1: \documentstyle[aps,epsfig,preprint,prb]{revtex}
2: %\documentstyle[aps,epsfig,twocolumn,prb]{revtex}
3:
4: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5:
6: \begin{document}
7:
8: %\twocolumn[\hsize\textwidth\columnwidth\hsize\csname %
9: %@twocolumnfalse\endcsname
10:
11: \draft
12:
13: %\wideabs {
14:
15: \title{Scaling of $H_{c2 \perp}(T)$ in Nb/CuMn Multilayers}
16:
17: \author {A. Angrisani Armenio, L. V. Mercaldo, S. L. Prischepa,\thanks{
18: Permanent address: State university of Informatics and
19: RadioElectronics, P. Brovka Str. 6, 220600 Minsk, Belarus} M.
20: Salvato, C. Attanasio\thanks{Corresponding author.
21: Telephone:+39-089-965313; Fax: +39-089-953804; E-mail:
22: attanasio@sa.infn.it}, and L. Maritato}
23: \address{Dipartimento di Fisica and INFM, Università
24: degli Studi di Salerno, Baronissi (Sa), I--84081, Italy.}
25:
26:
27:
28: \date{\today}
29: \maketitle
30:
31: \begin{abstract}
32:
33: Measurements of the perpendicular upper critical magnetic field
34: $H_{c2 \perp}(T)$ are reported for several Nb/CuMn multilayers.
35: It is found that, despite the magnetic nature of the samples, the
36: data for samples with low Mn percentage in the CuMn layers are
37: simply described by the Werthamer-Helfand-Honenberg theory for
38: conventional type-II superconductors, neglecting both Pauli spin
39: paramagnetism and spin orbit impurity scattering. For high Mn
40: concentration a different theoretical aprroach is needed.
41: \end{abstract}
42:
43:
44: \pacs{Key words: superconductivity, spin glass, multilayers,
45: critical magnetic field. Running head: Nb/CuMn multilayers}
46:
47: \section{INTRODUCTION}
48:
49: The issue of the proximity effect in superconducting multilayers
50: has been intensely studied since the early
51: sixties\cite{Degennes,Hauser1}. In particular since
52: superconductivity and magnetism are two mutually excluding
53: phenomena a lot of interest has been devoted to the analysis of
54: superconducting (S)/magnetic (M) multilayers\cite{Hauser2,Jin}.
55: Several theoretical and experimental studies have been done on
56: these systems, in particular about the existence of the so-called
57: $\pi-phase$ state which manifests itself in a nonmonotonic
58: behavior of the transition temperature $T_c$ of the S/M
59: multilayers as a function of the magnetic layers thickness
60: $d_M$\cite{Radovic1}. Numerous experiments about the behavior of
61: $T_c$ versus $d_M$ have been reported on different S/M
62: multilayers\cite{Korevaar,Verbank,Strunk,Jiang,Mughe,Mercaldo,Attanasio}:
63: However the presence of $T_c$ oscillations is still an open
64: question and further theoretical investigation is needed. In
65: particular in the case of Nb/CuMn multilayers (where CuMn is a
66: spin glass) the presence of a small Mn percentage ($\geq$ 0.7 \%)
67: in copper gives rise to a nonmonotonic behavior of $T_c$ versus
68: $d_M$ \cite{Mercaldo,Attanasio} which cannot be explained in the
69: framework of the conventional proximity theory even when taking
70: into account a paramagnetic pair breaking mechamism\cite{Hauser2}.
71:
72: Another interesting feature of S/M multilayers is the temperature
73: behavior of the upper critical magnetic field, both in the
74: direction parallel, $H_{c2 \parallel}(T)$, and perpendicular,
75: $H_{c2 \perp}(T)$, to the plane of the film, which shows
76: deviations from the S/N case (here N is a normal metal).
77: Measurements performed on V/Fe multilayers\cite{Korevaar}
78: revealed a good agreement with the theoretical predictions for
79: S/M multilayers\cite{Radovic2}. Both $H_{c2 \parallel}(T)$ and
80: $H_{c2 \perp}(T)$ could be consistently described using the same
81: value for the only free parameter of the theory. On the other
82: hand the same measurements performed on Nb/CuMn samples with 2.7
83: \% and 4.5 \% of Mn could be only qualitatively described by the
84: same theory, probably due to the much more complicated nature of
85: a spin glass system with respect to the ferromagnetic
86: case\cite{PhysicaC1}.
