Doping effect of Pr6O11 on superconductivity and flux pinning of MgB2 bulk

Article abstract of DOI:10.1002/pssa.200622505

Bulk samples of MgB₂ doped with 0, 1, 3, and 5 wt% Pr ₆O₁₁ nanopowder were prepared using a solid-state reaction route. The lattice constants of Pr₆O₁₁-doped MgB ₂ systematically increase with increasing doping level, indicating a partial substitution of Pr for Mg in the crystal structure, and consequently, the superconducting transition temperature, T_c, of MgB₂ is slightly suppressed. At a low doping level of 1 wt% Pr₆O ₁₁, the critical current density, J_c, and the irreversibility field, H_irr are improved. However, at higher doping levels, Pr₆O₁₁ doping seems harmful for the performance of MgB₂ in high magnetic fields. It is argued that the effect of the doping mechanism of Pr₆O₁₁ on T₁₁ and J _c is quite different from those of other rare-earth elements, such as Y or Dy.

Full text of DOI:10.1002/pssa.200622505

Original
Paper
phys. stat. sol. (a) 204, No. 5, 1555–1560 (2007) / DOI 10.1002/pssa.200622505
Doping effect of Pr O on superconductivity and flux pinning
6 11
of MgB bulk
2
1
1
1
X. F. Pan , T. M. Shen , G. Li , C. H. Cheng , and Y. Zhao
2
*, 1, 2
1
Key Laboratory of Advanced Technologies of Materials (Ministry of Education of China),
Superconductivity R&D Center (SRDC), Mail Stop 165#, Southwest Jiaotong University,
610031 Chengdu, Sichuan, P.R. China
2
School of Materials Science and Engineering, University of New South Wales, 2052 Sydney,
NSW, Australia
Received 22 October 2006, revised 21 January 2007, accepted 24 January 2007
Published online 13 March 2007
PACS 61.10.Nz, 61.72.Ww, 74.25.Ha, 74.25.Qt, 74.70.Ad
Bulk samples of MgB doped with 0, 1, 3, and 5 wt% Pr O nanopowder were prepared using a solid-state
11
2
6
reaction route. The lattice constants of Pr O -doped MgB systematically increase with increasing doping
6
11
2
level, indicating a partial substitution of Pr for Mg in the crystal structure, and consequently, the super-
conducting transition temperature, T , of MgB is slightly suppressed. At a low doping level of
c
2
1 wt% Pr O , the critical current density, J , and the irreversibility field, H are improved. However, at
irr
6
11
c
higher doping levels, Pr O doping seems harmful for the performance of MgB in high magnetic fields.
2
6
11
It is argued that the effect of the doping mechanism of Pr O on T and J is quite different from those of
c
6
11
c
other rare-earth elements, such as Y or Dy.
© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1 Introduction
The discovery of the MgB superconductivity has generated great interest in the field of applied super-
2
conductivity because MgB has a critical transition temperature T of 39 K [1], reaching the record T of
2 c c
noncuprate superconductors. Compared with the high-temperature superconductors (HTS), MgB has
2
simple structure, low processing cost, and does not have the weak-link problem at grain boundaries that
hampers the applications of HTS-based superconductivity technologies. The advantages of MgB make it
2
a promising candidate for engineering applications in the temperature range 20–30 K, in which the tradi-
tional low-temperature superconductors (LTS) cannot play any roles.
One of the major problems of MgB -based superconductivity technologies is that the critical current
2
density J of MgB materials is still not high enough to satisfy the industrial applications, especially
c 2
under high magnetic field. A rapid drop of J with the increase of applied magnetic field indicates a lack
c
of effective pinning centers, such as second-phase particles, crystal defects, dislocations, and so on, in
MgB . Thus, it is necessary to improve the pinning behavior of MgB superconductor by all possible
2 2
means. Chemical doping often provides an effective way to meet this requirement, for example, the flux-
pinning behavior and J have been improved by chemical dopants, such as Zr, Ti, nanodiamond, C, SiC,
c
Y O [2–7]. Furthermore, magnetic-elements such as Mn, Fe, Co, Ni [8–10] have also been doped into
2 3
MgB , but these magnetic elements often suppress the superconductivity and degrade its performance in
2
*
Corresponding author: e-mail: yzhao@swjtu.edu.cn, Phone: +86 28 8760 0786, Fax: +86 28 8760 1824
© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1556
X. F. Pan et al.: Doping effect of Pr O on superconductivity and flux pinning of MgB
6 11 2
a magnetic field due to the existence of a local magnetic field. Rare-earth elements (RE) often possess a
strong magnetic moment, however, it seems that they do not suppress the superconductivity of MgB
2
[6, 11]. Recently, it is reported that Dy O -doped MgB shows a significantly enhanced J in a low or
2 3 2 c
medium field (2–4 T) [11].
