Health & Lifestyle

Giant magneto-resistive granular layers deposited by TVA method

Giant magneto-resistive granular layers deposited by TVA method
of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Vacuum 76 (2004) 131–134 Giant magneto-resistive granular layersdeposited by TVA method I. Mustata a , C.P. Lungu a,  , A.M. Lungu a , V. Zaroski a , M. Blideran a , V. Ciupina b a National Institute for Lasers, Plasma and Radiation Physics, PO Box MG-36, Magurele-Bucharest, Romania b Department of Physics, ‘‘Ovidius University’’, Constanta, Romania Abstract The purity degree of metallic giant magneto-resistive (GMR) thin layers is a parameter, that can drastically influencetheir performance. As known, when impurities are present, the thin layer resistance increases up to a level where thespin interactions responsible for the GMR effect are hindered. By using the thermionic vacuum arc (TVA) method, themetal deposition takes place in high or ultrahigh vacuum conditions, no presence of any gas being necessary as forexample in the case of sputter depositions techniques.In order to test the TVA method, simultaneous discharges in Co and Cu and also in Co and Ag vapors were ignited.The measured values of the GMR effects (defined as [ R ( H  )–  R (0)]/ R (0), %) were found situated between 5% and 33%and showed a particular behavior presented in the paper. r 2004 Published by Elsevier Ltd. Keywords:  GMR effect; Granular layers; Thermionic vacuum arc (TVA); Co/Ag; Co/Cu 1. Introduction The granular giant magneto-resistive (GMR)layers can be obtained by simultaneous deposi-tions of magnetic and non-magnetic metals usingmethods like electron beam evaporation, sputter-ing, vacuum arc deposition [1–4]. After deposition,thermal treatment of the layers facilitates magneticcluster formation in the non-magnetic crystallinenetwork. In order to ensure its anti-parallelorientation in the absence of the external magneticfields, the mean distance between magnetic clustersmust be small [4].The free electrons traveling through the layersuffer different collisions depending on theirparallel or anti-parallel spin orientation relativeto the clustered spins. Indeed, the collisionfrequencies of the electrons carrying a parallelspin are much lower than those of the electronswith anti-parallel spins.An external applied magnetic field can impose atotal parallel orientation of all clusters so that theelectrons having the spin parallel oriented on this ARTICLE IN PRESS$-see front matter r 2004 Published by Elsevier Ltd.doi:10.1016/j.vacuum.2004.07.003  Corresponding author. Fax: +40-21-457-4468. E-mail address: (C.P. Lungu).  direction suffer collisions of very low frequencyresulting in a very low electrical resistance. Thiscreates a short cut in the electronic current anddecreases the total layer resistance. Usually, thiseffect is measured by the rate of the resistancevariation due to the magnetic field to the resistancein the absence of the magnetic field that is [2,3]: ½ R ð H  Þ 2 R ð 0 Þ = R ð 0 Þ ; % :  ð 1 Þ For preparation of GMR granular layers, ansrcinal method, which is an intermediate betweenelectron beam evaporation and electrical vacuumarc discharge, known as thermionic vacuum arc(TVA) discharge was used [5–7]. The method usesan electron beam emitted by an externally heatedcathode (a tungsten grounded filament) acceler-ated by a high anodic voltage. The electron beamcan evaporate the anode materials as neutral pureparticles and facilitate their deposition on thesubstrate when the electron energy and currentintensity are not too high. When the anodepotential is increased up to a certain value, theevaporation rate increases as much as to allow anelectrical discharge to be ignited in the evaporatedpure material and the discharge is maintained evenwhen the discharge current is as low as a fewhundred mA. When increasing the supply voltage,the discharge current can increase up to 4–5A. Atthese intermediate arc currents, the discharge canburn only in the presence of the external cathodeheating source. 2. Experimental methods  2.1. Experimental arrangement The experimental setup is given in Fig. 1 where,for brevity reasons, the vacuum chamber is notpresented.In a vacuum chamber with a minimum pressureof 10  5 Pa, two independent groups of depositionwere mounted, one for Co and the other for Cu orAg. Each group had its own filament and anodesupply separated by an electrical screen which wasmounted in order to minimize the reciprocalinfluence of the two plasmas.With these precautions we were able to simulta-neously maintain two arc discharges in twodifferent metals, for as long as necessary to obtaina desired film thickness. In order to obtaindifferent relative concentrations in the as depositedgranular layers, sets of glass substrate sampleswere positioned between the two plasma sources(Co and Cu (Ag)). The first sample was positionedabove the Co discharge (specimen 1, length  L  ¼  0)and the last one was placed above the Cu (Ag)discharge (specimen 16), as shown in Fig. 1. Theglass substrates had a rectangular shape of 80mm  10mm  1mm. In this way, we were ableto obtain a range of concentrations from highcobalt content (Co 4 80at%, specimen 1) towardslow cobalt content (Co o 15at%, specimen 16), asdetermined by XPS analysis of the films, using MgK a  ( h n =1253.6eV) as excitation source.  2.2. Post-deposition thermal treatment It is already known that granular-type GMRdepositions have to be thermally treated in orderto ensure proper cluster dimensions and distancesbetween them [1].Some of the deposited specimens were thermallytreated in an oven at 450 1 C for 60min in air.Other samples were not thermally treated becausewe believe that heating of the samples by theplasma thermal radiation and also by ion bom-bardment are sufficiently high to obtain sampleswith GMR effect. ARTICLE IN PRESS Fig. 1. Experimental setup. I. Mustata et al. / Vacuum 76 (2004) 131–134 132  3. Results and discussion The resistance measurements were performed bythe four points probing method. The samples weremounted between the two magnetic poles of anelectromagnet and the magnetic field was variedfrom 0 to 1.5T.The experimental setup for the measurement of the GMR effect was of current-in-plane (CIP)type. The magnetic field was in the plane of thespecimen and perpendicular to the electric currentdirection. The curves experimentally obtained arepresented in Figs. 2–4.Let us consider the curve presented in Fig. 2which represents the GMR effect for a Co–Agspecimen positioned 30cm from the Co crucibleand 34cm from the Ag one is presented (about40at% Co). A small difference between the twocurves corresponding to the two different magneticfield variations can be observed in this figure. Thevariations are rapid, indicating a uniform distribu-tion of the cobalt clusters formed during thedeposition. Due to the high density, uniformityand high purity of the thin films deposited by theTVA technique, the post thermal formed clustersare expected to have more uniform dimensionsthan those obtained by other deposition techni-ques [1,3]. Therefore, the exchange forces betweentwo adjacent clusters are expected to be practicallyof similar intensity for all cluster pairs. This meansthat, at a certain value of the external magneticfield induction, practically all clusters are reor-iented on the magnetic field direction and as aconsequence, the film resistance decreases veryrapidly. These variations appear for higher mag-netic fields, demonstrating a strong couplingbetween the clusters.A similar curve was obtained when we used theCo–Cu discharges as shown in Fig. 3. A current of the Co discharge of 250mA and a voltage of 1550V were used, whereas for the Cu discharge, acurrent of 240mA and a voltage of 600V were alsoused. The total thickness was 404nm (measured insitu with a Cressington thickness monitor) and thedeposition time was 10min. The specimen was notthermally treated after deposition. The distancebetween the sample and the Co crucible was 26cm ARTICLE IN PRESS -1.0-    (   R    H   -   R    0    )   /   R    0  ,   % H, T  Forward field Reverse field Fig. 2. GMR effect of the Co–Ag film: distance to Co anode30cm, distance to Ag anode 34cm. -1.0-    (   R    H   -   R    0    )   /   R    0  ,   % H, T  Forward field Reverse field Fig. 3. GMR effect of the Co–Cu film: distance to Co anode26cm, distance to Cu anode 26cm. -1.0-    (   R    H   -   R    0    )   /   R    0  ,   % H, T  Forward field Reverse field Fig. 4. GMR effect of the Co–Cu film: distance to Co anode24cm, distance to Cu anode 32cm. I. Mustata et al. / Vacuum 76 (2004) 131–134  133  and that from the Cu one was 26cm relating to aCo concentration of about 35at%). A maximumGMR effect of 10% was obtained.A high value of the GMR effect was alsoobserved for the samples that have not beenthermally treated after depositions. This highvalue is probably due to the thermal heating of the sample during deposition when bombarded byhigh energetic ions coming from the two plasmasand by the thermal radiation of the two anodes. Inthis case, cluster formation has taken place duringdeposition and no post-deposition treatment wasnecessary.In Fig. 4, we present the GMR effect of a Co–Cu specimen prepared in the same conditionsas the specimen characterized in Fig. 3 but withthermal treatment. This specimen was positioned24cm from Co crucible and 32cm from Cucrucible, i.e. closer to the Co source than theformer specimen (about 42at% Co).One can easily observe a great differencebetween forward and reversed field variation, thisindicating a great coercitive magnetic force. Also,the gradients are very steep and of great values,which makes these samples very good for auto-mated devices. Indeed, the film resistance re-mained practically constant up to a highmagnetic induction and it decreased very rapidlygiving a high electrical signal. The maximumGMR effect of the Co–Cu specimen post-deposi-tion treated was 33% in this case.We consider that the ease in achieving the GMReffect is due to the peculiarities of the TVAmethod: the atom evaporation is determined byelectron bombardment and not by ion sputteringas in classical cases. Due to the fact that theelectron mass is smaller than that of ions, thecluster evaporation from the anode is practicallyzero. The evaporating atoms are ionized anddirected to the substrates. The deposited layer iscontinuously bombarded by Co, Ag or Cu ionswith energies of a few hundred eV [6]. As aconsequence, fine, dense and well-structured filmswere prepared. Plasma radiation combined withion bombardment ensures high layer temperaturesand this fact allows magnetic clusters to appear.Further studies are necessary to be made inorder to find other interesting characteristics of theGMR granular layers deposited by this method. 4. Conclusions The data presented show that the TVA dis-charge can be considered as a good candidate forGMR film depositions.Magneto-resistive granular layers with a GMReffect as high as 5–10% without post-depositionthermal treatment and 33% with post-depositiontreatment were obtained.The treated specimens show very rapid resis-tance variations for small magnetic field domainsof variation, which can be very good for magneticsensitive sensors.The position of the specimens relative to the twodischarges is very critical and must be studiedmore carefully in the future by different measure-ment techniques. References [1] Xiao JQ, Jiang JS, Chien CL. J Phys Rev 1992;B46:9266.[2] Berkowitz E, Mitchell J, Carey M, Young P, Zhang S,Spada F, Parker F. Phys Rev Lett 1992;68:3745.[3] Xiao JQ, Jiang JS, Chien CL. Phys Rev Lett 1992;68:3749.[4] Rubin S. Riesenmagnetowiderstandseffect in granularSystemen aufgebaut aus wohldefinierten Co Clustern ineiner Ag-Matrix. PhD thesis, Ko ¨ln, Germany, 1997.[5] Musa G, Ehrich H, Mausbach M. J Vac Sci Technol1994;A12:2887.[6] Mustata I, Cretu M, Cudalbu C, Lungu CP, Musa G,Ehrich H, Schumann J, Hegemann T. Proceedings of the25th ICPIG, Nagoya, Japan, Vol. 3, 2001. 313.[7] Biloiu C, Ehrich H, Musa G. J Vac Sci Technol2001;A19:757. ARTICLE IN PRESS I. Mustata et al. / Vacuum 76 (2004) 131–134 134
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks