- Email: [email protected]
Local magnetic properties of (RE)12Co5Bi studied by μSR
ARTICLE IN PRESS Physica B 404 (2009) 611–614
Contents lists available at ScienceDirect
Physica B journal homepage: www.elsevier.com/locate/physb
Local magnetic properties of ðREÞ12 Co5 Bi studied by mSR M. Egilmez a, I. Fan a, K.H. Chow a,, W.A. MacFarlane b, A.I. Mansour a, D.S. Martin b, J. Jung a, B. Hitti c, D. Arseneau c, A.V. Tkachuk d, A. Mar d a
Department of Physics, University of Alberta, Edmonton, AB, Canada T6G 2G7 Department of Chemistry, University of British Columbia, Vancouver, BC, Canada V6T 1Z3 TRIUMF, 4004 Wesbrook Mall, Vancouver, BC, Canada V6T 2A3 d Department of Chemistry, University of Alberta, Edmonton, Canada T6G 2G2 b c
a r t i c l e in fo
Keywords: mSR Rare-earth intermetallic compounds Local magnetism
abstract The recently discovered ðREÞ12 Co5 Bi (where RE is a rare-earth element) system has a rich magnetic phase diagram. In this paper, we report studies on polycrystalline samples of RE ¼ Tb, Gd and Tm in order to obtain local magnetic information on the samples. In low fields, the Tb and Gd compounds show a magnetic transition at 75 and 100 K, respectively, while there is no evidence of a magnetic transition down to 3 K in the Tm sample. These results are compared with bulk magnetization measurements. The magnetic ground state in Tm12 Co5 Bi appears to be dramatically different in low compared to high fields, indicative of a meta-magnetic transition. & 2008 Elsevier B.V. All rights reserved.
1. Introduction Over the decades there has been enormous interest on rareearth oxides [1–5] and intermetallic compounds [6,7] due to their rich phase diagrams. In particular, coupling of the f and d electrons in rare-earth transition metal intermetallics give rise to many instances of complex magnetic phase diagrams and intriguing unresolved magnetic and transport phenomena. One such example is the newly discovered series of ðREÞ12 Co5 Bi where RE is a rare-earth element. In this system, Bi occupies certain sites leading to a highly ordered structure with negligible quenched disorder due to its size [6]. Bulk magnetic studies suggest a variety of magnetic phases including ferrimagnetism and antiferromagnetism [6]. As usual, it is useful to complement the bulk magnetization studies with a local probe of magnetism such as mSR. MacFarlane et al. [8] have carried out such studies on the Ho12 Co5 Bi compound. These measurements revealed a bulk transition at T c ¼ 30 K to a magnetically ordered state, and it was shown that in the paramagnetic region, the sample follows a Curie–Weiss law with slow magnetic fluctuations which persist far above T c . Keeping in mind the fact that magnetic properties in ðREÞ12 Co5 Bi could be dramatically modified by substitution of different RE atoms, it is also be interesting to study other members of the series. In this paper, we report studies of the local magnetism in polycrystalline samples of selected ðREÞ12 Co5 Bi compounds for
Corresponding author. Tel.: +1780 492 4573; fax: +1780 492 0714.
E-mail address: [email protected] (K.H. Chow). 0921-4526/$ - see front matter & 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.physb.2008.11.112
RE ¼ Tb, Gd and Tm. Our results indicate magnetic transitions in the RE ¼ Tb and Gd samples at around 75 and 100 K, respectively. On the other hand, the RE ¼ Tm sample does not show any magnetic ordering down to 3 K in low magnetic fields. However, the effects of high magnetic field on the mSR spectra strongly suggest the existence of a meta-magnetic transition in Tm12 Co5 Bi where the magnetic ground state in low applied longitudinal magnetic fields is dramatically different from the one at high fields.
2. Experimental details The samples of Tb12 Co5 Bi, Gd12 Co5 Bi and Tm12 Co5 Bi were prepared by arc-melting of cold-pressed pellets prepared from stoichiometric amounts of elemental components with an excess of 2.5 wt% Bismuth added to compensate for vaporization during the reaction. The reaction was carried out in an Edmund Buhler MAM-1 compact arc furnace on a water-cooled copper plate in a Ti-gettered argon atmosphere. Afterward, the resulting ingots were annealed in sealed evacuated silica tubes at 870 K for at least one week and then quenched in cold water. All the samples were characterized by powder X-ray diffraction on an Inel powder diffractometer equipped with a CPS 120 detector and found to be single phase. Further details of sample preparation and structural characterization can be found in elsewhere [6]. The magnetization measurements were carried out using a Quantum Design PPMS system. The mSR measurements were carried out on the M20B beamline at TRIUMF in Vancouver, Canada, where mSR and
ARTICLE IN PRESS 612
M. Egilmez et al. / Physica B 404 (2009) 611–614
bNMR experiments can be carried out [10–12]. In this experiment, a beam of nearly 100% spin-polarized positive muons (lifetime 2:2 ms) of nominal momentum 28 MeV=c is implanted into the sample. The sample was mounted in a packet made from highpurity Ag foil (thickness 25 mm). The packet was attached to a high-purity silver backing with Apiezon N grease in a helium gas flow cryostat. The experiments described in this paper were carried out in the standard longitudinal field ðLF-mSRÞ configuration [9,13,14]. A small longitudinal field (LF) of 1 mT was applied to ‘‘decouple’’ relaxation effects due to nuclear moments (ARE, 59 Co, and 209Bi, where A is the atomic number) and any small stray magnetic fields, and hence we are only sensitive to the stronger
5
2.0
3 1.0 2 0.5
M/H (emu/f.u.)
H/M (f.u./emu)
4
Tb12Co5Bi
1.5
1
electronic moments [12]. A signal that relaxes or oscillates faster than the deadtime of the spectrometer ð10 nsÞ is detected only as a missing fraction of the total asymmetry [8].
3. Results and discussion Fig. 1 summarizes the temperature dependence of the bulk DC susceptibility w and inverse susceptibility of the three samples, measured in a field of 0.5 T. These results are discussed in more detail in Ref. [6] and we only briefly summarize them here. There is a peak in the wðTÞ of Tb12 Co5 Bi at approximately 75 K this is believed to be a consequence of an antiferromagnetic transition. Above this temperature, the Tb12 Co5 Bi sample is in the paramagnetic state. In the case of Gd12 Co5 Bi, the susceptibility rises rapidly below 120 K. By also taking into account the magnetic field dependence of the DC and AC susceptibility data, it was suggested that the sharp rise is a result of ferrimagnetic ordering in this sample [6]. The Tm12 Co5 Bi appears to be in a paramagnetic state down to 3 K (see Fig. 1), but there may be an antiferromagnetic transition at TN 2:7 K. Let us now discuss our mSR results on these samples. In all three compounds, no coherent precession signals were observed at any temperature, similar to the situation in Ho12 Co5 Bi reported previously by MacFarlane et al. [8]. In Fig. 2, we show representative LF-mSR spectra in Tb12 Co5 Bi. The data are well-described by a single component exponential relaxation together with a non-relaxing component (most likely due to
0
0.0 2.5
25
Cd12Co5Bi
2.0
1.5 15 1.0
10
0.5
5
0.0
0
M/H (emu/f.u.)
H/M (f.u./emu)
20
4 12
9 2
6
M/H (emu/f.u.)
H/M (f.u./emu)
Tm12Co5Bi
3
0 0
50
100 150 200 Temperature (K)
250
0 300
Fig. 1. Temperature dependence of the field cooled DC susceptibility and inverse susceptibility of all samples measured in a field of 0.5 T.
Fig. 2. Typical raw data of the time dependence of LF relaxation in a field of 10 G for Re ¼ Tb sample. The solid curves are best fits to Eq. (1).
ARTICLE IN PRESS M. Egilmez et al. / Physica B 404 (2009) 611–614
muons that stop in the surrounding Ag): AðtÞ ¼ As elt þ AAg
(1)
By fitting the LF ¼ 10 G data to Eq. (1) we were able to extract the spin relaxation rate l and the amplitude As . Their temperature dependences are shown in Fig. 3 for all three compounds. Note that, as temperature is decreased, l shows a peak for
Fig. 3. (a)–(c) Temperature dependence of relaxation rate l in a longitudinal field of 10 G for RE ¼ Tb, Gd and Tm. Insets show the corresponding relaxing asymmetry for each sample. The solid curves are guide to the eye.