The doping effects of rare-earth elements in MgB have not been so intensively studied. More studies
2
are necessary to explore the possible new approaches of improving the performance of MgB by rare-
2
earth element doping. For this purpose, the doping effect of Pr O on the superconductivity and flux
6 11
pinning of MgB is studied in this work. The reasons that Pr O is chosen as the dopant in this work
2 6 11
include: (1) Pr often exhibits a complex valence state, different from Y O , thus Pr doping may result in
2 3
some unexpected results; (2) Pr often exhibits a strong magnetic moment, thus it is helpful to investigate
the effect of magnetic ions of a rare-earth element on T and other superconducting properties of MgB ;
c
2
(3) Pr has a strong suppression on T of HTS, thus it is interesting to note if such a suppression exists in
c
the MgB system.
2
2 Experimental
A series of Pr O -doped MgB samples with mass ratios of MgB :Pr O = (1 – x):x (x = 0, 0.01,
6 11 2 2 6
11
0.03, and 0.05) were prepared by solid-state reaction at ambient pressure. Mg (99%), B (amorphous,
99.9%), Pr O (99.9%) nanopowder (100 nm) were mixed and pressed into cylinders with a diameter
6 11
of 10 mm under a pressure of 10 MPa. Then the pressed samples were wrapped in Fe foil and sintered
at 750 °C for 1 h in flowing Ar, and finally cooled to room temperature in a tube furnace. An excess of
Mg powder was added simultaneously around the samples to compensate the loss of Mg due to its
3
evaporation at high temperature. In the experiment, the MgB samples with a density of 1.35 g/cm
2
3
in undoped and approximately 1.42 g/cm in the doped samples have been obtained. The crystal structure
was investigated by XRD (X’Pert Pro, PANalytical) with Cu K radiation. The magnetic property
α
of samples were measured over a temperature range of 5–45 K using a physical properties measurement
system (PPMS, Quantum Design), and the magnetization measurements were performed with a
1 mT applied field in zero-field cooling (ZFC). The magnetic J was calculated from the width, ∆M,
c
of the magnetization loops (M–H) using the Bean model [12], J = 20 ∆M/[a′(1 – a′/3b′)], where a′ and
c
b′ are the dimensions of the sample perpendicular to the direction of the applied magnetic field
with a′ < b′. H values were determined from the closure of the hysteresis loops with a criterion of
irr
2 −2
J = 10 A cm .
c
3 Result and discussion
Figure 1a shows the XRD patterns of the Pr O -doped and undoped MgB samples. Besides MgB
6 11 2 2
phase, small quantities of MgO and PrB are indicated. There are no peaks corresponding to Mg, B or
6
Pr O observed. This is quite different from the situation in Y O -doped MgB in which most of the
6 11 2 3 2
Y O dopant is not decomposed [6]. This result is also different from Dy O -doped MgB in which the
2 3 2 3 2
main impurity phase is DyB , other than DyB [8], indicating that the Pr doping has a different effect on
4 6
the phase structure of the doped MgB . The detailed XRD results show (Fig. 1b), that (100) and (002)
2
reflections shift towards low diffraction angles for Pr O -doped samples. The doping-level dependence
6 11
of the lattice constants for the Pr O -doped MgB is shown in Fig. 2 from which an expansion
6 11 2
of the a-axis and c-axis lattice parameters can be clearly seen. This indicates that Pr has partially substi-
tuted Mg, forming a (Mg,Pr)B alloy. Although the lattice parameters increase with doping level, this
2
increase soon becomes saturated, indicating a small solid solubility of Pr in MgB . This may be why the
2
amount of PrB increases rapidly with doping level, as shown by the rapid increase of the intensity of
6
PrB reflections in the XRD patterns shown in Fig. 1a. It is evident that the influence of Pr O doping on
6 6 11
the crystal structure of MgB is quite different from that by doping with Y O or Dy O since Y O or
2 2 3 2 3
2
3
Dy O cannot be doped into the MgB lattice but only form impurity phases mixed within the MgB
2 3 2
2
matrix.
© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com

Original
Paper
phys. stat. sol. (a) 204, No. 5 (2007)
1557
Fig. 1 (a) XRD patterns of Pr O -doped MgB with doping levels of 0, 1, 3, and 5 wt%. (b) The [100]
6
11
2
and [002] reflections for Pr O -doped and undoped MgB samples, which shows that the reflection shift
11
6
2
to lower diffraction angle with increasing doping level.
Figure 3 shows typical results of the temperature dependence of magnetization for the Pr O -doped
6 11
MgB samples. All of these samples show a sharp superconducting transition with a T spanning a small
2 c
range of 35–38 K, indicating a good quality and uniformity of the samples. With increasing doping
level, T is slowly but steadily depressed (see the inset of Fig. 3). This slow suppression of T is consis-
c
c
tent with the slow increase of lattice parameters with doping level, and may be regarded as the conse-
quence of the partial substitution of Pr at Mg site. It is obvious that the suppression of T by Pr doping is
c
not serious because Pr doping does not directly affect the boron sheet that is believed to be responsible
for superconductivity in MgB . For example, compared to the undoped MgB sample, the T of the
2 2 c
1 wt% Pr O -doped sample drops by 0.2 K. This very moderate suppression of T by Pr doping might
6 11 c
suggest that the pair-breaking effect of the magnetic moment of the rare-earth element is not striking in
MgB . However, as reflected by the changes of T and the lattice parameters, Pr doping is quite different
2
c
from Y or Dy doping, which have no influence on the crystal structure and T of MgB .
c 2
The variation of J with applied magnetic field at 10, 20, and 30 K for the doped and undoped samples
c
is shown in Fig. 4a. At the low doping level of 1 wt%, Pr O doping has clearly improved the J per-
6 11 c
formance at all temperatures comparing with the undoped sample, although the T of Pr O -doped sam-
c 6 11
ple is slightly lower than that of the undoped one. For example, at 10 K, the J for the 1 wt% Pr O -
c 6 11
5
2
4
2
doped one reaches 1.6 × 10 A/cm at 0.5 T and 4.5 × 10 A/cm at 2 T, respectively, whereas the J of
c
5
the undoped was 1.2 × 10 A/cm at 0.5 T and 3.3 × 10 A/cm at 2 T. However, as the doping level in-
2
4
2
creased, the improvement of J (H) by Pr O doping is no longer pronounced. At a doping level of
c 6 11
3 wt%, the doped sample has a slightly higher J than the undoped one in the low-field region, but the J
c
c
0.352
c
0.351
0.267
a
0.266
Fig. 2 Variation of the lattice parameters a and c
with doping level for Pr O -doped MgB . The solid
6
11
2
0.265
lines are guides to the eye.
0
2
4
6
Doping Level (wt%)
www.pss-a.com
© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1558
X. F. Pan et al.: Doping effect of Pr O on superconductivity and flux pinning of MgB
6 11 2
Fig. 3 Volume magnetization vs. temperature
curves for Pr O -doped MgB with doping levels of
6
11
2
0, 1, 3, and 5 wt%. Measurement was done in 1 mT
in the ZFC process. Inset: Variation of supercon-
ducting transition temperature T (at onset and mid-
c
dle point) with doping level for Pr O -doped MgB .
2
6
11
in high field is lower than that of the undoped sample. At a doping level of 5 wt%, the Pr O -doped
6 11
sample performs worse than the undoped one at all temperatures and in the whole magnetic field range in
this study. The Pr O -doping effect on the irreversibility behavior of MgB shows a similar feature as
6
11
2
that on J . As can be seen in Fig. 4b that illustrates the variation of H (10 K) with doping level, the H
c irr irr
(10 K) value increases at low doping level, reaches 7.2 T at the 1 wt% doping level (for comparison, the
undoped one has a H of 6.9 T at 10 K), but soon decreases with further increasing doping level. An-
irr
other phenomenon worthy of note is that there is a correlation between the full width at half-maximum
(FWHM) of the (100) peak in the XRD pattern with the irreversibility field for the samples studied in
this work. As shown in Fig. 1b, the FWHM of the (100) peak varies with the doping level. What is inter-
esting is that both the FWHM of the (100) peak and H vary with doping level in a very similar way, as
irr
shown in Fig. 4b. When H is high, the (100) peak of the corresponding sample is significantly broad-
irr
ened, and vice versa. Because the (100) peak of MgB reflects the lattice constant of a honeycomb boron
2
sheet in the MgB structure, the broadening of this peak may suggest the occurrence of some distortion
2
of the sheet. This result is consistent with the report of Yamamoto et al. [13] who observed that the
FWHM of the (110) peak, which also corresponds to distortion of the honeycomb boron sheet in MgB ,
2
has a positive correlation with the H of MgB , that is, H increases with increasing FWHM of the (110)
irr 2 irr
peak. Because the honeycomb boron sheet is responsible for the superconductivity in MgB , a distortion
2
of the sheet may result in imperfection-induced-scattering for charge carriers, thus an enhanced H .
c2
Therefore, the underpinning mechanism behind the correlation between FWHM of the (100) peak and
H should be the impurity-induced enhancement of H . For the sake of comparison, the mass density,
irr c2
T , ∆T , H and J data for samples studied in this work are summarized in Table 1.
c c irr c
As is well known, the behavior of J of a superconductor at high magnetic field is largely governed by
c
both the flux-pinning force and the strength of upper critical field. For MgB , carbon doping usually
2
Table 1 Summary of mass density, T , ∆T , H and J data for samples studied in this work.
c
c
irr
c
mid
sample
Pr O
6
density
T
c
∆T
H
J
c,10 K, 2 T
11
c
irr, 10 K
3
(g/cm )
2
(A/cm )
(wt%)
(K)
(K)
(T)
4
4
4
4
1
2
3
4
0
1
3
5
1.35
1.42
1.40
1.40
36.4
36.2
36.0
35.6
1.9
1.8
1.8
2.1
6.9
7.2
6.6
6.5
3.0 × 10
4.5 × 10
3.3 × 10
2.1 × 10
© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com

Original
Paper
phys. stat. sol. (a) 204, No. 5 (2007)
1559
Fig. 4 (a) Magnetic-field dependence of critical current density J at various temperatures for Pr O -doped MgB
2
c
6
11
with various doping levels. (b) Variations of irreversibility field H (10 K) and FWHM of (100) peak with doping
irr
level Pr O -doped MgB . A positive correlation between H and the FWHM of the (100) peak can be clearly
irr
6
11
2
seen.
affects the conduction band, resulting in an enhancement of impurity scattering, and, consequently, an
improvement the H [14]. Therefore, although the T of MaB is suppressed by carbon doping, the J (H)
c2
c
2
c
behavior is largely improved, especially at high fields. In the present study, the dopant occupies the Mg
site other than the boron sheet that is responsible for electron conduction and the superconductivity.
Therefore, Pr doping in MgB may only slightly affect the electron scattering. As a consequence, the T
2
c
is depressed slightly, and H may also be affected insignificantly. This may explain why the improve-
c2
ment of J and H in Pr O -doped MgB is not as pronounced as that in carbon-doped MgB . Also, dif-
c irr 6 11 2
2
ferent from Y O - or Dy O - doped MgB [6, 11] where Y O (or DyB ) nanoparticles serve as pinning
3 2 2 3 4
2
3
2
center for magnetic flux, Pr O nanoparticles has decomposed and formed PrB , as revealed by XRD
6 11 6
analysis (see also Fig. 1a), thus they cannot be expected to serve as nano-pinning centers in the present
system. TEM analysis (the results are not shown here) indicating that the PrB is mainly segregated at
6
grain boundary region of MgB , which is quite different from the microstructure of nanodiamond or
2
Y O -doped MgB , where the nanoparticles are inserted in the MgB matrix. Therefore, the improved
2 3 2 2
flux-pinning behavior in our samples can be mainly attributed to the Pr-doping effect in Mg sites, other
than the flux pinning being severed by secondary phases. In addition, the segregation of PrB at grain
6
boundaries may degrade the MgB grain connection, which may be why J decreases with an increased
2 c
amount of PrB in Pr O -doped MgB .
6 6 11 2
4 Summary
In summary, Pr O -doped MgB samples with various doping levels were prepared by solid-state reac-
6 11 2
tion at ambient pressure. Pr O doping affects the crystal structure of MgB , resulting in a systematic
6 11 2
increase of the lattice constants with increasing doping level, thus suggesting a partial substitution of Pr
for Mg in MgB crystal structure. As a result, the superconducting transition temperature, T , of MgB is
2
c
2
slightly but steadily suppressed. At a low doping level of 1 wt% Pr O , the critical current density, J ,
6 11 c
and the irreversibility field, H are improved. High-level doping (>3 wt%) of Pr O in MgB is harmful
irr 6 11 2
for the performance of MgB in high magnetic fields due to the segregation of PrB at the grain bounda-
2 6
ries of MgB grains, which degrades the grain connection.
2
Acknowledgements The authors are grateful to the National Science Foundation of China (grant No. 50372052,
50430137, 50588201); the Southwest Jiaotong University R&D Program (grant No. 2004A02), the Sichuan Province
Government Research Funding (2004A02), and the Australian Research Council (grant No. DP0452522,
DP0559872) for the financial support.
www.pss-a.com
© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1560
X. F. Pan et al.: Doping effect of Pr O on superconductivity and flux pinning of MgB
6 11 2
References
[1] J. Nagamatsu, N. Nakagawa, T. Muranaka, Y. Zenitani, and J. Akimitsu, Nature 410, 63 (2001).
[2] Y. Zhao, Y. Feng, C. H. Cheng, L. Zhou, Y. Wu, T. Machi, Y. Fudamoto, N. Koshizuka, and M. Murakami,
Appl. Phys. Lett. 79, 1154 (2001).
[3] Y. Zhao, D. X. Huang, Y. Feng, C. Cheng, T. Machi, N. Koshizuka, and M. Murakami, Appl. Phys. Lett. 80,
1640 (2002).
[4] Y. Zhao, C. H. Cheng, X. F. Rui, H. Zhang, P. Munroe, H. M. Zeng, N. Koshizuka, and M. Murakami, Appl.
Phys. Lett. 83, 2916 (2003).
[5] T. Takenobu, T. Ito, Dam Hieu Chi, K. Prassides, and Y. Iwasa, Phys. Rev. B 64, 134513 (2001).
[6] J. Wang, Y. Bugoslavsky, A. Berenov, L. Cowey, A. D. Caplin, L. F. Cohen, L. D. Cooley, X. Song, and D. C.
Larbalestier, Appl. Phys. Lett. 81, 2026 (2002).
[7] S. X. Dou, S. Soltanian, J. Horvat, X. L. Wang, S. H. Zhou, M. Ionescu, H. K. Liu, P. Munroe, and M. Tomsic,
Appl. Phys. Lett. 81, 3419 (2002).
[8] M. Kuhberger and G. Gritzner, Physica C 370, 39 (2002).
[9] H. Kitaguchi and H. Kumakura, Supercond. Sci. Technol. 18, S284 (2005).
[10] C. H. Cheng, Y. Zhao, X. T. Zhu, J. Nowotny, C. C. Sorrell, T. Finlayson, H. Zhang, Physica C 386, 588
(2003).
[11] S. K. Chen, M. Wei, and J. L. MacManus-Driscoll, Appl. Phys. Lett. 88, 192512 (2006).
[12] C. P. Bean, Rev. Mod. Phys. 36, 31 (1964).
[13] A. Yamamoto, J. Shimoyama, S. Ueda, Y. Katsura, I. Iwayama, S. Horii, and K. Kishio, Appl. Phys. Lett. 86,
212502 (2005).
[14] A. Gurevich, Phys. Rev. B 67, 184515 (2003).
© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com