throbber
VOLUME 80, NUMBER 9
`
`P H Y S I C A L R E V I E W L E T T E R S
`
`2 MARCH 1998
`
`Magnetic Moment of fcc Fe(111) Ultrathin Films by Ultrafast Deposition on Cu(111)
`
`J. Shen, P. Ohresser, Ch. V. Mohan, M. Klaua, J. Barthel, and J. Kirschner
`Max-Planck-Institut für Mikrostrukturphysik, Weinberg 2, D-06120 Halle/Saale, Germany
`(Received 19 September 1997)
`Thermally deposited ultrathin Fe films on Cu(111) at room temperature or below have been known
`to grow in a multilayer mode with a low net magnetic moment 共⬃(cid:48)(cid:46)(cid:53)m(cid:66)兲. In the present work we have
`used pulsed laser deposition to produce isotropic fcc Fe兾Cu共(cid:49)(cid:49)(cid:49)
`films which grow layer by layer with
`a magnetic moment of more than (cid:50)m(cid:66). We attribute the larger magnetic moment of the pulsed laser
`deposited films to their structural perfection as seen by low energy electron diffraction and scanning
`tunneling microscopy.
`[S0031-9007(98)05386-1]
`

`
`PACS numbers: 75.70.Ak, 68.55.–a, 75.30.Cr, 81.15.Fg
`
`fcc g(cid:45)Fe is known to have various magnetic structures.
`These include paramagnetic, antiferromagnetic, ferrimag-
`netic, low-moment ferromagnetic, or high-moment ferro-
`magnetic states depending on the lattice constant [1,2].
`Generally for g(cid:45)Fe a larger lattice constant favors a larger
`moment as well as the ferromagnetic alignment. How-
`ever, the most interesting high-moment 共.(cid:50)m(cid:66)兲 ferromag-
`netic phase has never been experimentally observed in
`any isotropic fcc Fe system despite theoretical predictions
`[3,4]. In bulk, fcc g(cid:45)Fe has zero net moment: It can only
`be stabilized either above 1200 K in a paramagnetic state,
`or quenched as antiferromagnetic inclusions in a Cu matrix
`[5].
`In the case of ultrathin films, a paradoxical situation
`arises: On the one hand, one has to epitaxially grow Fe
`on an fcc substrate with relatively large lattice constant to
`achieve large atomic volume and thus the high-moment
`phase of Fe; on the other hand, growing Fe on such sub-
`strates will unavoidably lead to a vertical expansion of the
`Fe lattice constant, which makes the films no longer fcc-
`like. A possible substrate for growing isotropic fcc Fe is
`copper due to the small lattice misfit. On the (100) sur-
`face, the high-moment phase was observed to exist up to
`(cid:52) ML [6] but in this thickness region the films are tetrag-
`onally distorted [7]. The net moment of Fe films falls
`immediately [8] after they become fcc-like (5 to 10 ML)
`[7]. On the (111) surface, the Fe films appear to have near
`isotropic fcc structure [9] but magnetic measurements on
`films capped by copper [10] as well as on films supported
`by a stepped substrate [11] indicate that Fe possesses a sig-
`nificantly smaller net moment of about (cid:48)(cid:46)(cid:53)m(cid:66). Therefore,
`the very fundamental question, i.e., whether an isotropic
`fcc Fe system can be high-moment ferromagnetic or not,
`has yet to be answered experimentally.
`In this Letter we demonstrate the first experimental
`finding of an isotropic fcc Fe system which has a high-
`moment ferromagnetic phase. The system is ultrathin
`Fe film on Cu(111) prepared by pulsed laser deposition
`(PLD). The PLD technique greatly increases the instan-
`taneous deposition rate by 5 or 6 orders relative to that of
`thermal deposition (TD). The PLD films have an isotropic
`fcc structure and a layer-by-layer morphology up to 6 ML
`
`as compared to the multilayer-island morphology of the
`TD Fe films. Most importantly, we have observed the
`high-moment ferromagnetic phase in the PLD Fe films up
`to 3 ML. We will discuss the mechanisms supporting the
`high-moment phase later in the paper.
`The experiments were performed in an ultrahigh vacuum
`(UHV) multichamber system including facilities for scan-
`ning tunneling microscopy (STM), Auger electron spec-
`troscopy (AES), low energy electron diffraction (LEED),
`and magneto-optical Kerr effect (MOKE). The base pres-
`sure is better than (cid:53) 3 (cid:49)(cid:48)2(cid:49)(cid:49) mbar and never exceeds
`(cid:50) 3 (cid:49)(cid:48)2(cid:49)(cid:48) mbar during deposition. Prior to film deposi-
`tion the copper(111) substrate 共miscut ,(cid:48)(cid:46)(cid:49)–兲 was cleaned
`by cycles of Ar1 sputtering followed by 700 K anneal-
`ing until a contamination free Auger spectrum and a sharp
`LEED 共(cid:49) 3 (cid:49)兲 pattern have been achieved. The substrate
`was then kept at 220 K and placed 100 mm away from an
`Fe target (99.99% purity). The output of an excimer laser
`with KrF (248 nm wavelength, 34 ns pulse length, typical
`pulse energy (cid:50)(cid:55)(cid:48) (cid:51)(cid:48)(cid:48) mJ, and repetition rate 5 Hz) was
`focused onto the Fe target, resulting in an instantaneous de-
`position rate, i.e., the deposition rate during the duration of
`the plasma plume (on the order of microsecond) created by
`each laser pulse, of Fe on Cu(111) of about (cid:49)(cid:48)(cid:54) ML兾min.
`The film thickness was controlled by reflection high energy
`electron diffraction and cross-examined by STM and AES
`afterwards. For comparison, films on the same substrate at
`the same temperature were also prepared by thermal evap-
`oration from an iron wire (5N purity) heated by e-beam
`bombardment. The magnetic properties of both types of
`the films were recorded by MOKE.
`By pulsed laser deposition,
`the morphology of the
`Fe兾Cu共(cid:49)(cid:49)(cid:49)兲 films has been improved remarkably towards
`layer by layer. Figure 1 shows a side-by-side comparison
`of the morphology of 1.0, 2.0, and (cid:51)(cid:46)(cid:53) ML Fe兾Cu共(cid:49)(cid:49)(cid:49)兲
`films prepared by thermal (left column) and pulsed laser
`(right column) deposition. The thermally deposited Fe
`films have a typical multilayer morphology: The substrate
`has not been wetted after 2 ML of Fe deposition. All the
`islands are bilayer (marked as 2) or trilayer (marked as
`3) high at nominal thickness of 1 ML, and contain five
`
`1980
`
`0031-9007兾98兾80(9)兾1980(4)$15.00
`
`© 1998 The American Physical Society
`
`Lambeth Magnetic Structures, LLC Exhibit 2001
`
`LMBTH-000001
`
`

`
`VOLUME 80, NUMBER 9
`
`P H Y S I C A L R E V I E W L E T T E R S
`
`2 MARCH 1998
`
`pulsed laser deposited films have a significantly more
`stable fcc structure with respect to the fcc ! bcc phase
`transformation. Figure 2 shows the interlayer spacing
`of the PLD films as a function of the thickness as
`calculated from LEED intensity vs energy curves based
`on a kinematic model. Below 6 ML the interlayer spacing
`is close to that of the substrate and above 6 ML it
`decreases to a distinctly smaller value. The inset LEED
`patterns indicate that the films are in fcc(111) structure
`below 6 ML and transform into the bcc(110) structure
`with Kurdjumov-Sachs orientation at higher thickness.
`Therefore, both our LEED and STM studies suggest that
`the pulsed laser deposited films remain isotropic fcc below
`6 ML. Figures 1 and 2 unambiguously prove that by
`means of ultrafast deposition, i.e., pulsed laser deposition,
`the morphology of the Fe兾Cu共(cid:49)(cid:49)(cid:49)兲 films is indeed largely
`improved towards layer by layer, and the fcc ! bcc phase
`transformation has been considerably delayed.
`Knowing the morphology and structure of the films,
`we now compare the magnetic properties of thermally
`and pulsed laser deposited films. As the present Let-
`ter is focused on the magnetic moment, we will concen-
`trate here only on the saturation magnetization while we
`leave most of the magnetic properties such as Curie tem-
`perature, magnetic anisotropy, and the spin reorientation
`to a forthcoming paper [12]. The saturation magnetiza-
`tion was obtained from saturated MOKE hysteresis loops,
`which were measured always in the same polar geometry
`

`
`the pulsed laser deposited
`spacing of
`Interlayer
`FIG. 2.
`Fe兾Cu共(cid:49)(cid:49)(cid:49)
`films as a function of thickness. The inset pictures
`are typical LEED patterns corresponding to films below and
`above 6 ML, respectively. The decrease of the interlayer spac-
`ing at 6 ML is consistent with the LEED patterns indicating a
`structural transformation from fcc(111) to bcc(110).
`
`1981
`

`
`FIG. 1. STM topography images of Fe兾Cu共(cid:49)(cid:49)(cid:49)
`films pre-
`pared by thermal deposition (left column) and pulsed laser
`deposition (right column). The pulsed laser deposited films
`show typical morphology of a good layer-by-layer growth. The
`thermally deposited films grow in a multilayer mode. At thick-
`ness as low as 1 ML, the thermal film is formed by bilayer
`islands (marked as 2), trilayer islands (marked as 3), and bcc
`precipitates. At 2 ML, the islands typically contain five layers
`in height. Note at 3.5 ML the pulsed laser deposited film stays
`fcc-like, while the thermal film has become mainly bcc-like.
`
`In addition, some
`layers at nominal thickness of 2 ML.
`elongated ridgelike structures have been observed in the
`films. These structures are recognized as bcc(110) precipi-
`tates because they have the same elongated morphology
`(along 具(cid:48)(cid:49)(cid:49)典 directions) as that of the thick films 共.(cid:52) ML兲
`whose structure has been determined to be bcc(110) by
`LEED. When the thickness increases to 3.5 ML, the
`bcc precipitates become dominant and the LEED pattern
`(not shown here) accordingly shows structure of bcc(110)
`with Kurdjumov-Sachs orientation. The fcc ! bcc phase
`transformation quickly proceeds between 2 and 4 ML,
`above which the films have been fully transformed into
`bcc morphology according to our STM studies.
`For the pulsed laser deposited films, the morphology
`is close to a layer-by-layer one. At 1 ML, more than
`90% of the substrate surface has been covered by the
`Fe atoms, while in the 2 ML film the second layer
`contributes more than 85% of the total surface area.
`Moreover, we have obtained clear evidence that
`the
`
`LMBTH-000002
`
`

`
`VOLUME 80, NUMBER 9
`
`P H Y S I C A L R E V I E W L E T T E R S
`
`2 MARCH 1998
`
`irrespective whether the easy magnetization axis is paral-
`lel or perpendicular to the film surface. Typical hysteresis
`loops of both thermally and pulsed laser deposited films
`at 1.6 ML are shown in Fig. 3. The two loops, measured
`at 40 K, differ strongly. The PLD loop has a well-defined
`rectangular shape and a very small coercivity of 135 Oe.
`The TD loop is significantly tilted with a large coercivity
`of 1150 Oe, reflecting the imperfection of the TD films as
`shown in Fig. 1. In this respect, the very small coercivity
`and rectangular shape of the hysteresis of the PLD film is
`yet another proof of its structural perfection.
`The most remarkable difference, however, is the satu-
`ration magnetization. At the same nominal thickness of
`1.6 ML, the saturation magnetization of the PLD film is
`almost 4 times larger than that of the TD film.
`In Fig. 4
`the saturation magnetization is displayed as a function of
`thickness. Here all the data points, except two points for
`the PLD films below 1.5 ML, were acquired at 150 K by
`liquid (cid:78)(cid:50) cooling instead of 40 K by liquid He cooling.
`Since the Curie temperature of both PLD and TD films is
`above 200 K [12], 150 K turns out to be sufficiently be-
`low the Curie temperature where the magnetization falls
`rapidly. Below 1.5 ML, the Curie temperature of the PLD
`films is lower than 150 K; therefore, the two data points
`were measured at 40 K. It is also important to note here
`that we consider the measured Kerr intensity, no matter
`from the fcc or transformed bcc films, to be proportional
`to the magnetization [11].
`For the thermally deposited films in Fig. 4, their mag-
`netization is strongly correlated with the fcc ! bcc phase
`transformation. In the thickness range where the fcc phase
`is dominant 共,(cid:50) ML兲, the magnetization of the films has
`low values increasing near linearly with film thickness.
`Between 2 and 4 ML, the magnetization increases steeply,
`in accordance with the rapid progress of the fcc ! bcc
`phase transformation in this region.
`In the bcc region
`

`
`FIG. 3. Comparison of Kerr hysteresis loops of a 1.6 ML
`Fe兾Cu共(cid:49)(cid:49)(cid:49)
`film prepared by thermal deposition and pulsed
`laser deposition. The loops were measured at 40 K. The
`saturation signal of the pulsed laser deposited film is about
`4 times larger than that of the thermal film. The former has
`also a square shape with a much smaller coercivity reflecting
`the improved morphology shown in Fig. 1.
`
`1982
`
`共.(cid:52) ML兲, the magnetization of the films again rises pro-
`portionally with thickness but is nearly 4 to 5 times larger
`than the expected values based on the linear extrapolation
`from the fcc films 共,(cid:50) ML兲. This means that by thermal
`growth, the net magnetic moment of the fcc films is about
`4 to 5 times smaller than that of the bcc films. Assum-
`ing the typical moment of (cid:50)(cid:46)(cid:50)m(cid:66) for the transformed bcc
`films, we estimate the net moment of the TD films in the
`fcc regime to be about (cid:48)(cid:46)(cid:53)m(cid:66), which is consistent with
`previous results from the copper capped films [10] and the
`films supported by a stepped substrate [11].
`two different
`For the pulsed laser deposited films,
`regions can be distinguished in Fig. 4. Below 3 ML
`(region I),
`the magnetization increases linearly with
`increasing thickness, reaching its maximum at about
`3 ML. The magnetic moment in this region appears to
`be close to that of the transformed TD bcc films, i.e.,
`(cid:50)(cid:46)(cid:50)m(cid:66) per atom. This moment value corresponds to the
`high-moment ferromagnetic phase since it is clearly larger
`than the net moment of any other magnetic structure
`predicted by the theories [1,2]. Above (cid:51) ML (region II),
`the magnetization falls abruptly to 30% to 40% of that
`of the 3 ML film. Upon further thickness increasing, the
`

`
`FIG. 4. Saturation magnetization as a function of the thick-
`ness of the thermally (open circles) and pulsed laser deposited
`(full circles) Fe兾Cu共(cid:49)(cid:49)(cid:49)
`films. The thermal films have low
`magnetization below 2 ML ((cid:48)(cid:46)(cid:53)m(cid:66) per atom per monolayer).
`Between 2 and 4 ML, their magnetization increases quickly,
`in accordance with the rapid progress of the fcc ! bcc phase
`In the bcc region 共.(cid:52) ML兲, the
`transformation in this region.
`magnetization of the films is nearly 4 times larger than the ex-
`pected values based on the linear extrapolation from the fcc
`films 共,(cid:50) ML兲. For the pulsed laser deposited films, the mag-
`netization initially increases linearly at a rate of about (cid:50)m(cid:66) per
`atom per monolayer reaching the maximum at about 3 ML.
`Above 3 ML, the magnetization first falls down to 30% to 40%
`of that of the 3 ML film, and then linearly increases with a
`slope of about (cid:48)(cid:46)(cid:55)m(cid:66) per atom per monolayer. Note that at a
`given nominal thickness below 3 ML, the magnetization of the
`pulsed laser deposited films is strongly enhanced, up to a factor
`of 4 to 5, as compared to that of the thermal films.
`
`LMBTH-000003
`
`

`
`VOLUME 80, NUMBER 9
`
`P H Y S I C A L R E V I E W L E T T E R S
`
`2 MARCH 1998
`
`magnetization again rises linearly with a 3 times smaller
`slope than the initial slope below 3 ML. We estimate the
`net moment of the films in region II to be about (cid:48)(cid:46)(cid:55)m(cid:66) per
`atom averaged over the film.
`As mentioned, the transition from low-moment to high-
`moment phase in the TD Fe films is the direct result of the
`fcc ! bcc phase transformation in this system. For the
`PLD films, the structural origin of the abrupt decrease of
`the net magnetic moment around 3 ML, however, appears
`to be less clear. Figure 2 indicates that neither a structural
`transformation nor a distinct change of the lattice constant
`occurs around 3 ML thickness.
`In this respect,
`the
`PLD Fe兾Cu共(cid:49)(cid:49)(cid:49)兲 system has to be distinguished from
`the well-studied TD Fe兾Cu共(cid:49)(cid:48)(cid:48)兲 system, where an abrupt
`drop of the magnetization has also been observed around
`4 ML thickness [8]. The latter is generally considered as
`a result of the structural relaxation from fct to fcc which
`apparently does not occur in the PLD Fe兾Cu共(cid:49)(cid:49)(cid:49)兲 case.
`An obvious question is as follows: Why do the ther-
`mally grown Fe兾Cu共(cid:49)(cid:49)(cid:49)兲 films have a low net moment in
`the first few monolayers? Two possible mechanisms may
`be responsible for the small moment. First, there exists
`two additional paths for the strain relaxation in the TD Fe
`films: (1) via the edge atoms [13] of the multilayer island
`and (2) via the bcc precipitates [14]. Therefore strain re-
`laxation is more likely to occur in the TD Fe films than
`in the PLD films, which could result in a smaller atomic
`volume, thus a smaller moment, for the TD films. Sec-
`ond, the thermally grown Fe films may have mixed signif-
`icantly with copper due to the high diffusivity of the (111)
`surface. The net moment of Fe will be reduced by copper
`diffusion: In an extreme case, the films become nonmag-
`netic when deposited at 370 K where a significant inter-
`diffusion occurs [15]. The pulsed laser deposited films,
`however, are less likely affected by copper diffusion be-
`cause of the fast deposition rate.
`The finding of the high-moment phase for the pulsed
`laser deposited Fe兾Cu共(cid:49)(cid:49)(cid:49)兲 films 共,(cid:51) ML兲 has also clearly
`demonstrated that it is not essential for the fcc Fe films to
`expand tetragonally to stabilize the high-moment phase.
`Both the tetragonally expanded Fe兾Cu共(cid:49)(cid:48)(cid:48)兲 films and the
`isotropic Fe兾Cu共(cid:49)(cid:49)(cid:49)兲 films have high-moment ferromag-
`netic phase at low thicknesses, but the transition from the
`high-moment to a low-moment phase has different origins
`in the two systems. The instability of the moment of fcc Fe
`may be understood in the following way: If changes of the
`lattice constant do occur, as in the Fe兾Cu共(cid:49)(cid:48)(cid:48)兲 system, the
`magnetic moment of the fcc Fe films will change accord-
`ingly; but even if the lattice constant remains unchanged,
`the magnetic moment of the film as a whole will still be
`
`changed if the thickness is high enough and the influence
`of the Fe兾Cu interface is low enough. Our experimental
`data (Fig. 4) suggest that in the fcc Fe(111) films the trans-
`formed phase is a low-moment ferromagnetic or ferrimag-
`netic phase in contrast to the antiferromagnetic phase in the
`fcc Fe(100) films. The determination of the detailed mag-
`netic structure of the Fe兾Cu共(cid:49)(cid:49)(cid:49)兲 films, however, needs
`theoretical studies using ab initio calculations.
`In summary, we have successfully modified the mor-
`phology of the Fe兾Cu共(cid:49)(cid:49)(cid:49)兲 films by increasing the de-
`position rate by about 5 to 6 orders of magnitude using
`pulsed laser deposition. The pulsed laser deposited films
`共,(cid:51) ML兲 have a significantly enlarged magnetic moment
`indicating that the isotropic fcc Fe兾Cu共(cid:49)(cid:49)(cid:49)兲 ultrathin films
`are high-moment ferromagnetic.
`The authors thank G. Kroder and F. Pabisch for their
`technical support.
`
`[1] V. L. Moruzzi, P. M. Markus, and J. Kubler, Phys. Rev. B
`39, 6957 (1989).
`[2] Yu-Mei Zhou, Wen-qing Zhang, Lie-ping Zhong, and
`Ding-sheng Wang, J. Magn. Magn. Mater. 145, L273
`(1995).
`[3] R. Lorenz and J. Hafner, Phys. Rev. B 54, 15 937 (1996).
`[4] T. Asada and S. Blügel, Phys. Rev. Lett. 79, 507 (1997).
`[5] S. C. Abrahams, L. Cuttman, and J. S. Kasper, Phys. Rev.
`127, 2052 (1962); G. J. Johanson, M. B. McGirr, and D. A.
`Wheeler, Phys. Rev. B 1, 3208 (1970).
`[6] R. D. Ellerbrock, A. Fuest, A. Schatz, W. Keune, and R. A.
`Brand, Phys. Rev. Lett. 74, 3053 (1995).
`[7] S. Müller, P. Bayer, C. Reischl, K. Heinz, B. Feldmann,
`H. Zillgen, and M. Wuttig, Phys. Rev. Lett. 74, 765
`(1995).
`[8] J. Thomassen, F. May, B. Feldmann, M. Wuttig, and H.
`Ibach, Phys. Rev. Lett. 69, 3831 (1992).
`[9] D. Tian, F. Jona, and P. M. Marcus, Phys. Rev. B 45,
`11 216 (1992).
`[10] W. Kümmerle and U. Gradmann, Phys. Status Solidi A
`45, 171 (1978).
`[11] J. Shen, M. Klaua, P. Ohresser, H. Jenniches, J. Barthel,
`Ch. V. Mohan, and J. Kirschner, Phys. Rev. B 56, 11 134
`(1997).
`[12] P. Ohresser, J. Shen, Ch. V. Mohan, M. Klaua, J. Barthel,
`and J. Kirschner (unpublished).
`[13] J. Fassbender, U. May, B. Schirmer, R. M. Jungblut, B.
`Hillebrands, and G. Güntherodt, Phys. Rev. Lett. 75, 4476
`(1995).
`[14] J. Shen, C. Schmidthals, J. Woltersdorf, and J. Kirschner,
`Surf. Sci. (to be published).
`[15] J. Shen, M. Klaua, P. Ohresser, Ch. V. Mohan, J. Barthel,
`and J. Kirschner (unpublished).
`
`1983
`
`LMBTH-000004

This document is available on Docket Alarm but you must sign up to view it.


Or .

Accessing this document will incur an additional charge of $.

After purchase, you can access this document again without charge.

Accept $ Charge
throbber

Still Working On It

This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.

Give it another minute or two to complete, and then try the refresh button.

throbber

A few More Minutes ... Still Working

It can take up to 5 minutes for us to download a document if the court servers are running slowly.

Thank you for your continued patience.

This document could not be displayed.

We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.

Your account does not support viewing this document.

You need a Paid Account to view this document. Click here to change your account type.

Your account does not support viewing this document.

Set your membership status to view this document.

With a Docket Alarm membership, you'll get a whole lot more, including:

  • Up-to-date information for this case.
  • Email alerts whenever there is an update.
  • Full text search for other cases.
  • Get email alerts whenever a new case matches your search.

Become a Member

One Moment Please

The filing “” is large (MB) and is being downloaded.

Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.

Your document is on its way!

If you do not receive the document in five minutes, contact support at support@docketalarm.com.

Sealed Document

We are unable to display this document, it may be under a court ordered seal.

If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.


Access Government Site

We are redirecting you
to a mobile optimized page.





Document Unreadable or Corrupt

Refresh this Document
Go to the Docket

We are unable to display this document.

Refresh this Document
Go to the Docket