87:
88: In this paper we report on measurements of the perpendicular
89: upper critical magnetic field as a function of temperature in
90: Nb/CuMn multilayers. A large number of samples have been measured
91: with different Mn percentage and different layering, also in the
92: presence of a regular array of
93: antidots\cite{PhysicaC1,Jap,Philmag,Prbsubmitted}. Regardeless of
94: the specific nature of the multilayers, the measurements for the
95: samples with low percentage of Mn in the CuMn layers (up to 2.7
96: \%) are in agreement with the Werthamer-Helfand-Honenberg (WHH)
97: theory, which describes the behavior of conventional type-II
98: superconductors\cite{WHH}. This result indicates that $H_{c2
99: \perp}(T)$ measurements are less sensitive to the presence of Mn
100: than the measurements of $T_c$ versus the magnetic layers
101: thickness. However for sufficiently high Mn concentration a
102: different theoretical approach is needed to describe the data.
103:
104:
105: \section{EXPERIMENT}
106:
107:
108: Nb/CuMn multilayers have been fabricated by using a dual-source
109: magnetically enhanced dc triode sputtering system with a movable
110: substrate holder on Silicon (100) substrates\cite{Mercaldo}. The
111: bottom layer is CuMn and the top layer is Nb for all the samples.
112: Some of the samples are patterned into 200$\times$200 $\mu$m$^2$
113: zones with a regular array of antidots and suitable contact pads.
114: The preparation details for these samples, obtained by lift-off
115: procedure, are reported elsewhere\cite{Philmag,Prbsubmitted}. All
116: the samples present good superconducting properties and a well
117: defined layered structure as shown by low angle X-ray diffraction
118: patterns\cite{PhysicaC1}.
119:
120: Transport measurements have been performed with a standard {\sl
121: dc} four probe technique with magnetic field applied
122: perpendicular to the plane of the film. In figure \ref{fig1} the
123: resistive transitions of one of the analyzed samples (sample
124: A(20)1) are reported at different values of the external
125: perpendicular magnetic field. The $H_{c2 \perp}(T)$ values have
126: been obtained from the $R(T)$ curves, at different applied
127: magnetic fields, using the 50\% $R_N$ criterion, where $R_N$ is
128: the normal state resistance just before the transition to the
129: superconducting state. However, even if different criteria are
130: used to extract the $H_{c2 \perp}(T)$ from $R(T,H)$ curves, no
131: substantial differences are observed in the results. We have also
132: occasionally performed $R(H)$ measurements at different
133: temperatures to extract the $H_{c2 \perp}$ value for each
134: temperature at the intersection point between the flux flow
135: regime and the normal state resistance\cite{Larkin}. The critical
136: magnetic field values obtained with the two different methods are
137: always in good agreement with each other.
138:
139: Table I shows the sample characteristics: Nb thickness $d_{S}$,
140: CuMn (Cu) thickness $d_M$, Mn percentage, number of bilayers,
141: superconducting critical temperature $T_c(K)$ and anisotropy
142: ratio $\zeta = H_{c2 \perp}(0)/H_{c2 \parallel}(0)$. A column is
143: added in the end to point out the patterned samples with the
144: array of antidots.
145:
146: \section{RESULTS AND DISCUSSION}
147:
148: The upper critical field measurement allows us to investigate the
149: nature of the pair-breaking mechanism present in our
150: superconducting-spin glass multilayers. Figure \ref{fig2} shows
151: the reduced perpendicular critical magnetic field $h_{c2}=H_{c2
152: \perp}/[T_c(-dH_{c2 \perp}/dT)|_{T=T_c}]$ as a function of the
153: reduced temperature $t=T/T_c$ for all the investigated samples.
154: The data have been analyzed in the framework of the WHH theory
155: which widely describes the $H_{c2}(T)$ behavior of bulk type-II
156: superconductors, including the case where the effect of applied
157: magnetic field on the electron spin magnetic moments cannot be
158: neglected. In particular Pauli spin paramagnetism and spin-orbit
159: scattering are taken into account, respectively through the
160: parameters $\alpha$ and $\lambda_{so}$, which appear in the
161: implicit equation for the reduced field $h_{c2}$\cite{WHH}:
162:
163: \begin{equation}
164: ln\Bigg({1 \over t}\Bigg)=\Bigg({1 \over 2} + {i \lambda_{so}
165: \over 2 \gamma}\Bigg) \Psi\Bigg({1 \over 2} + {\bar h_{c2} + {1
166: \over 2}\lambda_{so} + i \gamma \over 2 t}\Bigg)+ \Bigg({1 \over
167: 2} - {i \lambda_{so} \over 2 \gamma}\Bigg) \Psi\Bigg({1 \over 2}
168: + {\bar h_{c2} + {1 \over 2}\lambda_{so} - i \gamma \over 2
169: t}\Bigg)-\Psi\Bigg({1 \over 2}\Bigg)
170: \end{equation}
171:
172: \noindent where $\psi$ is the digamma function, $\bar h_{c2}=(4/
173: \pi^2)h_{c2}$ and $\gamma=[(\alpha \bar
174: h_{c2})^2-((1/2)\lambda{so})^2]^{1/2}$.
175:
176: While data for all the multilayers with high Mn percentage are
177: not described by the WHH theory, for the samples with Mn
178: percentage up to 2.7 all the experimental points collapse on the
179: WHH curve calculated for the case $\alpha=\lambda_{so}=0$. This
180: quite surprising result indicates that a small percentage of Mn
181: does not significantly influence the $H_{c2 \perp}(T)$ curves and
182: Nb/CuMn multilayers behave, at least for temperatures down to
183: $t=0.3$, like ordinary type-II superconductors. $H_{c2 \perp}(T)$
184: measurements are then less sensitive to Mn concentration than
185: measurements of $T_c$ versus $d_M$. A nonmonotonic $T_c(d_M)$
186: dependence was observed even for 0.7 \% of Mn\cite{Mercaldo},
187: revealing an unconventional proximity effect in the system, while
188: 2.7 \% of Mn is still not sufficient to cause an appreciable
189: deviation from a conventional $H_{c2 \perp}(T)$ behavior.
190:
191: In figure \ref{fig2} measurements for a Nb/Cu multilayer, the
192: sample M(0), are also shown. Again these data collapse on the WHH
193: curve with $\alpha=\lambda_{so}=0$. Same results have been
194: obtained for Nb/Pd\cite{Cocco} multilayers. On the other hand
195: data from samples with high Mn percentage cannot be fitted to the
196: WHH theory even in the case $\alpha \neq 0$ and $\lambda_{so} \neq
197: 0$. In fact in the $h_{c2}-t$ plane all these data lie above the
198: points obtained in the small Mn percentage case, while
199: theoretical curves with $\alpha$, $\lambda_{so} \neq 0$ are
200: always below the $\alpha = \lambda_{so} = 0$ curve. Similar
201: results apply for V/Fe\cite{Korevaar} and Nb/Gd\cite{Strunk}
202: multilayers and Nb/Pd$_{1-x}$Fe$_x$/Nb triple
203: layers\cite{Vonlohn}, with $x \neq 0$, when plotted in the WHH
204: fashion. Also in these cases the data lie above the WHH curve
205: with $\alpha=\lambda_{so}=0$. These results show that an
206: additional pair breaking mechanism, which is not taken into
207: account in the WHH theory, is present both in
208: superconducting/ferromagnetic and some superconducting/spin glass
209: multilayers, such as Nb/CuMn with high Mn percentage ($>$ 2.7
210: \%). In this case a realistic explanation requires theories which
211: explicitly take into account the magnetic nature of the non
212: superconducting material in the multilayers\cite{Radovic2}.
213:
214:
215:
216: \section{CONCLUSIONS}
217:
218: Measurements of $H_{c2 \perp}(T)$ have been performed on several
219: Nb/CuMn multilayers with different Mn percentage in the CuMn
220: layers, also in the presence of regular array of antidots. It is
221: found that the $H_{c2 \perp}(T)$ curves for samples having low Mn
222: percentage are described by a conventional theory for type-II
223: superconductors despite the magnetic nature of the samples,
224: regardless of the layering and of a more complicated structure,
225: i.e. if a regular array of antidots is present.
226:
227:
228: \begin{references}
229:
230: \bibitem{Degennes} P. G. de Gennes and E. Guyon, Phys. Lett. {\bf 3}, 168 (1963).
231:
232: \bibitem{Hauser1} J. J. Hauser, H. C. Theuerer, and N. R. Werthamer, Phys. Rev. {\bf 136}, A637 (1964).
233:
234: \bibitem{Hauser2} J. J. Hauser, H. C. Theuerer, and N. R. Werthamer, Phys. Rev. {\bf 142}, 118 (1966).
235:
236: \bibitem{Jin} B. Y. Jin and J. B. Ketterson, Adv. Phys. {\bf 38}, 189 (1989).
237:
238: \bibitem{Radovic1} Z. Radovi\'c, M. Ledvij, L. Dobrosavlijevi\'c-Gruij\'c, A. I. Buzdin, and J. R. Clem, Phys. Rev. B {\bf 44}, 759 (1991).
239:
240: \bibitem{Korevaar} P. Koorevaar, Y. Suzuki, R. Coehoorn, and J. Aarts, Phys. Rev. B {\bf 49}, 441 (1994).
241:
242: \bibitem{Verbank} G. Verbanck, C. D. Potter, R. Schad, P. Belien, V. V. Moshchalkov, and Y. Bruynseraede, Physica C {\bf 235-240}, 3295 (1994).
243:
244: \bibitem{Strunk} C. Strunk, C. S\"urgers, U. Paschen, and H.v. L\"ohneysen, Phys. Rev. B {\bf 49}, 4053 (1994).
245:
246: \bibitem{Jiang} J. S. Jiang, D. Davidovi\'c, D. H. Reich, and C. L. Chien, Phys. Rev. Lett. {\bf 74}, 314 (1995).
247:
248: \bibitem{Mughe} Th. M\"uhge, N. N. Garif'yanov, Yu. V. Goryunov, G. G. Khaliullin, L. R. Tagirov, K. Westrholt, I. A. Garifullin, and H. Zabel, Phys. Rev. Lett. {\bf 77}, 1857 (1996).
249:
250: \bibitem{Mercaldo} L. V. Mercaldo, C. Attanasio, C. Coccorese, L,
251: Maritato, S. L. Prischepa, and M. Salvato, Phys. Rev. B {\bf 53},
252: 14040 (1996).
253:
254: \bibitem{Attanasio} C. Attanasio, C. Coccorese, L. V. Mercaldo, S. L. Prischepa, M. Salvato, and L. Maritato, Phys. Rev. B {\bf 57}, 14411 (1998).
255:
256: \bibitem{Radovic2} Z. Radovi\'c, L. Dobrosavlijevi\'c-Gruij\'c, A. I. Buzdin, and J. R. Clem, Phys. Rev. B {\bf 38}, 2388 (1988).
257:
258: \bibitem{PhysicaC1} C. Attanasio, C. Coccorese, L. V. Mercaldo, M. Salvato, L. Maritato, S. L. Prischepa, C. Giannini, C. Tapfer, L. Ortega, and
259: F. Comin , Physica C {\bf 312}, 112 (1999).
260:
261: \bibitem{Jap} C. Attanasio, L. Maritato, M. Salvato, S. L. Prischepa,
262: B. N. Engel, and C. M. Falco, J. Appl. Phys. {\bf 77}, 2081 (1995).
263:
264: \bibitem{Philmag} C. Attanasio, T. Di Luccio, L. V. Mercaldo, S. L. Prischepa, R. Russo, M. Salvato, L. Maritato, and S. Barbanera, Philosophical Magazine {\bf 80}, 875 (2000).
265:
266: \bibitem{Prbsubmitted} C. Attanasio, T. Di Luccio, L.V. Mercaldo, S.L. Prischepa, R. Russo, M. Salvato, L. Maritato, S. Barbanera, and A. Tuissi, Phys. Rev. B accepted for publication.
267:
268: \bibitem{WHH} N. R. Werthamer, E. Helfand, and P. C. Honenberg, Phys. Rev. {\bf 147}, 295 (1966).
269:
270: \bibitem{Larkin} A. I. Larkin and Yu. N. Ovchinnikov, in {\sl Non Equilibrium Superconductivity}, edited by P. N. Langenberg and A. I. Larkin (North Holland, Amsterdam, 1986).
271:
272: \bibitem{Cocco} C. Coccorese, C. Attanasio, L. V. Mercaldo, M. Salvato, L. Maritato, J. M. Slaughter, C. M. Falco, S. L. Prischepa, and B. I. Ivlev,
273: Phys. Rev. B {\bf 575}, 7922 (1998).