613
Tb12 Co5 Bi and Gd12 Co5 Bi, clear evidence for a magnetic transition is taking place. As usual, the high temperature side corresponds to a paramagnetic state, while below the peak, there is static magnetic order. Consistent with this, As decreases dramatically at the transition temperature T c 75 and 100 K for Tb and Gd (vertical line in Fig. 3(a) and (b)), confirming that the majority of the sample has entered a magnetically ordered state. Note that these transition temperatures are roughly consistent with the bulk magnetization measurements [6]. On the other hand, there is no evidence in the Tm12 Co5 Bi sample for a magnetic transition within the temperature range of 3–300 K. According to the DC susceptibility measurements for the RE ¼ Tm sample (in a field of 0.5 T), antiferromagnetic ordering takes place at temperatures lower than 3 K, outside our accessible temperature range. The relaxation rates at high temperatures, i.e. in the paramagnetic state, for the RE ¼ Gd sample are substantially lower than for the RE ¼ Tb and Tm samples. Similar behavior has been observed in rare-earth Gd and Tb quasicrystals [15]. We speculate that the same qualitative mechanism proposed there might also explain our results: in the fast fluctuation regime, the observed muon spin relaxation rate is determined by the ratio of the second moment of the field distribution at the muon site, hB2m i, and the field fluctuation rate. If the site is assumed to be the same in the three compounds, then hB2m i is determined by the magnitude of
Fig. 4. (a)–(b) Temperature dependence of relaxation rate l in longitudinal fields of 10 G and 5 T for RE ¼ Gd and Tm. The solid curves are guide to the eye.
ARTICLE IN PRESS 614
M. Egilmez et al. / Physica B 404 (2009) 611–614 3þ
the surrounding moments. The free-ion moments for Tb , 3þ Tm3þ and Gd are comparable, so the large differences in relaxation rates may indicate that the Tb and Tm moment fluctuation rates are much less than in Gd. These differences in the fluctuation rates may be a consequence of crystalline electric field (CEF) splittings originating from the low rare-earth site symmetry. Note that for Gd12 Co5 Bi, the trivalent Gd is in a state of zero total angular momentum [14]. In this case the ion is spherically symmetric so it is not affected by the CEF [14–16]. However, CEF effects would be important in Tb12 Co5 Bi and Tm12 Co5 Bi. Measurements such as inelastic neutron scattering, heat capacity, etc. may be useful for elucidating the nature of the crystal field splittings, but are not currently available for these materials. We also performed LF-mSR measurements in a magnetic field of 5 T. In Fig. 4, we compare the data at 10 G with the 5 T data for Gd12 Co5 Bi and Tm12 Co5 Bi. In Gd12 Co5 Bi, the magnetic transition temperature at 10 G and 5 T are the same ð100 KÞ. Furthermore, the paramagnetic relaxation rate in high fields is smaller than at low fields, the standard well-known behavior associated with probing the spectral density of magnetic fluctuations at a constant temperature [9,14]. In the Gd12 Co5 Bi sample, there appears to be relaxation still present (in the 1/3rd tail) at low temperatures in both 10 G and 5 T. This indicates that significant dynamical fluctuations are still present. By contrast, in Tm12 Co5 Bi, the relaxation rates at a particular temperature increase with increasing field (see Fig. 4b). This is clear indication that the magnetic states in low fields and high fields are different. Bulk magnetization studies on this sample also led Tkachuk et al. [6] to propose the existence of a meta-magnetic transition above a field of 2 T at 2 K, believed to originate from the collapse/melting of the antiferromagnetic ground state with applied field. It will be fruitful to investigate the low temperature behavior of the RE ¼ Tm sample using mSR. There are several questions regarding the observed unusual behavior: Will the magnetic ordering temperature change with increasing magnetic field? What is the threshold field for the proposed meta-magnetic transition?
Acknowledgements This research is partially supported by the Center for Materials and Molecular Research at TRIUMF and the NSERC. TRIUMF receives federal funding via a contribution through the NSERC.
References [1] J. Jung, et al., Appl. Phys. Lett. 87 (2005) 262506; A.I. Mansour, et al., Appl. Phys. Lett. 90 (2007) 162511; M. Abdelhadi, et al., Appl. Phys. Lett. 88 (2006) 102501. [2] M. Egilmez, et al., Appl. Phys. Lett. 92 (2008) 162515; M. Egilmez, et al., Appl. Phys. Lett. 90 (2007) 162508; M. Egilmez, et al., Appl. Phys. Lett. 89 (2006) 062511. [3] M. Egilmez, et al., Appl. Phys. Lett. 90 (2007) 232506; R. Patterson, et al., Appl. Phys. Lett. 88 (2006) 172509. [4] M. Egilmez, et al., Appl. Phys. Lett. 92 (2008) 022501. [5] M. Egilmez, et al., Appl. Phys. Lett. 92 (2008) 132505. [6] A.V. Tkachuk, et al., Inorg. Chem. 44 (2005) 2272. [7] V. Babyuk, et al., J. Alloys Compd. 365 (2004) 21. [8] W.A. MacFarlane, et al., J. Phys. Chem. C 111 (2007) 5526. [9] S.J. Blundell, Contemp. Phys. 40 (1999) 175; P. Dalmas De Reotier, et al., J. Phys. Condens. Matter 9 (1997) 9113; A. Amoto, Rev. Modern Phys. 69 (1997) 1119. [10] K.H. Chow, et al., Phys. Rev. B 53 (1996) R14725; Th. Jestadt, et al., Phys. Rev. B 59 (1999) 3775; K.H. Chow, B. Hitti, R.F. Kiefl, Identification of defects in semiconductors, in: M. Stavola (Ed.), Semiconductors and Semimetals, vol. 51A, Academic Press, New York, 1998, p. 137; J. Sugiyama, et al., Phys. Rev. B 73 (2006) 224437; I. Fan, et al., Phys. Rev. B 77 (2008) 035233; K.H. Chow, et al., Phys. Rev. B 73 (2006) 113202; K.H. Chow, et al., Phys. Rev. B 51 (1995) 14762. [11] K.H. Chow, et al., Phys. Rev. Lett. 84 (2000) 2251; B.E. Schultz, et al., Phys. Rev. B 72 (2005) 033201; B.E. Schultz, et al., Phys. Rev. Lett. 95 (2005) 086404; K.H. Chow, et al., Phys. Rev. B 50 (1994) 8918; Z. Salman, et al., Nano Lett. 7 (2007) 1551; Z. Salman, et al., Phys. Rev. Lett. 98 (2007) 167001; G.D. Morris, et al., Phys. Rev. Lett. 93 (2004) 157601; Z. Salman, et al., Phys. Rev. B 70 (2004) 104404; Z. Salman, et al., Phys. Rev. Lett. 96 (2006) 147601; K.H. Chow, et al., Physica B 340 (2003) 1151. [12] A.I. Mansour, et al., Phys. Rev. Lett. 100 (2008) 257602; R.C. Vilao, et al., Phys. Rev. B 77 (2008) 235212; T.J. Parolin, et al., Phys. Rev. B 77 (2008) 214107; Z. Salman, et al., Phys. Rev. B 77 (2008) 214415; T.A. Keeler, et al., Phys. Rev. B 77 (2008) 144429; K. Mukai, et al., Phys. Rev. Lett. 99 (2007) 087601; R.L. Lichti, et al., Phys. Rev. B 76 (2007) 045221; J. Sugiyama, et al., Phys. Rev. B 75 (2007) 174424; Z. Salman, et al., Phys. Rev. B 75 (2007) 073405; K. Shimomura, et al., Phys. Rev. Lett. 92 (2004) 135505; R.L. Lichti, et al., Phys. Rev. B 70 (2004) 165204; K.H. Chow, et al., Phys. Rev. Lett. 76 (1996) 3790; K.H. Chow, et al., Phys. Rev. Lett. 87 (2001) 216403; S.F.J. Cox, et al., Phys. Rev. Lett. 86 (2001) 2601; Z. Salman, et al., Phys. Rev. B 73 (2006) 174427. [13] R.S. Hayano, et al., Phys. Rev. B 20 (1979) 850. [14] G.M. Kalvius, D.R. Noakes, O. Hartmann, in: K.A. Gschneidner Jr., L. Eyring, G.H. Lander (Eds.), Handbook on the Physics and Chemistry of the Rare Earths, vol. 32, North-Holland, Amsterdam, The Netherlands, 2001, pp. 55–451. [15] D.R. Noakes, et al., Phys. Lett. A 238 (1998) 197. [16] B.D. Dunlap, et al., Phys. Rev. B 29 (1984) 6243.
Contents lists available at ScienceDirect
Physica B journal homepage: www.elsevier.com/locate/physb
Local magnetic properties of ðREÞ12 Co5 Bi studied by mSR M. Egilmez a, I. Fan a, K.H. Chow a,, W.A. MacFarlane b, A.I. Mansour a, D.S. Martin b, J. Jung a, B. Hitti c, D. Arseneau c, A.V. Tkachuk d, A. Mar d a
Department of Physics, University of Alberta, Edmonton, AB, Canada T6G 2G7 Department of Chemistry, University of British Columbia, Vancouver, BC, Canada V6T 1Z3 TRIUMF, 4004 Wesbrook Mall, Vancouver, BC, Canada V6T 2A3 d Department of Chemistry, University of Alberta, Edmonton, Canada T6G 2G2 b c
a r t i c l e in fo
Keywords: mSR Rare-earth intermetallic compounds Local magnetism
abstract The recently discovered ðREÞ12 Co5 Bi (where RE is a rare-earth element) system has a rich magnetic phase diagram. In this paper, we report studies on polycrystalline samples of RE ¼ Tb, Gd and Tm in order to obtain local magnetic information on the samples. In low fields, the Tb and Gd compounds show a magnetic transition at 75 and 100 K, respectively, while there is no evidence of a magnetic transition down to 3 K in the Tm sample. These results are compared with bulk magnetization measurements. The magnetic ground state in Tm12 Co5 Bi appears to be dramatically different in low compared to high fields, indicative of a meta-magnetic transition. & 2008 Elsevier B.V. All rights reserved.
1. Introduction Over the decades there has been enormous interest on rareearth oxides [1–5] and intermetallic compounds [6,7] due to their rich phase diagrams. In particular, coupling of the f and d electrons in rare-earth transition metal intermetallics give rise to many instances of complex magnetic phase diagrams and intriguing unresolved magnetic and transport phenomena. One such example is the newly discovered series of ðREÞ12 Co5 Bi where RE is a rare-earth element. In this system, Bi occupies certain sites leading to a highly ordered structure with negligible quenched disorder due to its size [6]. Bulk magnetic studies suggest a variety of magnetic phases including ferrimagnetism and antiferromagnetism [6]. As usual, it is useful to complement the bulk magnetization studies with a local probe of magnetism such as mSR. MacFarlane et al. [8] have carried out such studies on the Ho12 Co5 Bi compound. These measurements revealed a bulk transition at T c ¼ 30 K to a magnetically ordered state, and it was shown that in the paramagnetic region, the sample follows a Curie–Weiss law with slow magnetic fluctuations which persist far above T c . Keeping in mind the fact that magnetic properties in ðREÞ12 Co5 Bi could be dramatically modified by substitution of different RE atoms, it is also be interesting to study other members of the series. In this paper, we report studies of the local magnetism in polycrystalline samples of selected ðREÞ12 Co5 Bi compounds for
Corresponding author. Tel.: +1780 492 4573; fax: +1780 492 0714.
E-mail address: [email protected] (K.H. Chow). 0921-4526/$ - see front matter & 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.physb.2008.11.112
RE ¼ Tb, Gd and Tm. Our results indicate magnetic transitions in the RE ¼ Tb and Gd samples at around 75 and 100 K, respectively. On the other hand, the RE ¼ Tm sample does not show any magnetic ordering down to 3 K in low magnetic fields. However, the effects of high magnetic field on the mSR spectra strongly suggest the existence of a meta-magnetic transition in Tm12 Co5 Bi where the magnetic ground state in low applied longitudinal magnetic fields is dramatically different from the one at high fields.
2. Experimental details The samples of Tb12 Co5 Bi, Gd12 Co5 Bi and Tm12 Co5 Bi were prepared by arc-melting of cold-pressed pellets prepared from stoichiometric amounts of elemental components with an excess of 2.5 wt% Bismuth added to compensate for vaporization during the reaction. The reaction was carried out in an Edmund Buhler MAM-1 compact arc furnace on a water-cooled copper plate in a Ti-gettered argon atmosphere. Afterward, the resulting ingots were annealed in sealed evacuated silica tubes at 870 K for at least one week and then quenched in cold water. All the samples were characterized by powder X-ray diffraction on an Inel powder diffractometer equipped with a CPS 120 detector and found to be single phase. Further details of sample preparation and structural characterization can be found in elsewhere [6]. The magnetization measurements were carried out using a Quantum Design PPMS system. The mSR measurements were carried out on the M20B beamline at TRIUMF in Vancouver, Canada, where mSR and
ARTICLE IN PRESS 612
M. Egilmez et al. / Physica B 404 (2009) 611–614
bNMR experiments can be carried out [10–12]. In this experiment, a beam of nearly 100% spin-polarized positive muons (lifetime 2:2 ms) of nominal momentum 28 MeV=c is implanted into the sample. The sample was mounted in a packet made from highpurity Ag foil (thickness 25 mm). The packet was attached to a high-purity silver backing with Apiezon N grease in a helium gas flow cryostat. The experiments described in this paper were carried out in the standard longitudinal field ðLF-mSRÞ configuration [9,13,14]. A small longitudinal field (LF) of 1 mT was applied to ‘‘decouple’’ relaxation effects due to nuclear moments (ARE, 59 Co, and 209Bi, where A is the atomic number) and any small stray magnetic fields, and hence we are only sensitive to the stronger
5
2.0
3 1.0 2 0.5
M/H (emu/f.u.)
H/M (f.u./emu)
4
Tb12Co5Bi
1.5
1
electronic moments [12]. A signal that relaxes or oscillates faster than the deadtime of the spectrometer ð10 nsÞ is detected only as a missing fraction of the total asymmetry [8].
3. Results and discussion Fig. 1 summarizes the temperature dependence of the bulk DC susceptibility w and inverse susceptibility of the three samples, measured in a field of 0.5 T. These results are discussed in more detail in Ref. [6] and we only briefly summarize them here. There is a peak in the wðTÞ of Tb12 Co5 Bi at approximately 75 K this is believed to be a consequence of an antiferromagnetic transition. Above this temperature, the Tb12 Co5 Bi sample is in the paramagnetic state. In the case of Gd12 Co5 Bi, the susceptibility rises rapidly below 120 K. By also taking into account the magnetic field dependence of the DC and AC susceptibility data, it was suggested that the sharp rise is a result of ferrimagnetic ordering in this sample [6]. The Tm12 Co5 Bi appears to be in a paramagnetic state down to 3 K (see Fig. 1), but there may be an antiferromagnetic transition at TN 2:7 K. Let us now discuss our mSR results on these samples. In all three compounds, no coherent precession signals were observed at any temperature, similar to the situation in Ho12 Co5 Bi reported previously by MacFarlane et al. [8]. In Fig. 2, we show representative LF-mSR spectra in Tb12 Co5 Bi. The data are well-described by a single component exponential relaxation together with a non-relaxing component (most likely due to
0
0.0 2.5
25
Cd12Co5Bi
2.0
1.5 15 1.0
10
0.5
5
0.0
0
M/H (emu/f.u.)
H/M (f.u./emu)
20
4 12
9 2
6
M/H (emu/f.u.)
H/M (f.u./emu)
Tm12Co5Bi
3
0 0
50
100 150 200 Temperature (K)
250
0 300
Fig. 1. Temperature dependence of the field cooled DC susceptibility and inverse susceptibility of all samples measured in a field of 0.5 T.
Fig. 2. Typical raw data of the time dependence of LF relaxation in a field of 10 G for Re ¼ Tb sample. The solid curves are best fits to Eq. (1).
ARTICLE IN PRESS M. Egilmez et al. / Physica B 404 (2009) 611–614
muons that stop in the surrounding Ag): AðtÞ ¼ As elt þ AAg
(1)
By fitting the LF ¼ 10 G data to Eq. (1) we were able to extract the spin relaxation rate l and the amplitude As . Their temperature dependences are shown in Fig. 3 for all three compounds. Note that, as temperature is decreased, l shows a peak for
Fig. 3. (a)–(c) Temperature dependence of relaxation rate l in a longitudinal field of 10 G for RE ¼ Tb, Gd and Tm. Insets show the corresponding relaxing asymmetry for each sample. The solid curves are guide to the eye.
613
Tb12 Co5 Bi and Gd12 Co5 Bi, clear evidence for a magnetic transition is taking place. As usual, the high temperature side corresponds to a paramagnetic state, while below the peak, there is static magnetic order. Consistent with this, As decreases dramatically at the transition temperature T c 75 and 100 K for Tb and Gd (vertical line in Fig. 3(a) and (b)), confirming that the majority of the sample has entered a magnetically ordered state. Note that these transition temperatures are roughly consistent with the bulk magnetization measurements [6]. On the other hand, there is no evidence in the Tm12 Co5 Bi sample for a magnetic transition within the temperature range of 3–300 K. According to the DC susceptibility measurements for the RE ¼ Tm sample (in a field of 0.5 T), antiferromagnetic ordering takes place at temperatures lower than 3 K, outside our accessible temperature range. The relaxation rates at high temperatures, i.e. in the paramagnetic state, for the RE ¼ Gd sample are substantially lower than for the RE ¼ Tb and Tm samples. Similar behavior has been observed in rare-earth Gd and Tb quasicrystals [15]. We speculate that the same qualitative mechanism proposed there might also explain our results: in the fast fluctuation regime, the observed muon spin relaxation rate is determined by the ratio of the second moment of the field distribution at the muon site, hB2m i, and the field fluctuation rate. If the site is assumed to be the same in the three compounds, then hB2m i is determined by the magnitude of
Fig. 4. (a)–(b) Temperature dependence of relaxation rate l in longitudinal fields of 10 G and 5 T for RE ¼ Gd and Tm. The solid curves are guide to the eye.
ARTICLE IN PRESS 614
M. Egilmez et al. / Physica B 404 (2009) 611–614 3þ
the surrounding moments. The free-ion moments for Tb , 3þ Tm3þ and Gd are comparable, so the large differences in relaxation rates may indicate that the Tb and Tm moment fluctuation rates are much less than in Gd. These differences in the fluctuation rates may be a consequence of crystalline electric field (CEF) splittings originating from the low rare-earth site symmetry. Note that for Gd12 Co5 Bi, the trivalent Gd is in a state of zero total angular momentum [14]. In this case the ion is spherically symmetric so it is not affected by the CEF [14–16]. However, CEF effects would be important in Tb12 Co5 Bi and Tm12 Co5 Bi. Measurements such as inelastic neutron scattering, heat capacity, etc. may be useful for elucidating the nature of the crystal field splittings, but are not currently available for these materials. We also performed LF-mSR measurements in a magnetic field of 5 T. In Fig. 4, we compare the data at 10 G with the 5 T data for Gd12 Co5 Bi and Tm12 Co5 Bi. In Gd12 Co5 Bi, the magnetic transition temperature at 10 G and 5 T are the same ð100 KÞ. Furthermore, the paramagnetic relaxation rate in high fields is smaller than at low fields, the standard well-known behavior associated with probing the spectral density of magnetic fluctuations at a constant temperature [9,14]. In the Gd12 Co5 Bi sample, there appears to be relaxation still present (in the 1/3rd tail) at low temperatures in both 10 G and 5 T. This indicates that significant dynamical fluctuations are still present. By contrast, in Tm12 Co5 Bi, the relaxation rates at a particular temperature increase with increasing field (see Fig. 4b). This is clear indication that the magnetic states in low fields and high fields are different. Bulk magnetization studies on this sample also led Tkachuk et al. [6] to propose the existence of a meta-magnetic transition above a field of 2 T at 2 K, believed to originate from the collapse/melting of the antiferromagnetic ground state with applied field. It will be fruitful to investigate the low temperature behavior of the RE ¼ Tm sample using mSR. There are several questions regarding the observed unusual behavior: Will the magnetic ordering temperature change with increasing magnetic field? What is the threshold field for the proposed meta-magnetic transition?
Acknowledgements This research is partially supported by the Center for Materials and Molecular Research at TRIUMF and the NSERC. TRIUMF receives federal funding via a contribution through the NSERC.
References [1] J. Jung, et al., Appl. Phys. Lett. 87 (2005) 262506; A.I. Mansour, et al., Appl. Phys. Lett. 90 (2007) 162511; M. Abdelhadi, et al., Appl. Phys. Lett. 88 (2006) 102501. [2] M. Egilmez, et al., Appl. Phys. Lett. 92 (2008) 162515; M. Egilmez, et al., Appl. Phys. Lett. 90 (2007) 162508; M. Egilmez, et al., Appl. Phys. Lett. 89 (2006) 062511. [3] M. Egilmez, et al., Appl. Phys. Lett. 90 (2007) 232506; R. Patterson, et al., Appl. Phys. Lett. 88 (2006) 172509. [4] M. Egilmez, et al., Appl. Phys. Lett. 92 (2008) 022501. [5] M. Egilmez, et al., Appl. Phys. Lett. 92 (2008) 132505. [6] A.V. Tkachuk, et al., Inorg. Chem. 44 (2005) 2272. [7] V. Babyuk, et al., J. Alloys Compd. 365 (2004) 21. [8] W.A. MacFarlane, et al., J. Phys. Chem. C 111 (2007) 5526. [9] S.J. Blundell, Contemp. Phys. 40 (1999) 175; P. Dalmas De Reotier, et al., J. Phys. Condens. Matter 9 (1997) 9113; A. Amoto, Rev. Modern Phys. 69 (1997) 1119. [10] K.H. Chow, et al., Phys. Rev. B 53 (1996) R14725; Th. Jestadt, et al., Phys. Rev. B 59 (1999) 3775; K.H. Chow, B. Hitti, R.F. Kiefl, Identification of defects in semiconductors, in: M. Stavola (Ed.), Semiconductors and Semimetals, vol. 51A, Academic Press, New York, 1998, p. 137; J. Sugiyama, et al., Phys. Rev. B 73 (2006) 224437; I. Fan, et al., Phys. Rev. B 77 (2008) 035233; K.H. Chow, et al., Phys. Rev. B 73 (2006) 113202; K.H. Chow, et al., Phys. Rev. B 51 (1995) 14762. [11] K.H. Chow, et al., Phys. Rev. Lett. 84 (2000) 2251; B.E. Schultz, et al., Phys. Rev. B 72 (2005) 033201; B.E. Schultz, et al., Phys. Rev. Lett. 95 (2005) 086404; K.H. Chow, et al., Phys. Rev. B 50 (1994) 8918; Z. Salman, et al., Nano Lett. 7 (2007) 1551; Z. Salman, et al., Phys. Rev. Lett. 98 (2007) 167001; G.D. Morris, et al., Phys. Rev. Lett. 93 (2004) 157601; Z. Salman, et al., Phys. Rev. B 70 (2004) 104404; Z. Salman, et al., Phys. Rev. Lett. 96 (2006) 147601; K.H. Chow, et al., Physica B 340 (2003) 1151. [12] A.I. Mansour, et al., Phys. Rev. Lett. 100 (2008) 257602; R.C. Vilao, et al., Phys. Rev. B 77 (2008) 235212; T.J. Parolin, et al., Phys. Rev. B 77 (2008) 214107; Z. Salman, et al., Phys. Rev. B 77 (2008) 214415; T.A. Keeler, et al., Phys. Rev. B 77 (2008) 144429; K. Mukai, et al., Phys. Rev. Lett. 99 (2007) 087601; R.L. Lichti, et al., Phys. Rev. B 76 (2007) 045221; J. Sugiyama, et al., Phys. Rev. B 75 (2007) 174424; Z. Salman, et al., Phys. Rev. B 75 (2007) 073405; K. Shimomura, et al., Phys. Rev. Lett. 92 (2004) 135505; R.L. Lichti, et al., Phys. Rev. B 70 (2004) 165204; K.H. Chow, et al., Phys. Rev. Lett. 76 (1996) 3790; K.H. Chow, et al., Phys. Rev. Lett. 87 (2001) 216403; S.F.J. Cox, et al., Phys. Rev. Lett. 86 (2001) 2601; Z. Salman, et al., Phys. Rev. B 73 (2006) 174427. [13] R.S. Hayano, et al., Phys. Rev. B 20 (1979) 850. [14] G.M. Kalvius, D.R. Noakes, O. Hartmann, in: K.A. Gschneidner Jr., L. Eyring, G.H. Lander (Eds.), Handbook on the Physics and Chemistry of the Rare Earths, vol. 32, North-Holland, Amsterdam, The Netherlands, 2001, pp. 55–451. [15] D.R. Noakes, et al., Phys. Lett. A 238 (1998) 197. [16] B.D. Dunlap, et al., Phys. Rev. B 29 (1984) 6243.