Superconductivity in Dense $MgB_2$ Wires
Extracted source text · arXiv cond-mat/0102289
%&latex209
\documentstyle[twocolumn,prb,aps,epsfig]{revtex}
% \draft command makes pacs numbers print
\draft
\begin{document}
\title{Superconductivity in Dense $MgB_2$ Wires}
% repeat the \author\address pair as needed
\author{P. C. Canfield, D. K. Finnemore, S. L. Bud'ko, J. E. Ostenson,
G. Lapertot,\thanks{On
leave from Commissariat a l'Energie Atomique, DRFMC-SPSMS,
38054 Grenoble, France} C. E. Cunningham,\thanks{On leave from Dept. of Physics,
Grinnell College, Grinnell, IA 50112} and C. Petrovic }
\address{Ames Laboratory, U.S. Department of Energy and Department of Physics and Astronomy\\
Iowa State University, Ames, Iowa 50011}
\date{\today}
\maketitle
\begin{abstract}
% insert abstract here
$MgB_2$ becomes superconducting just below 40 K. Whereas porous polycrystalline
samples of $MgB_2$ can be synthesized from boron powders, in this letter we
demonstrate that dense wires of $MgB_2$ can be prepared by exposing boron
filaments to $Mg$ vapor. The resulting wires have a diameter of 160 ${\mu}m$,
are better than 80\% dense and manifest the full $\chi = -1/4{\pi}$
shielding in the superconducting state. Temperature-dependent resistivity
measurements indicate that $MgB_2$ is a highly conducting metal in the
normal state with $\rho (40~K)$ = 0.38 $\mu Ohm$-$cm$. Using this value,
an electronic mean free path, $l \approx 600~\AA$ can be estimated,
indicating that $MgB_2$ wires are well within the clean limit. $T_c$, $H_{c2}(T)$,
and $J_c$ data indicate that $MgB_2$
manifests comparable or better superconducting properties in dense wire
form than it manifests as a sintered pellet.
\end{abstract}
% insert suggested PACS numbers in braces on next line
\pacs{74.70.Ad, 74.25.Fy, 74.25.Ha, 74.60.Jg}
\section{Introduction}
The discovery of superconductivity in $MgB_2$ has caused a renaissance of
interest in intermetallic superconductivity.\cite {1} This, combined with the
discovery of superconductivity in $YPd_2B_2C$ and the $RNi_2B_2C$ series several
years ago,\cite {2-5} seems to indicate that the old idea of
looking for high intermetallic $T_c$ values in compounds rich in light elements is still a
valid guiding principle. Measurements of the boron isotope effect in this
compound\cite {6} are consistent with the superconductivity being mediated via
electron-phonon coupling, a conclusion that is also suported by
recent bandstructural calculations.\cite {7} Measurements of the upper
critical field, $H_{c2}(T)$, the thermodynamic critical field, $H_c(T)$, and the
critical current, $J_c$, indicate that $MgB_2$ is a type-II superconductor with
properties that are consistent with a intermetallic superconductor that has a
$T_c \approx 40~K$.\cite {8} For example, other than the
remarkably high $T_c$, $MgB_2$ appears to be
quite similar to $Nb_3Sn$. Given this similarity, and given the
far lower density of $MgB_2$ as well as the greater natural abundances of $Mg$ and $B$,
a logical question is whether wires of $MgB_2$ can be easily synthesized, and
if so what are their physical properties. In this letter we present a
remarkably simple method for the synthesis of $MgB_2$ wires from boron
filaments. In addition, we show that wires produced in this manner are
of high density and have impressively low normal state resistivity
and impurity scattering.
\section{Experimental Methods}
$MgB_2$ can be synthesized in powder form by reacting stoichiometric amounts
of powdered $B$ and $Mg$ at 950$^{\circ}$C for approximately an hour.\cite {6}
Given that at 950$^{\circ}$C the vapor pressure of $Mg$ is
approximately 200 Torr,\cite {9} it is believed that $MgB_2$ forms
via a process of diffusion of $Mg$ vapor into the boron grains.
Based on this observation, the possibility of using this technique
on other morphologies of boron appears to be promising.