274:
275: \bibitem{Vonlohn} M. Sch\"ock, C. S\"urgers, and H.v. L\"ohneysen, Eur.Phys. J. B {\bf 14}, 1 (2000).
276:
277:
278: \end{references}
279:
280: \vspace{10.in} Table I. Relevant sample parameters. See the text
281: for the meaning of the listed quantities.
282:
283: \begin{tabular}{|c||c|c|c|c|c|c|c|}
284: \hline\hline
285:
286: \rm Sample &
287: $d_{S}$ $(\rm \AA)$
288: & $d_{M}$ $(\rm \AA)$
289: & \% Mn
290: & $\rm N_{bil}$
291: & $T_c(K)$
292: & $\zeta$
293: & \rm Antidot lattice \\
294: \hline
295: M(27)1 & 260 & 6 & 2.7 & 10 & 6.02 & 1.0 & No \\
296: M(27)2 & 260 & 9 & 2.7 & 10 & 5.42 & 1.0 & No \\
297: M(27)3 & 260 & 11 & 2.7 & 10 & 4.96 & 1.2 & No \\
298: M(27)4 & 260 & 16 & 2.7 & 10 & 4.22 & 1.8 & No \\
299: M(27)5 & 260 & 19 & 2.7 & 10 & 4.03 & 3.3 & No \\
300: M(27)6 & 260 & 24 & 2.7 & 10 & 4.58 & 3.1 & No \\
301: M(27)7 & 260 & 29 & 2.7 & 10 & 4.50 & 5.0 & No \\
302: M(45)1 & 350 & 4 & 4.5 & 10 & 6.67 & 1.37 & No \\
303: M(45)2 & 350 & 11 & 4.5 & 10 & 5.46 & 1.74 & No \\
304: M(45)3 & 350 & 15 & 4.5 & 10 & 3.78 & 1.45 & No \\
305: M(45)4 & 350 & 29 & 4.5 & 10 & 3.67 & 7.23 & No \\
306: M(45)5 & 350 & 32 & 4.5 & 10 & 3.61 & 5.41 & No \\
307: A(20)1 & 250 & 8 & 2 & 6 & 7.54 & 1.46 & Yes \\
308: A(20)2 & 250 & 12 & 2 & 6 & 7.38 & 1.36 & Yes \\
309: A(20)3 & 250 & 20 & 2 & 6 & 6.96 & 1.41 & Yes \\
310: A(20)4 & 250 & 24 & 2 & 6 & 6.66 & 1.72 & Yes \\
311: A(20)5 & 250 & 28 & 2 & 6 & 6.5 & 1.76 & Yes \\
312: M(0)& 200 & 200 & 0 & 10 & 6.67 & 1.6 & No \\
313: \hline\hline
314: \end{tabular}
315:
316: %\newpage
317:
318: %\begin{center}
319:
320:
321: %FIGURES
322:
323: %\end{center}
324:
325: %\vspace{1cm}
326:
327: \begin{figure}[h]
328: \vskip2cm
329: \begin{center}
330: \leavevmode \epsfxsize=15cm \epsfysize=10cm \epsffile{whh1.eps}
331: \end{center}
332: \caption{\label{fig1}{Transition curves for different
333: perpendicular magnetic fields for the sample A(20)1. The curves
334: correspond to increasing fields, from right to left, equal to
335: 0.0, 0.4, 0.7, 0.8, 0.9, 1.0, 1.1, 1.2, 1.5, 1.6, 1.7, 1.8, 1.9,
336: 2.0, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8, 2.9, 3.0 T.}}
337: \end{figure}
338:
339: \vskip1cm
340:
341: \begin{figure}[h]
342: \begin{center}
343: \leavevmode \epsfxsize=15cm \epsfysize=10cm \epsffile{whh2.eps}
344: \end{center}
345: \caption{\label{fig2}{Reduced perpendicular magnetic field
346: $h_{c2}$ versus reduced temperature $t$ for all the analyzed
347: samples. Diamonds refer to the samples of the series M(45); open
348: squares to the samples of the series M(27); circles to the
349: samples of the series A(20) and full triangles to the sample
350: M(0). The solid line is the WHH theoretical curve obtained for
351: $\alpha=\lambda_{so}=0$.}}
352: \end{figure}
353:
354:
355: \end{document}
356: