K8.
Additionally, molybdates have attracted great attention due to the existence of many valences of the Mo ions (from 2+ to 6+)9. Recently, the study of the Mo valences in the Li0.9Mo6O17 compound unambiguously showed an interesting connection with the formation of 1D channels in this compound10.
Due to these characteristics of the molybdates, our group has studied the preparation and characterization of many anisotropic molybdenum oxides. In particular, this paper deals with the KMo4O6 phase which is reported to be a quasi-low dimensional system11-15. In fact, KMo4O6 can crystallize in two different phases, one is orthorhombic with space group Pbam and another is tetragonal with space group P or P4/mbm12,13. The most reported phase is the tetragonal P4/mbm which has high anisotropic electrical conductivity and shows a metal-insulator transition (MIT) near 110 K11. Ramanujachary and co-workers have argued this compound is paramagnetic and no magnetic ordering is related to the MIT11. The anisotropic low-D behaviors of the KMo4O6 compound have been attributed to the metal-metal bonding along c-axis of the crystalline structure13,15.
This paper reports some information on the microstructure of single crystals and additional physical properties of the tetragonal KMo4O6 phase. It is reported for the first time details about the mechanism of the crystal growth during the fused salt electrolysis. Furthermore, magnetization measurements show a magnetic ordering in the whole temperature range from 2 to 300 K.
Single crystals of the tetragonal KMo4O6 phase have been prepared by fused salt electrolysis using K2MoO4 and MoO3 in 6 to 1 molar ratio, respectively. The synthesis was performed during 52 hours at 930 °C applying an electrical current between 20 to 25 mA. Black single crystals with rectangular shape grew attached to the platinum cathode. To have a better idea about the cathode in Figure 1 is shown a typical platinum cathode before and after the electrolysis.
Some of the crystals were removed from the cathode, ground, and characterized by powder X-ray diffractometry. The cathode and some crystals were analyzed by scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS). In order to measure the electrical and magnetic properties some crystals were cleaned using diluted HCl solution. Electrical resistivity measurements were performed in a Physical Properties Measurement System (PPMS) from 2 to 300 K using the standard four-probe method along c-axis of the single crystal. Magnetization measurements were carried out in Vibrating Sample Magnetometer (VSM) coupled with 9T PPMS. The measurements were performed attaching several single crystals on the sample holder and applying magnetic field parallel and perpendicular to the longest axis of the crystals.
Results and Discussion
X-ray powder diffraction pattern of the ground single crystals removed from the cathode is shown in Figure 2. The peaks agree very well with those of the simulated for the P4/mbm phase12. Peaks marked with * symbols are due to the MoO2 phase despite the overlaps with the (220), (211) and (440) peaks of the KMo4O6 phase. The comparison between powder difractograms shows that the crystals are mainly due to the KMo4O6 tetragonal P4/mbm phase with some MoO2 impurities. Results of SEM performed in some as grown-single crystals agree with that. Micrograph of the deposited crystals, displayed in Figure 3a, shows that the single crystals grow perpendicularly to the spiral shaped platinum cathode. In Figure 3b, one can see that the crystals are rectangular bars 1 to 2 mm long with cross section up to 0.2 × 0.2 mm2.
One as-grown single crystal was selected and observed with higher magnification. Figure 3c displays the micrograph of a small part of the crystal. It is possible to observe several small impregnated crystals on the KMo4O6 bar. These small crystals have been attributed to the MoO2 impurities in agreement with the peaks of this phase observed in X-ray diffractogram shown in Figure 2. Probably, the small crystals on top of the KMo4O6 single crystal bars have grown during the cooling process in which MoO2 and MoO3 can exist in equilibrium at high temperatures16.
Figure 4 displays the electrical resistivity as a function of temperature for the longest axis of one single crystal (c-axis).
The electrical resistivity varies less than 30% in the range from 2 to 300 K. This is an indication that the KMo4O6 compound is a poor conducting metal. Furthermore, a MIT occurs near 118 K. The general behavior of the electrical resistivity is in good agreement with previous report11. The origin of this MIT is still under discussion. Ramanujachary et al.11 pointed out that this transition is not associated with any magnetic ordering. In order to clarify that we have performed magnetization measurements in the crystals with magnetic field applied parallel and perpendicular to the longest axes of the crystals. In Figure 5 are displayed magnetization measurements as a function of temperature, M(T). Main panel shows the M(T) for magnetic field of 2 T applied parallel to the longest axes of the single crystals.
At high temperature (above ~100K), the magnetization is almost constant and at low temperatures it shows an upturn. Although the M(T) curve looks like a conventional paramagnetic behavior, 1/M versus T plot (not shown) does not display any linear behavior in whole temperature range measured. The constant magnetization at higher temperatures can be more properly observed in inset of the Figure 5 for magnetic field of 1T applied parallel (black symbols) and perpendicular (red symbols) to c-axis of the single crystals. A magnetic anisotropy (Mparallel/Mperpendicular) with a factor of 2 has been found near room temperature for H = 1 T. The general behavior seems to agree with previous measurements11, but the magnetization versus applied magnetic field hysteresis curves, M(H), measured at several fixed temperatures suggest a different interpretation. Figure 6 displays many M(H) curves from 3 to 300 K. Above 80 K all the curves almost set on top of each other, show weak magnetic ordering at low magnetic fields, and linear behavior at magnetic fields above 2000 Oe. The M(H) curves for temperatures below 80 K show non-linear behaviors and are strongly temperature dependent. This indicates a different magnetic behavior above and below 80 K. Furthermore, a weak ferromagnetic ordering is observed at all temperatures (see Figure 6b). Remanent magnetization and coercive field are weak temperature dependent. On the other hand, the hysteresis curves at low field follow the same trend for the curves measured at high magnetic fields above and below 80 K. The origin of the magnetic ordering in the KMo4O6 is still under investigation but results performed in other molybdate compounds8,17 suggest that the presence of Mo ions with different valences could be responsible for the weak ferromagnetic moment observed in Figure 6.
The following conclusions have been obtained:
KMo4O6 single crystals have been successfully prepared by salt fused electrolysis;
Electrical resistivity measurement shows a similar behavior as reported previously and confirms the crossover from metal to insulating behavior near 118 K;
Magnetization measurements as a function of temperature agree with the previous results;
M(H) curves at high magnetic fields (H > 2000 Oe) are temperature dependent and constant below and above 80 K, respectively;
M(H) hysteresis curves unambiguously demonstrate the presence of a ferromagnetic ordering at all temperatures from 3 to 300 K. The origin of this magnetic ordering is still under investigation.
This material is based upon work supported by FAPESP (2009/14524-6, 2009/54001-2, and 2010/06637-2) and CNPq (490182/2009-7 and 309084/2010-5). M. Andrade is a CAPES fellowship.
1. Tarkiainen R, Ahlskog M, Penttila J, Roschier L, Hakonen P, Paalanen M et al.\Multiwalled carbon nanotube: Luttinger versus Fermi liquid. Physical Review B. 2001; 64:195412. http://dx.doi.org/10.1103/PhysRevB.64.195412 [ Links ]
2. Artemenko SN and Remizov SV. Low-temperature conductivity of quasi-one-dimensional conductors: Luttinger liquid stabilized by impurities. Physical Review B. 2005; 72:125118. http://dx.doi.org/10.1103/PhysRevB.72.125118 [ Links ]
3. Dos Santos CAM, White BD, Yu Y-K, Neumeier JJ and Souza JA. Dimensional Crossover in the Purple Bronze Li0.9Mo6O17. Physical Review Letters 2007; 98:266405. http://dx.doi.org/10.1103/PhysRevLett.98.266405 [ Links ]
6. Da Luz MS, Dos Santos CAM, Moreno J, White BD and Neumeier JJ. Anisotropic electrical resistivity of quasi-one-dimensional Li0.9Mo6O17 determined by the Montgomery method. Physical Review B. 2007; 76:233105. http://dx.doi.org/10.1103/PhysRevB.76.233105 [ Links ]
7. Dos Santos CAM, Da Luz MS, Yu Y-K, Neumeier JJ, Moreno J. and White BD. Electrical transport in single-crystalline Li0.9Mo6O17: A two-band Luttinger liquid exhibiting Bose metal behavior. Physical Review B. 2008; 77:193106. http://dx.doi.org/10.1103/PhysRevB.77.193106 [ Links ]
8. Alves LMS, Damasceno VI, Dos Santos CAM, Bortolozo AD, Suzuki PA, Izario Filho HJ et al.Unconventional metallic behavior and superconductivity in the K-Mo-O system. Physical Review B. 2010; 81:174532. http://dx.doi.org/10.1103/PhysRevB.81.174532 [ Links ]
9. Moshier WC, Davis GD and Cote GO. Surface Chemistry of Sputter-Deposited Al-Mo and Al-Cr Alloys Polarized in 0.1N KCl Electrochemical Society. 1989; 136(2):356-62. http://dx.doi.org/10.1149/1.2096635 [ Links ]
10. Da Luz MS, Neumeier JJ, Dos Santos CAM, White BD, Izario Filho HJ, Leão JB et al. Neutron diffraction study of quasi-one-dimensional lithium purple bronze: Possible mechanism for dimensional crossover. Physical Review B. 2011; 84:014108. http://dx.doi.org/10.1103/PhysRevB.84.014108 [ Links ]
11. Ramanujachary KV, Greenblatt M, Jones EB and McCarrol WH. Synthesis and Characterization of a New Modification of the Quasi-Low-Dimensional Compound KMo4O6. Journal of Solid State Chemistry. 1993; 102(1):69-78. http://dx.doi.org/10.1006/jssc.1993.1008 [ Links ]
12. McCarrol WH, Ramanujachary KV, Greenblatt M and Marsch RE. On the Tetragonal Forms of KMo4O6. Journal of Solid State Chemistry. 1995; 117(1):217-18. http://dx.doi.org/10.1006/jssc.1995.1266 [ Links ]
13. Hughbanks T and Hoffmann R. Chains of trans-edge-sharing molybdenum octahedra: metal-metal bonding in extended systems. Journal of the American Chemical Society. 1983; 105(11):3528-37. http://dx.doi.org/10.1021/ja00349a027 [ Links ]
14. Jung D, Lee B-H, Kim S-J and Kang W. Synthesis, Characterization, and Electronic Structure of a New Molybdenum Bronze SnMo4O6. Chemistry of Materials. 2001; 13(5):1625-29. http://dx.doi.org/10.1021/cm000646q [ Links ]
15. Torardi CC and McCarley RE. Sodium tetramolybdenum hexoxide (NaMo4O6). A metallic infinite-chain polymer derived by condensation of octahedral clusters. Journal of the American Chemical Society. 1979; 101(14):3963-4. http://dx.doi.org/10.1021/ja00508a043 [ Links ]
16. Wold A, Kunnmann W, Arnott RJ and Ferretti A. Preparation and Properties of Sodium and Potassium Molybdenum Bronze Crystals. Inorganic Chemistry. 1964; 3(4):545-7. http://dx.doi.org/10.1021/ic50014a022 [ Links ]
17. Thakur P, Cezar JC, Brookes NB, Choudhary RJ, Prakash R, Phase DM et al. Direct observation of oxygen induced room temperature ferromagnetism in MoO2 thin films by x-ray magnetic circular dichroism characterizations. Applied Physics Letters. 2009; 94(6):06250. http://dx.doi.org/10.1063/1.3080679 [ Links ]
Received: March 1, 2012
Revised: July 17, 2012