$MgB_2$ wire was produced by sealing 100 ${\mu}m$ diameter boron
fiber\cite {10} and $Mg$ into a $Ta$ tube with a nominal ratio of $Mg_2B$.
Given that $MgB_2$ is the most $Mg$ rich
binary $Mg-B$ compound known,\cite {11} it was felt that excess $Mg$ would aid in the
formation of the proper, stoichiometric phase. The sealed $Ta$ tube was sealed
in quartz and then placed into a 950$^{\circ}$C box furnace for
approximately an hour. The reaction ampoule was then removed from
the furnace and quenched to room temperature.
Measurement of temperature and field dependent electrical resistivity and
magnetization were performed in Quantum Design MPMS and PPMS systems.
Resistivity measurements were made in a standard four probe geometry using
Epotek H20E silver epoxy to make contacts. The contact resistance was
approximately 1 Ohm. Given the well defined geometry of the samples, accurate
measurements of resistivity were possible.
\section{Results}
Upon opening the $Ta$ tube it became clear that there had been a reaction
between the boron fiber and the $Mg$ vapor. Whereas the boron fibers were straight
and moderately flexible before the reaction, the $MgB_2$ wires in the $Ta$ tube after
the reaction were brittle and deformed. The inset of Figure 1 is a photograph of
the resulting wires. As can be seen, there has been significant warping
and bending of the fiber as a result
of the reaction with the $Mg$ vapor. Figure 2 shows scanning electron
microscope images of the fiber before the reaction as well as the wire after
the reaction. In both cases a tungsten core (approximately 15 ${\mu}m$
diameter) can be clearly seen. This core is part of the original boron fiber
and does not appear to be effected by the exposure of the fiber to $Mg$, nor,
as will be seen, does it seem to effect the superconducting properties of the
resulting $MgB_2$ wire. Whereas the boron fiber has a diameter of 100 ${\mu}m$ and
breaks with a smooth, clean surface (inset to Fig. 1), the $MgB_2$ wire has a diameter of
approximately 160 ${\mu}m$ and breaks with a rougher, grainier surface. The
increased diameter of the wire is consistent with observations that there is
an expansion associated with the formation of the $MgB_2$ powders during
synthesis.\cite {6} Although the $MgB_2$ wires are somewhat brittle, the integrity
of the filament segments was preserved during the exposure to the $Mg$ vapor,
i.e. the fibers did not decompose or turn into powder.
Using a diameter of 160 ${\mu}m$ and measuring the length and mass
of several wire segments we determined the density of the
wire to be 2.4 $g/cm^3$. This is to be
compared with a theoretical value of 2.55 $g/cm^3$
for a single crystal sample using
lattice parameters $a$ = 3.14 $\AA$ and $c$ = 3.52 $\AA$ for the hexagonal
unit cell.\cite {6} Given
the rather coarse nature of our measurements, this implies that the wire
samples are probably better than 80\% of the theoretical density. It should
be noted that the small tungsten core would come in as a roughly 10\%
correction, and therefore is within our level of uncertainty.
Figure 1 presents the temperature-dependent magnetization of $MgB_2$. The
data were taken after cooling in zero field and then warming in a field of 25
Oe. Given the aspect ratio of the wire segments used we were able to obtain
a susceptibility very close to $-1/4{\pi}$, the value expected for total shielding
and a demagnetization factor close to zero. $T_c$ = 39.4 K can be determined from these
data by using an onset criterion (2\% of $-1/4{\pi}$). The width of the transition
(10\% - 90\%) is 0.9 K.
Figure 3 presents the temperature-dependent electrical resistivity of
$MgB_2$ wires. The room temperature resistivity has a value of 9.6 ${\mu}Ohm$-$cm$ whereas
$\rho (77~K)$ = 0.6 ${\mu}Ohm$-$cm$ and $\rho (40~K)$ = 0.38 ${\mu}Ohm$-$cm$.
This leads to a residual resistivity ratio
of $RRR$ = 25.3. It should be noted that the shape of the resistivity curve and the
$RRR$ values are qualitatively the same as those observed for sintered pellets
of polycrystalline $Mg^{10}B_2$.\cite {8} The relatively low room temperature
resistivity value, along with the high $RRR$ are not unusual for diboride
samples.\cite {12} The resistivity of the sintered pellet samples\cite {8}
is approximately 1 ${\mu}Ohm$-$cm$ at 40 K. This somewhat higher
value of the calculated resistivity for the pellet is consistent with the sintered
sample having an actual density substantially lower than the theoretical value.
The temperature-dependent resistivity shown in Fig. 3 can be fit by $\rho =
\rho_0 + \rho_1 T^{\alpha}$ with $\alpha \approx 2.6$ between $T_c$ and
200 K. This is comparable to the power law $R~=~R_1 + R_1 T^{\alpha_1}$
with $\alpha_1 \approx 2.8$
found for the sintered $Mg^{10}B_2$ sample\cite {8} over
a comparable temperature range.
Given the similarity of the two power laws it seems clear that the
resistivity of $MgB_2$ will not have a linear slope for temperatures between
$T_c$ and 300 K.\cite {7} On the other hand, using an average Fermi
velocity\cite {7} of $v_F = 4.8 \cdot 10^7~cm/s$ and a carrier density of $6.7 \cdot 10^{22}
~el/cm^3$ (two free electrons per unit cell)
we can estimate the electronic mean free path to be approximately
600 $\AA$ at $T_c$. This is clearly an approximate value of the electronic mean
free path, but given the estimated superconducting coherence length of
approximately 50 $\AA$,\cite {8} these values place
$MgB_2$ wires well within the
clean limit. Given a $\kappa \approx 26$,\cite {8} this implies that,
much like the case of the $RNi_2B_2C$
materials,\cite {5} there may be significant non-local effects
associated with $MgB_2$.
The superconducting transition temperature, $T_c$ = 39.4 K, can be determined from
both the magnetization and resistivity data shown in Figs. 1 and 3. This value
is slightly higher than the $T_c$ = 39.2 K value determined for isotopically
pure $Mg^{11}B_2$, but is significantly lower that $T_c$ = 40.2 K for $Mg^{10}B_2$.
This is consistent with an approximate 80 \% natural abundance of $^{11}B$.
It is noteworthy that the superconducting transition is both relatively high
and sharp in the wire samples. This means that either very few impurities are
being incorporated into the $MgB_2$ or that what few impurities are being
incorporated are having very little effect on either resistivity or $T_c$.
The temperature dependence of the upper critical field, $H_{c2}(T)$, is shown
in the inset to Fig. 3. For each field three data points are shown: onset
temperature, temperature for maximum $d{\rho}/dT$, and completion temperature.
Qualitatively these data are remarkably similar to the $H_{c2}(T)$ data
inferred from measurements on $Mg^{10}B_2$ sintered pellets\cite {8}
as well as recent measurements on hot-pressed powders.\cite {13}
Quantitatively,
at $H$ = 9 T, the width of the resistive transition for the wire sample
is roughly half of the width found for the sintered sample.
These data are consistent
with the wire sample being of comparable or better quality as the sintered
powder samples.
Figure 4 presents data on the critical current $J_c$. The open symbols are $J_c$
values extracted from direct measurements of the current dependent voltage
across the sample at given temperature and applied field values. The
filled symbols are $J_c$ values inferred from magnetization loops by
application of the Bean model.\cite {8,14} The direct measurement
of $J_c$ was limited to values below approximately 200 $A/cm^2$ due to resistive
heating from the sample leads and contact resistance. As can be seen, the
extrapolations of the directly measured, low $J_c$, data and the
Bean-model-inferred, high $J_c$, data match up moderately well.
In comparison to the $J_c$ data presented for a sintered pellet of
$Mg^{10}B_2$,\cite {8} $J_c$ for the wire sample is roughly a factor
of two higher at low fields and temperatures and over an order of magnitude
higher at high fields.
\section{Conclusions}
We have devised a simple technique of producing low resistivity, high density,
high $T_c$ $MgB_2$ in wire form via exposure of boron filaments to $M$g vapor.
The resulting wire has better than 80\% the theoretical density of
$MgB_2$ and measurements
of the temperature dependent resistivity reveal that $MgB_2$ is highly conducting
in the normal state. The room temperature resistivity has a value of
9.6 ${\mu}Ohm$-$cm$ whereas the resistivity at $T$ = 40 K is 0.38 ${\mu}Ohm$-$cm$.
This means that even in the normal state wires of $MgB_2$ can carry
significant current densities. This should be compared with the resistivity of $Nb_3Sn$
$\rho (20~K)$ = 11 $\mu Ohm$-$cm$ and $\rho (300~K)$ = 80 $\mu Ohm$-$cm$.\cite {15}
Given the well-defined geometry of the wire samples we have been able
to directly measure a full $-1/4{\pi}$ susceptibility in the superconducting
state. The values of $T_c$ for this wire sample are slightly higher than
the $T_c$ values for isotopically pure $Mg^{11}B_2$ sintered powders and the
width of the resistive superconducting transition is smaller that that
seen for $Mg^{10}B_2$ sintered powders. In addition, $H_{c2}(T)$ for the
wire sample is virtually the same as that found for $Mg^{10}B_2$ sintered
pellets. Based on all of these observations it appears that $MgB_2$ wires
provide dense, high quality samples of $MgB_2$. By comparing our estimate
of the electronic mean free path $l \approx 600~\AA$ to the
superconducting coherence length $\xi \approx 50~\AA$\cite {8}
we can see that $MgB_2$ wires are well within the clean limit.
All of the above, of course, presents the possibilities of using
such wires for both research
and applied purposes. For basic research the possibilities of making weak link
Josephson junctions, SLUGS, and other devices are currently being pursued.
On the applied side, given that boron filaments are produced in a variety of sizes
and of arbitrary lengths, the possibility of converting boron filament into
$MgB_2$ wire as part of a continuous process leads to the possibility of
simple manufacturing of light weight, high $T_c$, wires with remarkably small
normal state resistivities. In addition, this process could be used to
turn boron coatings on tapes, cavities, or other devices into
high-quality superconducting films. Although the low
temperature $J_c$ values are currently smaller than those for
$Nb_3Sn$,\cite {8} as of yet very little effort has been put
into optimizing $J_c$. A multi-dimensional phase space of filament
purity, diameter, treatment time and temperature has yet to be explored.
Both basic and applied directions of research will have to be explored in
detail over the coming months and years.
\section{Acknowledgments}
We would like to thank V. Crist and M. Kramer for taking the electron
microscope images as well as J. R. Clem and R. W. McCallum for
useful discussions and N. Anderson
and N. Kelso for assistance in the lab. Ames Laboratory is operated for
the U. S. Department of Energy by Iowa State University under Contract
No. W-7405-Eng.-82. This work was supported by the director fo Energy
Research, Office of Basic Energy Sciences.
\begin{references}
\bibitem{1} J. Akimiitsu, Symposium on Transition Metal Oxides, Sendai, January 10, 2001;
J. Nagamatsu, N. Nakagawa, T. Muranaka, Y. Zenitani, and J. Akimitsu (to be published).
\bibitem{2} R. Nagarajan, C. Mazumdar, Z. Hossain, S. K. Dhar,
K. V. Gopalakrishnan, L. C. Gupta, C. Godart, B. D. Padalia, and
R. Vijayaraghavan, Phys. Rev. Lett. {\bf 72}, 274
(1994).
\bibitem{3} R. J. Cava, H. Takagi, B. Battlog, H. W. Zandbergen, J. J. Krajewski,
W. F. Peck, Jr., R. B. van Dover, R. J. Felder, T. Siegrist,
K. Mizuhashi, J. O. Lee, H. Eisaki, S. A. Carter, and S. Uchida,
Nature {\bf 376}, 146 (1994).
\bibitem{4} R. J. Cava, H. Takagi, H. W. Zandbergen, J. J. Krajewski,
W. F. Peck, Jr., T. Siegrist, B. Battlog, R. B. van Dover, R. J. Felder,
K. Mizuhashi, J. O. Lee, H. Eisaki, and S. Uchida,
Nature {\bf 376}, 252 (1994).
\bibitem{5} For review see: P. C. Canfield, P. L. Gammel, and D. J. Bishop,
Physics Today {\bf 51}, 40 (1998) and references therein.
\bibitem{6} S. L. Bud'ko, G. Lapertot, C. Petrovic, C. E. Cunningham, N. Anderson,
and P. C. Canfield, Phys. Rev. Lett. (in press); cond-mat/0101463.
\bibitem{7} J. Kortus, I. I. Mazin, K. D. Belashchenko, V. P. Antropov, and L. L. Boyer,
cond-mat/0101446.
\bibitem{8} D. K. Finnemore, J. E. Ostenson, S. L. Bud'ko,
G. Lapertot, and P. C. Canfield, cond-mat/0102114.
\bibitem{9} R. E. Honig, and D. A. Kramer, Vapor Pressure Curves
of the Elements, Fall 1968, RCA Laboratories, Princeton N.J.
\bibitem{10} Textron Systems, 201 Lowell St., Wilmington, MA 01887.
\bibitem{11} Binary Alloy Phase Diagrams, Second Edition,
Edited by T. Massalski, (A.S.M. International, 1990).
\bibitem{12} A. N. Emel'yanov, V. P. Kobyakov, V. I. Ponomarev,
T. G. Utkina, and A. S. Shteinberg, Phys. Met. Met. {\bf 83}, 608 (1997).
\bibitem{13} D. C. Larbalestier, M. Rikel, L. D. Cooley, A. A. Polyanskii, J. Y. Jiang,
S. Patnaik, X. Y. Cai, D. M. Feldmann, A. Gurevich, A. A. Squitier, M. T.
Naus, C. B. Eom, E. E. Hellstrom, R. J. Cava, K. A. Regan, N. Rogado, M. A.
Hayward, T. He, J. S. Slusky, P. Khalifah, K. Inumaru, and M. Hass,
unpublished.
\bibitem{14} C. P. Bean, Phys. Rev. Lett. {\bf 8}, 225 (1962).
\bibitem{15} Study of Transition Temperatures in Superconductors,
Final Report, prepared by L. J. Vieland, and R. W. Cohen, RCA
Laboratories, Princeton, NJ 08540, 1970.
\end{references}
%
\begin{figure}
\epsfxsize=0.9\hsize
\vbox{
\centerline{
\epsffile{F1bC.eps}
}}
\caption{Magnetization divided by applied field (25 $Oe$) for zero field
cooled wire sample. Field was applied parallel to the wire length, leading
to a demagnetization factor close to zero. Inset: photograph of wires as
they appear after removal from $Ta$ tube and part of U. S. dime for scale.}
\label{F1}
\end{figure}
%
\begin{figure}
\epsfxsize=0.9\hsize
\vbox{
\centerline{
\epsffile{F2C.eps}
}}
\caption{Electron microscope image of cross section of grown $MgB_2$
wire. The diameter of the wire is 160 ${\mu}m$. Inset: electron microscope
image of the un-reacted boron filament. The diameter of the filament is
100 ${\mu}m$. For both images the wire / filament was snapped
in-situ. Note: In both images a central core of tungsten wire
(diameter $\approx 15~{\mu}m$) can
be clearly seen. }
\label{F2}
\end{figure}
%
\begin{figure}
\epsfxsize=0.9\hsize
\vbox{
\centerline{
\epsffile{F3a.eps}
}}
\caption{Temperature dependent electrical resistivty of $MgB_2$ wire. Lower
inset: Expanded view for temperatures near $T_c$. Upper inset: $H_{c2}(T)$ data
inferred from temperature dependent resistivity data taken in constant
applied field upon cooling. The three symbols are for onset,
maximum slope and completion temperatures. }
\label{F3}
\end{figure}
%
\begin{figure}
\epsfxsize=0.9\hsize
\vbox{
\centerline{
\epsffile{F4.eps}
}}
\caption{Superconducting critical current density, $J_c$, as a function of
applied field every 5 K in 5 - 35 K range. Open symbols were
taken via direct measurement of current dependent voltage of the wire. Filled
symbols were determined via a Bean model analysis of
magnetization data from wire samples with the applied field parallel to the
wire length. The dashed lines simply connect data sets taken at the same
temperature.}
\label{F4}
\end{figure}
%
\end{document}