Enano newsletter issue 23
1 like1,837 views
The document provides an overview of current research on nanostructured active magneto-plasmonic materials and their applications in sensing and information technology. It highlights the goals of the nanomagma EU project, which include the development of active materials with tailored properties and proposed devices benefiting from the interplay between magneto-optics and plasmonics. Various studies and articles presented in the document explore advancements in magneto-plasmonic technologies, including advanced sensing mechanisms and novel device applications.
![nanoresearch
Ultrafast acousto-magneto-plasmonics
in hybrid metal-ferromagnet
multilayer structures
Vasily Temnov, MIT Chemistry Department
77 Massachusetts Avenue, 02139 Cambridge, MA.
temnov@mit.edu
Nanostructured metal surfaces are
presently used to effectively couple light
to surface plasmons. This technology
is also key to on-chip miniaturization of
plasmonic sensors. Strong sub-wavelength
confinement of optical surface plasmon fields
combined with their macroscopic propagation
distances exceeding tens of micrometers
makes it possible to perform sophisticated
nanoplasmonic experiments using
conventional far-field optical microscopy in
novel hybrid nanostructures. Combining these
nano-optical experiments with femtosecond
time-resolved spectroscopic pump-probe
techniques opens the door to fundamental
studies at the nano-scale and ultrafast
characterization of nano-optical devices.
Here we present a new plasmonic sensor,
based on a tilted slit-groove interferometer
(Fig. 1) milled by a focused ion beam into
a single noble metal film [1] or into a hybrid
metal-ferromagnet multilayer structure [2].
Surface plasmons excited at the groove
Fig. 1 > (a) Scanning electron microscopy image
propagate towards the slit, where they
of a slit-groove microinterferometer in a 200 nm
interfere with incident light (Fig. 1b). Due thin gold film. The width of the slit is 100 nm, the
to the tilt angle the optical transmission groove is 200 nm wide and 100nm deep. (b) The
through the slit shows a pronounced entire area of the microinterferometer is illuminated
periodic interference pattern (Fig. 1c). A by a spatially coherent laser beam. Surface plasmon
small modulation of the complex surface is excited at the groove, propagates towards the slit,
plasmon wave vector is accumulated over a where it is converted into free space radiation and
long propagation distance between the slit interferes with directly transmitted light. (c) Optical
and the groove and results into measurable transmission shows a pronounced interference
changes in the contrast and phase shift of pattern along the slit axis, see Ref. [1] for details./
the plasmonic interference pattern [1]. There
exist different ways to modulate the wave In hybrid magneto-plasmonic gold-cobalt-
vector of surface plasmons using multilayer gold trilayers a few nanometer thin cobalt
structures. layer is sandwiched between two gold layers 05](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-5-320.jpg)
![nanomagma
through the (111) gold
layer at the speed of
sound cS(111)=3450
m/s is converted into
a tensile pulse upon
reflection from the
gold-air interface. The
acoustic pulse creates
a transient multilayer
structure with higher
(lower) density of
free electrons for the
compressive (tensile)
acoustic pulses.
The wave vector of
femtosecond surface
plasmon probe pulses
propagating along the
Fig. 2 > Active magneto-plasmonic interferometry gold-air interface serves as a sensitive probe
in tilted slit-groove interferometers patterned in Au/
to the local perturbations of the electron
Co/Au multilayer structures. The magnetic field of
an electromagnet switches the magnetization in a density within the skin depth skin=13nm
cobalt layer and thus changes the wave vector of a induced by the acoustic pulse. Varying the
surface plasmon propagating between the slit and pump-probe delay time makes it possible to
the groove, see Ref. [2] for details./ monitor the dynamics of acoustic reflection
in the plasmonic pump-probe interferogram
within the skin depth of surface plasmon: (Fig. 3b) and extract the pump-induced
h< skin=13 nm (Fig. 2). A weak external modulation ’+i ’’ of surface dielectric
magnetic field can be used to switch the function (Fig. 3c). On top of the slowly
magnetization in a ferromagnetic cobalt increasing thermal background due to the
layer and thus modify the wave vector of temperature rise at gold-air interface the
surface plasmons [2]. Magneto-plasmonic apparent acoustic echo in ’ is observed
modulation depth of up to 2% is achieved is indicating the change of surface plasmon
this geometry. It can be further increased by wave vector ksp= ’/2| |2.
covering the microinterferometer with high- Straightforward mathematical analysis
index dielectric material [3]. delivers the exponential shape of the acoustic
When combined with time-resolved optical strain pulse with the amplitude of ~10-3, see
pump-probe spectroscopy, femtosecond Fig. 4. The exponential shape of acoustic
surface plasmon interferometry captures pulses provides the heat penetration depth
the dynamics of ultrafast electronic in cobalt within 15 nm, slightly exceeding the
excitations and coherent lattice vibrations skin depth of optical pump pulses. A sharp
within skin=13nm skin depth in gold with sub-picosecond back front of the acoustic
femtosecond time resolution [1]. Using a pulse indicates that the bandwidth of the
sapphire/cobalt/gold multilayer structure acoustic phonons exceeds 1 THz.
we generate ultrashort acoustic pulses by The 600 fs temporal resolution in our
thermal expansion of a cobalt film impulsively experiment is limited by 2nm (RMS) surface
heated by femtosecond laser pump pulses roughness (SR) at gold-air interface, which
through the sapphire substrate (Fig. 3a). acts as acoustic delay line for ultrashort
06 The compressive acoustic pulse propagates acoustic pulses and prevents observation](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-6-320.jpg)
![nanomagma
(a) of higher frequencies above 1 THz. A good
agreement between theory and experiment
is obtained by solving linear acoustic wave
equation taking into account dispersion
of high-frequency phonons and surface
roughness. Using higher excitation fluence
we were able to generate much larger
acoustic pulses with strain amplitudes
exceeding 1% (corresponding uniaxial strain
of ~2 GPa). These giant acoustic pulses
experience significant non-linear reshaping
after propagation though 120 nm and 220
nm thin gold films at room temperature [4].
Numerical solutions of the Korteveg-de
Vries equation provide an accurate and
nearly fit-free approximation of experimental
strain pulses obtained at different
excitation levels. We could not observe
any signatures of ultrasonic attenuation
in these measurements suggesting
that THz phonons in gold propagate
(b) over macroscopic distances most likely
exceeding 1 micron at phonon frequency of
1 THz. Technological challenge of fabricating
atomically smooth metal interfaces
should be tackled in order to safely detect
the acoustic phonons with frequencies
exceeding 1 THz and thus provide access
(c) to the mean free path of acoustic phonons
over the entire Brillouin zone. Femtosecond
time-resolved pump-probe measurements
in such structures may lead to the new
type of acoustic spectroscopy in solids with
ultrahigh (μeV) spectral resolution. Given
the large amplitude, short duration and
eventually loss-free propagation of acoustic
pulses generated in hybrid gold-cobalt
multilayer structures we envision many
interesting applications for the nonlinear
Fig. 3 > Femtosecond ultrasonics probed with
ultrashort surface plasmon pulses. (a) An ultrashort acoustics at the nano-scale. The possibility
compressive acoustic pulse is generated by thermal to switch magnetization in the magneto-
expansion of 35 nm thin fs-laser heated cobalt plasmonic and magneto-optical devices
transducer and propagates through a 120 nm thin by giant acoustic pulses represents just
gold film at the speed of sound. The dynamics one example with potentially high impact
of acoustic reflection is captured in a plasmonic in the field of ultrafast telecommunication
pump-probe interferogram (b) and results into the
technology.
pronounced modulation of the wave vector for a
time-delayed femtosecond surface plasmon probe I am indebted for the invaluable
08 pulse (c)./ contributions to this research project by](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-8-320.jpg)
![(a)
nanomagma
my collaboration partners from
IMM-CSIC Madrid, TU Berlin, TU
Chemnitz, Konstanz University
and Massachusetts Institute of
Technology. Financial support
by the German Research
Foundation and European
Networks of Excellence
‘Nanomagma’ and ‘Phoremost’
is deeply appreciated.
References
[1] V.V. Temnov, K.A. Nelson, G.
Armelles, A. Cebollada, T. Thomay.
A. Leitenstorfer, R. Bratschitsch,
(b) Optics Express 17 (2009) 8423.
[2] V.V. Temnov, G. Armelles,
U. Woggon, D. Guzatov, A.
Cebollada, A. Garcia-Martin,
J.M. Garcia-Martin, T. Thomay,
A. Leitenstorfer, R. Bratschitsch,
Nature Photonics 4 (2010) 107.
[3] D. Martin-Becerra, J.B. Gonzalez-
Diaz, V.V. Temnov, A. Cebollada,
G. Armelles, T. Thomay, A.
Leitenstorfer, R. Bratschitsch, A.
Garcia-Martin, M. Ujue-Gonzalez,
Appl. Phys. Lett. 97 (2010)
183114.
[4] V.V. Temnov, C. Klieber,
K.A. Nelson, T. Thomay, A.
Leitenstorfer, D. Makarov, M.
Albrecht, R. Bratschitsch (to be
published).
Fig. 4 > (a) The exponential acoustic pulse
generated by thermal expansion of fs-laser-
heated cobalt transducer is modified due to
the dispersive propagation through a 120
nm thin gold layer. Surface roughness (SR) at
gold-air interface smears out high-frequency
components in the pulse. (b) The measured
acoustic pulse (extracted from curves in
Fig.3c) preserves its exponential shape and
sub-picosecond acoustic front demonstrating
THz bandwidth of acoustic generation. Black
horizontal arrows in (a) and (b) indicate the
propagation direction of the acoustic pulse, see
Ref. [4] for details./ 09](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-9-320.jpg)
![nanomagma
The open format of the HORIBA Scientific-
GenOptics instruments makes MS coupling
easier and faster. The possibility of direct MS
analysis on the SPRi sensor was recently
shown [9]. The SuPRa-MS platform (Surface
Plasmon Resonance in arrays coupled with
Mass Spectrometry) combines SPRi and MS
in a single biochip. The biochip used for SPRi
(SPRi-Slide) is directly transferred to the MS
instrument. There is no need to neither elute
nor re-deposit the bound analyte. The MS
enzymatic digestion and the deposition of the
Fig. 1 > SPRi-PlexII system
MALDI matrix are performed directly on the
SPRi-Slide. The latter is then directly placed
The applications of SPRi are vast and include on the MS plate holder (Figure 2).
for example protein:protein [1], DNA:DNA [2,3], A proof-of-concept study of SPRi-MS
peptide:protein [4], polysaccharides:proteins imaging coupling was performed for the
[5] or protein:cells [6,7] interactions. The detection of LAG3 recombinant protein in
flexibility of the HORIBA Scientific-GenOptics plasma. The solution fraction of this protein is
instruments enables complex samples such a potential biomarker for breast cancer [10].
as serum and plasma to be analyzed for For this purpose, a mouse antibody (IgG2A)
clinical applications. directed against LAG3 was immobilized
The coupling of SPRi biosensors and matrix- on a SPRi-Slide using a dedicated surface
assisted laser desorption ionization mass chemistry compatible with MS analysis (NHS
spectrometry (MALDI-MS) is an innovative chemistry). Before injecting LAG3, rat serum
approach for biomarker discovery in biological albumin (RSA) was used to avoid non-specific
fluids. It permits analytes captured by SPRi binding on the surface of the biochip. Then,
to be identified and characterized by their the specific interaction of LAG3 (added in
molecular weight and peptide sequence. plasma) and IgG2A was monitored using
SPRi-MS opens a new method of detection, SPRi and images of the interaction were
quantification and structural characterization of studied. Several femtomoles/mm² of LAG3
proteins of interest. In the future,
it could help better discriminate
between sub-species within a
family of biomarkers.
In this context, the complexity lies
in the coupling of both techniques
[8]. Most strategies require the
elution of the bound analyte and
its analysis by ESI- (electrospay
ionisation) or MALDI-MS. This
procedure has many drawbacks
(analysis time, no multiplexing
capabilities, decreased sensitivity,
additional cross-contamination
risks, etc.) which delayed the
development of SPR-MS in the
diagnostic field. 11](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-11-320.jpg)
![the workflow and reduce consumable costs the measurement of glycosaminoglycan binding
during optimization processes. The coupling interactions. Anal. Chem. 80(9): 3476-3482.
with MS analysis is straightforward and easier, [6] Roupioz Y. and al. (2009). “Individual Blood-Cell
which makes it a valuable tool for biomarker Capture and 2D Organization on Microarrays” Small
identification. 2009, 5, No. 13, 1493–1497.
[7] Suraniti and al. (2007) Real-time detection of
lymphocytes binding on an antibody chip using
References SPR imaging. Lab Chip. 7: 1206-1208.
[1] Uzun and al. (2009) Production of surface plasmon [8] Boireau and al. (2009) Revisited BIA-MS
resonance based assay kit for hepatitis diagnosis. combinationb: Entire “on-a-chip” processing
Biosensors and Bioelectronics. 24(9): 2878-2884. leading to the proteins identification at low
[2] Spadavecchia and al. (2009) New cysteamine femtomole to sub-femtomole leveks? Biosensors
based functionalization for biochip applications. and Bioelectronics 24: 1121-1127.[9]
Sensors and Actuators B. 143(1):139-143. Bellon and al (2009) Hyphenation of Surface
[3] Corne and al. (2008) SPR imaging for label-free Plasmon Resonance Imaging to Matrix-Assisted
multiplexed analyses of DNA N-glycosylase Laser Desorption Ionization Mass Spectrometry by
interactions with damaged DNA duplexes. Analyst. On-Chip Mass Spectrometry and Tandem Mass
133: 1036-1045. Spectrometry Analysis. Anal. Chem. 81: 7695–
[4] Prieto and al. (2009) Synaptonemal complex 7702.
assembly and H3K4Me3 demethylation determine [10] Triebel and al (2006) A soluble lymphocyte
DIDO3 localization in meiosis. Chromosoma. 118: activation gene-3 (sLAG-3) protein as a prognostic
617-632. factor in human breast cancer expressing estrogen
[5] Mercey and al. (2008) Polypyrrole oligosaccharide or progesterone receptors. Cancer Letters.
array and surface plasmon resonance imaging for 235(1):147-53.
13](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-13-320.jpg)
![nanoresearch
Light localization on a gold nanodisk
array probed by near-field optics
Loïc Lalouat1, Lionel Aigouy1, P Prieto2, A.
. to know the position and the shape of the
Vitrey2, J. Anguita2, A. Cebollada2, M.U. field enhancement zones with regard to the
González2, A. García-Martín2 particles, as well as their vertical localization,
1 LPEM, UMR 8213 CNRS-ESPCI, Ecole and the influence of the incident polarization
Supérieure de Physique et de Chimie Industrielle, direction. For that, theoretical simulations
10 rue Vauquelin, 75231 Paris cedex 5, France. with finite difference time domain (FDTD),
2 IMM-Instituto de Microelectrónica de Madrid
finite boundary element or Green dyadic
(CNM-CSIC), Isaac Newton 8, PTM, Tres Cantos,
methods are often used, but the amount of
E-28760 Madrid, Spain.
experimental data available in the literature is
quite reduced. In this work, we present an
Arrays of metallic nanoparticles are artificial experimental study of the field localization on
structures that can find many applications in a disk array with a scanning near-field optical
physics and biology. When they are illuminated microscope (SNOM). Our experiments,
by an external light source, strong evanescent which are in good agreement with numerical
fields are localized in the near-field regions of simulations, show a strong localization of the
the particles. These strong local fields can electromagnetic field between the particles, in
be used for exciting single molecules, for the direction of the incident polarization.
performing Raman scattering, for developing The experimental SNOM set-up is shown in
biochemical sensors, or for performing Figure 1. In contrast to other SNOM techniques
nanolithography [1-4]. The knowledge of the [5,6], our probe is a submicron size fluorescent
local optical properties of these structures, particle glued at the end of a sharp tip [7]. In
like the electromagnetic field distribution, is contact with the sample surface, it absorbs
therefore of importance for developing such the local field at the excitation wavelength
applications. For instance, it is interesting and emits light at a different one. By collecting
the fluorescence as a
function of the tip position
on the surface, we obtain
a fluorescence image
which is directly related
to the intensity of the
electromagnetic field on
the surface. The particle,
which contains erbium
and ytterbium ions, is
excited at =975nm
and its fluorescence is
detected in the visible
range at =550nm. Since
this atypical excitation
Fig. 1 > Description of the
experimental set-up. The SNOM
probe is a fluorescent particle
14 glued at the end of a sharp tip.](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-14-320.jpg)
![nanomagma
process involves two photons, the collected
fluorescence is proportional to the square
of the total field intensity on the surface
[7]. The sample studied is an array of gold
nanodisks [diameter = 286nm, height=50nm,
period=500nm, (see Figure 2)] fabricated on
a glass substrate. The structure has a wide
plasmon resonance peak located just below
the excitation wavelength between 800 and
900nm.
Fig. 3 > (a,b) SNOM images measured in a
non-contact mode on the nanodisk array at
=975nm. The dotted circles indicate the
position of the disks. (c,d) FDTD calculations of
the near-field distribution on the structures. The
calculation represents the square of the intensity
of the total field which is the quantity measured
with the near-field fluorescent probe used in the
Fig. 2 > SEM picture of the studied gold nanodisk. experiments. The calculation has been performed
by taking into account of the probe size (a 160
nm large cube). The white arrows indicate the
We show in Figure 3(a) and (b), the incident polarization direction. The scale bar is
experimental near-field optical images of the 500 nm-long (taken from ref. 8).
structure measured in a non-contact mode
[8]. The incident polarization is linear and
theoretical ones which exhibit the same
indicated by the white arrow. The position of
periodic pattern, and the same polarization
the nanodisks is represented by the dotted
dependence.
white circles. The images show a periodic
pattern, with elongated bright spots. Each We show in Figure 4(a) and (b) higher
bright spot is in fact comprised of two lobes, resolution scans of the structure. Cross-
located between the disks, and aligned in the sections extracted from the experimental
incident polarization direction. and simulated images are shown in Figure
To check the validity of the SNOM results, 4(c) and (d). The curves are in excellent
we have performed an FDTD simulation agreement in terms of relative contrast. One
of the measured signal. To make a realistic can clearly see that all the electromagnetic
comparison, we calculated the square of field is concentrated between the disks in the
the total field intensity and integrated this direction of the incident polarization and that
quantity on a volume which represents the almost no light is located between the disks
fluorescent particle size. Such procedure only in the direction perpendicular to the incident
tends to broaden the size of the lobes but polarization.
does not change the shape of the pattern. Another interesting parameter which
The simulations are represented in Figure 3(c) characterizes nanodisks arrays is the vertical
and (d). An excellent agreement is observed extension of the electromagnetic field above
between the experimental images and the the surface. For instance, such parameter is 15](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-15-320.jpg)
![nanomagma
Fig. 5 > Experimental SNOM image measured in
planes perpendicular to the sample surface. The
scanning planes are indicated on the drawing.
The curve in the bottom shows the vertical
Fig. 4 > (a) : Experimental SNOM image measured decay of the measured signal (taken from ref. 8).
on the array of gold nanodisks ; (b) FDTD simulation
of the near-field optical signal; (c,d) cross-
sections parallel (direction A) and perpendicular References
(direction B) to the incident polarization The
position of the gold disks is the same than the [1] S. A. Maier, and H. A. Atwater, J. Appl. Phys.
ones shown in Fig. 3 (taken from ref. 8). 98, 011101 (2005).
[2] N. Fang, H. Lee, C. Sun, and X. Zhang, Science
308, 534 (2005).
important if we want to use the structures for
[3] S. S. Aćimović, M. P. Kreuzer, M. U. González,
performing nanolithography, because it will and R. Quidant, ACS Nano 3, 1231 (2009).
determine the penetration depth of the light [4] A. F. Koenderink, J. V. Hernández, F.
in a photoresist situated on top of the array. Robicheaux, L. D. Noordam, and A. Polman,
To have an idea of this localization, we have Nano Lett. 7, 745 (2007).
performed scans in planes perpendicular [5] M. Schnell, A. Garcia-Etwarri, A. J. Huber, K.
to the sample surface. We show in Figure 5 Crozier, J. Aizpurua, and R. Hillenbrand, Nat.
the experimentally measured signals which Photon. 3, 287 (2009).
indicate that the light is essentially confined [6] M. Salerno, N. Félidj, J. R. Krenn, A. Leitner, F.
at close distance from the surface. Above R. Aussenegg, and J. C. Weeber, Phys. Rev. B
200nm, no signal is detectable anymore. 63, 165422 (2001).
Such distance depends on the structure of [7] L. Aigouy, Y. De Wilde, and M. Mortier, Appl.
the array, and in particular on the disks size, Phys. Lett. 83, 147 (2003).
their thickness and their separation. [8] L. Aigouy, P. Prieto, A. Vitrey, J. Anguita, A.
To summarize, we have performed a study of Cebollada, M.U. González, A. García-Martín,
the light localization on a gold nanodisk array J. Labéguerie-Egéa, M. Mortier, J. Appl. Phys.
by near-field optics. The near field has been 110, 044308 (2011).
measured using a fluorescent particle glued at
the end of a sharp tip. The measured near-
field images, which represent the square of
the total field intensity, show that the light is
localized between the disks in the direction
of the incident polarization direction. The
results are in good agreement with numerical
simulations performed by finite difference time
16 domain method.](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-16-320.jpg)
![nanoresearch
Controlling fluorescence resonant
energy transfer with a magneto-optical
nanoantenna
R. Vincent and R. Carminati In the present work, we use an established
Institut Langevin, ESPCI ParisTech, CNRS, general framework for dipole-dipole energy
10 rue Vauquelin, 75231 Paris Cedex 05, France. transfer between an emitter and an absorber
in a nanostructured environment [5]. The
theory allows us to address FRET between
a donor and an acceptor in the presence
Energy transfer between a molecule in an excited
of a nanoparticle with an anisotropic
state (donor) and a molecule in the ground
electromagnetic response. For the case of
state (acceptor) underlies many significant
a nanoparticle with an anisotropic dielectric
photophysical and photochemical processes,
response (e.g., a nanoparticle made of a
from photosynthesis to fluorescence probing
ferromagnetic material exhibiting a magneto-
of biological systems. It is also of interest in
optical response), the distance dependence,
nanophotonics where efficient transfer of optical
the orientation dependence and the strength
excitations on subwavelength scales is a key
of the FRET efficiency can be changed
issue. Depending on the separation between substantially. In the case of a magneto-optical
the donor (D) and the acceptor (A), the process anisotropy, it can in principle be controlled
can be described accurately by various using the static magnetic field as an external
theories accounting for the electromagnetic control parameter.
interaction between the two species. For a
D-A distance range on the order of 2-10 nm,
which is relevant for photochemical studies
and nanophotonics, the well-established
Förster theory [1] based on quasi-static dipole-
dipole interaction has been very successful.
It shows that while Förster Resonant Energy
Transfer (FRET) is a very useful process
that can be used, for example, as a ruler for
spectroscopic measurements [2], it is a rather
weak process that goes down as the inverse Fig. 1 > Left panel: Schematic configuration of
the D-A system in the presence of a nanoparticle.
sixth power R6 of the D-A separation [3]. In
The different channels for energy transfer (direct
fact, one can introduce a length scale known or indirect) are indicated by dotted arrows. When
as the Förster distance R0 at which FRET is the transition dipoles are orthogonal, the direct
50% efficient and it is found that R0 is on the Förster transfer is disabled. Right panel: Energy-
order of a few nanometers in most practical level diagram of the FRET process between a do-
situations. For even smaller distances, Dexter nor and acceptor molecules.
[4] recognized that electronic exchange and
multipolar interactions become important and In principle, the presence of a nanostructure
a full quantum mechanical treatment must be close to a D-A couple will modify the emission
implemented. On the other hand, in the large and absorption by the transition dipoles; here
distance regime (non-negligible compared to we use the formalism to express explicitly
the wavelength), full electrodynamics is needed the FRET rate of a D-A couple interacting
to account for retardation effects. with a spherical nanoparticle exhibiting a 17](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-17-320.jpg)
![nanomagma
purely metallic response or a magneto-optical where RNP is the radius of the nanoparticle.
response. We have shown previously [5], that This simple expression shows that the ratio
the FRET rate mediated by the nanoparticle Rp/RNP is the crucial parameter that describes
can be expressed simply as follows the influence of the nanoparticle on the FRET
rate. For Rp>RNP, the nanoparticle enhances
the FRET transfer, while for Rp<<RNP, the
FRET becomes exclusively driven by the
direct transfer. Moreover, in previous work
In this expression is the energy transfer [5], we have shown that in the condition
rate from donor to acceptor mediated by the that the polarizability of the nanoparticle α(ω)
nanoparticle, 0 is the decay rate of the donor varies smoothly on the frequency range of the
in free space, R0 is the Förster radius for an spectral overlap between absorption cross
orientational factor equal to one, RA is the section of the acceptor and the normalized
distance of the acceptor to the nanoparticle, RD emission spectrum of the donor, the
the distance of the donor to the nanoparticle, polarization coupling radius Rp depends only
and Rp is the polarization coupling radius [5] on the polarizability tensor of the nanoparticle.
which describes the influence of the
nanoparticle on the FRET rate meditated by the
nanoparticle. This distance Rp defines an
influence radius of the nanoparticle, it allows to
compare the indirect FRET rate (i.e., mediated
by the nanoparticle) and the standard free-
space FRET rate .
Fig. 3 > Ratio Rp/RNP for Iron (blue solid line),
Nickel (gold dash-dotted line), and Cobalt
(red dashed line) as a function of the emission
wavelength lof the donor. RNP=10 nm. The
configuration is illustrated in the inset, showing
that the dipole are collinear and the couple
Donor-Acceptor and nanoparticle are aligned.
Fig. 2 > Two canonical configurations of the tree We illustrate the formalism for the well-known
body system Donor-NP-Acceptor studied in the metallic nanoparticle. Noble metals are known
present work. (a) Left panel: Aligned configuration. to hold plasmon resonances that enhance, for
(b) Right panel: Orthogonal configuration. The example, the polarizability of a nanoparticle.
arrows illustrate the molecular dipole orientations. Since the polarization coupling radius Rp
directly depends on the polarizability, one can
For the sake of illustration, let us consider the expect a substantial influence of the plasmon
situation in which the three bodies are aligned resonance on the FRET rate mediated by
with RD=RA=2RNP, and the transition dipole are the nanoparticle. This is indeed what we
aligned in the same direction [see Fig. 2(a)], in observe in Fig. 3, in which we have plotted
this case, we obtain the ratio Rp/RNP (with RNP=10 nm) versus the
emission wavelength of the donor for gold and
silver that are common materials in studies of
fluorescence enhancement or quenching. The
18 plasmon resonance is visible in both cases,](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-18-320.jpg)
![nanomagma
leading to an enhancement of Rp/RNP. For
instance in the case of silver, one reaches Rp/
RNP 3; for gold one has Rp/RNP 1.9. In the
particular conditions RD = RA = 2 RNP and R
= 4 RNP, we obtain an enhancement factor
of the FRET rate on the order of 180
for silver and 10 for gold. For a D-A
couple working at plasmon resonance with
these materials, we conclude that FRET is
mainly driven by the nanoparticle. Incidentally, Fig. 5 > Ratio Rp / RNP for Iron (blue solid line), Nickel
any change of the dielectric property of the (yellow dash-dotted line), and Cobalt (red dashed
nanoparticle will be reflected in a modulation line) as a function of the emission wavelength
of the FRET rate. Modulation of the dielectric of the donor in the presence of an external
response can be achieved, for example, magnetic field inducing a magnetization in the
through the magneto-optical effect that we direction orthogonal of the plane containing the
consider in the following. tree body D-A-NP. RNP =10 nm. The configuration
is illustrated in the inset.
with a change of the optical dielectric response
(anisotropic response). Therefore ferromagnetic
particles own an anisotropic dielectric response
controlled by an external magnetic field.
Fig. 4 is an illustration of the ratio Rp/RNP (with
RNP =10 nm) versus the emission wavelength
of the donor for different standard magneto-
optical materials: Nickel, Iron and Cobalt in an
Fig. 4 > Ratio Rp/RNP for Iron (blue solid line), Nickel aligned configuration (see Fig. 2(a) for a sketch
(gold dash-dotted line), and Cobalt (red dashed of the geometry). We observe a smoother
line) as a function of the emission wavelength behavior than in the case of noble metals. Its
of the donor. RNP=10 nm. The configuration is
origin lie in the stronger plasmon damping
illustrated in the inset, showing that the dipole
of magneto-optical materials comparing to
are collinear and the couple Donor-Acceptor
and nanoparticle are aligned. (b) Right panel:
metallic materials. For these materials, the
Canonical orthogonal configuration. amplification factor is around twenty, therefore
in this configuration, the FRET rate is still govern
by the nanoparticle.
Using experimental data for the dielectric
Figure 5 shows a computation of the ratio Rp/
function of different magneto-optical materials RNP with the same materials as in Fig. 4, but
[6], we compute the polarization coupling in the case of an orthogonal configuration
radius Rp normalized by the nanoparticle radius [see Fig. 2(b) for a sketch of the orthogonal
RNP as a function of several parameters: The geometry]. The magnetization is orthogonal
emission wavelength of the donor, the radius of to the plane containing the D-A couple and
the nanoparticle, and the material properties. the nanoparticle. Let us stress that in this
Ferromagnetic materials are materials with configuration the FRET rate vanishes in
magnetic anisotropy. Magnetic anisotropy is absence of an external static magnetic field
a consequence of the different directions of due to the orthogonality of the donor and
magnetization of the different magnetization acceptor transition dipoles. Although one
domains. At saturation, small nanoparticles observes that Rp/RNP remains smaller than
are customarily considered owning a single one, the possibility of inducing a FRET rate
domain. This change in magnetization comes driven only by the polarization anisotropy of the 19](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-19-320.jpg)
![nanomagma
nanoparticle is an interesting result, showing anisotropy can be controlled by an external
the potential of magneto-optical nanoparticles static magnetic field, and we have discussed
for FRET. On the one hand, the anisotropic potential application for FRET tuning and
response allows us to couple molecules for modulation. Here, we have presented a proof
which standard FRET gives a vanishing signal of concept. Further work should focus on
due to orientational mismatch (orthogonal enhancing the (weak) magneto-optical FRET
transition dipole). On the other hand, the signal. We have illustrated the effect also for the
possibility of controlling the magneto-optical well known metallic nanoparticle, showing that
response with a static magnetic field as an it furnishes insight in the understanding of the
external parameter could allow us to tune good quantities controlling this process.
or modulate the FRET rate, which can be
an advantage, for example, to increase the References
sensitivity of the detection process. [1] T. Förster, Ann. Phys. 437, 55 (1948); Discuss.
We have elucidated the Förster energy transfer Faraday Soc. 27, 7 (1959).
problem in a three body configuration, involving [2] L. Stryer, Annu. Rev. Biochem. 47, 819 (1978).
two fluorophores close to a nanoparticle [3] L. Novotny, B. Hecht, Principles of Nano-optics,
with an anisotropic dielectric response. We Cambridge University Press, (2006).
have shown that the distance dependence [4] D. L. Dexter, J. Chem. Phys. 21, 836 (1953).
is controlled by the Förster radius and a new [5] R. Vincent, and R. Carminati, Magneto-optical
distance that depends of the polarization control of Förster energy transfer, Phys. Rev. B 83,
properties of the nanoparticle. We have 165426 (2011).
illustrated the effects in the case of a magneto- [6] E. D. Palik, Handbook of Optical Constants of
optical nanoparticle for which the degree of Solids (Academic, New York, 1985).
20](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-20-320.jpg)
![www.nanociencia.imdea.org
RESEARCH PROGRAMMES
• Molecular nanoscience
IMDEA-Nanociencia is a private Foundation created by joint initia-
tive of the Comunidad de Madrid and the Ministry of Education of
• Scanning probe microscopies the Government of Spain in February 2007 to manage a new
and surfaces
research Institute in Nanoscience and Nanotechnology (IMDEA-
Nanociencia). The Institute is located at the campus of the Univer-
• Nanomagnetism sidad Autónoma de Madrid in Cantoblanco.
The Institute aims at performing research of excellence in selected
• Nanobiosystems: biomachines and
manipulation of macromolecules areas and offers attractive opportunities to develop a career in sci-
ence at various levels from Ph.D. students to senior staff positions.
• Nanoelectronics and
superconductivity The Madrid Institute for Advanced Studies in Nanoscience also
develops an important program of technology transfer and creation
of spin-off companies.
• Semiconducting nanostructures
and nanophotonics
E-mail contacto.nanociencia@imdea.org
Phone 34 91 497 68 49 / 68 51
Fax 34 91 497 68 55
• Nanofabrication and advanced
instrumentation
[Nanociencia y Nanotecnología: lo pequeño es diferente small is different
Nanoscience and Nanotechnology:
]](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-21-320.jpg)
![nanoresearch
Internal electromagnetic field
distribution and magneto-optical
activity of metal and metal-dielectric
magnetoplasmonic nanodisks
D. Meneses-Rodríguez, E. Ferreiro-Vila, J.C. structures respectively, with total heights
Banthí, P Prieto, J. Anguita, A. García-Martín,
. between 50 and 70 nm and diameters
M. U. González, J. M. García-Martín, A. between 110 and 140nm (Figure 1(a)). For
Cebollada and G. Armelles the sake of comparison, continuous thin films
IMM-Instituto de Microelectrónica de Madrid with identical composition have been also
(CNM-CSIC), Isaac Newton 8, PTM, E-28760 prepared.
Tres Cantos (Madrid), Spain.
The MO activity ( ) has been obtained
david.meneses@imm.cnm.csic.es
by measuring the MO Kerr effect in polar
configuration upon normal incidence
illumination, previously identifying the optical
Localized surface plasmon resonances
resonances through extinction spectra. In
(LSPRs) greatly influence the optical [1-4]
the fully metallic nanostructures, we find
and magneto-optical (MO) [5-10] properties
a distinctive evolution as a function of Co
of fully metallic and metal-dielectric position of the MO activity in the nanodiscs
nanostructures. The observed enhancement compared with that of the continuous layers,
in the MO activity when these LSPRs are with maximum values when the Co layer
excited is attributed to the high intensity of is located near the top or the bottom of
the electromagnetic (EM) field inside the the disks and minimum values in-between
global nanostructure when the LSPR occurs due to the LSPR excitation (Figure 1(b)).
[5,11]. Unfortunately, it is not straightforward This behavior is in contrast with the MO
to experimentally determine the intensity of activity exhibited by the continuous films,
the EM field inside a nanostructure. Here which increases monotonously as the Co
we show how the EM profile related to layer becomes closer to the top surface.
the LSPR can be probed locally inside the This indicates that the EM field inside the
nanostructure by measuring the MO activity nanodisks exhibits a nonuniform distribution
of the system as a function of the position a in plasmon resonance conditions. In fact,
MO active probe (a Co nanolayer) [12]. This the Co layer acts as a probe sensing the
will be done in full detail in metallic systems, EM field within the nanodisk, since the MO
and preliminary results will also be presented activity depends on the intensity of such field.
in more complex metal-dielectric magneto- Preliminary results on the possible influence
plasmonic nanodiscs. of multiple resonances in metal-dielectric
The magnetoplasmonic nanodisk arrays magnetoplasmonic nanodiscs will be also
have been fabricated in large area onto presented.
glass substrates by combining colloidal This information could be very relevant for
lithography with sputter, thermal and the design of magnetoplasmonic systems
electron beam deposition and lift-off offering optimum MO enhancement, for
techniques. Typical nanodisk structures are instance for sensing applications where
Au/Co/Au/Cr and Au/SiO2/Co/SiO2/Au/Ti, maximum sensitivity is expected in the areas
22 for the fully metallic and the metal-dielectric with higher EM field.](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-22-320.jpg)
![nanomagma
(a) [4] T. Pakizeh, A. Dimitriev, M. S. Abrishamian, N.
Granpayeh, and M. Häll, J. Opt. Soc. Am. B 25
(2008) 659.
[5] J. B. González-Díaz, A. García-Martín, J. M.
García-Martín, A. Cebollada, G. Armelles, B.
Sepúlveda, Y. Alaverdyan and M. Käll, Small 4
(2008) 202.
[6] G. A. Wurtz, W. Hendren, R. Pollard, R.
Atkinson, L. Le Guyader, A. Kirilyuk, Th. Rasing,
I. I. Smolyaninov and A. V. Zayats, New J. of
Phys. 10 (2008) 105012.
(b) [7] P. K. Jain, Y. Xiao, R. Walsworth, and A. E.
Cohen, Nanolett. 9 (2009) 1644.
[8] G. X. Du, T. Mori, M. Suzuki, S. Saito, H.
Fukuda, and M. Takahashi, Appl. Phys. Lett. 96
(2010) 081915.
[9] L. Wang, K. Yang, C. Clavero, A. J. Nelson, K.
J. Karroll, E. E. Carpenter, and R. A. Lukaszew,
J. Appl. Phys. 107 (2010) 09B303.
[10] G. X. Du, T. Mori, M. Suzuki, S. Saito, H.
Fukuda, and M. Takahashi, J. Appl. Phys. 107
(2010) 09A928.
[11] G. Armelles, A. Cebollada, A. García-Martín,
J. M. García-Martín, M. U. González, J. B.
González-Díaz, E. Ferreiro-Vila and J. F.
(c) Torrado, J. Opt. A: Pure Appl. Opt. 11 (2009)
114023.
[12] D. Meneses-Rodríguez, E. Ferreiro-Vila, P.
Prieto, J. Anguita, M. U. González, J. M. García-
Martín, A. Cebollada, A. García-Martín and G.
Armelles, Small, DOI: 10.1002/smll.201101060
(2011).
Fig. 1 > (a) Sketch of the fully metallic nanodiscs
(b) Maximum magneto-optical activity as a
function of the Co position for fully metallic
nanodiscs (c) SEM image of an array of metallic
nanodisc (Inset: extinction spectrum).
References
[1] S. A. Maier, Plasmonics: Fundamentals and
Applications (Springer, Berlin, 2007).
[2] S. A. Maier and H. A. Atwater, J. Appl. Phys.
98, (2005) 011101.
[3] K. H. Su, Q. H. Wei, and X. Zhang, Appl. Phys.
Lett. 88 (2006) 063118. 23](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-23-320.jpg)
![nanoresearch
Magneto-Optical properties of
nanoparticles
R. Gómez-Medina1, B. García-Cámara2, corrections to the electrostatic polarizability
I. Suárez-Lacalle1, L. S. Froufe-Pérez1, F. tensor.
González2, F. Moreno2, M. Nieto-Vesperinas3
and J. J. Sáenz1
1 Departamento de Física de la Materia Condensada
and Instituto “Nicolás Cabrera”, Universidad
Autónoma de Madrid, 28049 Madrid, Spain.
2 Grupo de Óptica, Departamento de Física
Aplicada, Universidad Cantabria, 39005 Santander,
Spain.
3 Instituto de Ciencia de Materiales de Madrid,
C.S.I.C., Campus de Cantoblanco, 28049 Madrid,
Spain.
juanjo.saenz@uam.es
Fig. 1 > Scattering cross section map of a non-
Electromagnetic scattering from nanometer- absorbing Mie sphere as a function of the
scale objects has long been a topic of refractive index m and the y parameter, y = mka =
large interest and relevance to fields from m(2 a/ ). Green areas correspond to parameter
astrophysics or meteorology to biophysics, ranges where the magnetic dipole contribution
medicine and material science [1-5]. In the dominates the total scattering cross section,
while red areas represent regions where the
last few years, small particles with resonant
electric dipole contribution is dominating. Higher
magnetic properties are being explored as order multipoles dominate the remaining blue-
constitutive elements of new metamaterials saturated areas. (Adapted from Ref. [2]).
and devices. The studies in the field often
involve randomly distributed small elements
or particles where the dipole approximation We will also explore the properties of high-
permittivity dielectric particles with resonant
may be sufficient to describe the optical
magnetic properties as constitutive elements
response. We will discuss the optical
of new metamaterials and devices [2].
response of disordered nano-materials where
Magnetic properties of low-loss dielectric
the constitutive nanoparticles can have a non-
nanoparticles in the visible or infrared are not
negligible response to static (Magneto-Optical
expected due to intrinsic low refractive index
active nanoparticles) or dynamic (Magneto-
of optical media in these regimes. Here we
dielectric nanoparticles) magnetic fields.
analyze the dipolar electric and magnetic
We will first analyze the peculiar scattering response of lossless dielectric spheres
properties of single nanoparticles. In made of moderate permittivity materials.
particular, we derive the radiative corrections For low material refractive index there are no
to the polarizability tensor of anisotropic sharp resonances due to strong overlapping
particles, a fundamental issue to understand between different multipole contributions.
the energy balance between absorption However, we find that Silicon particles with
and scattering processes [1]. As we will index of refraction 3.5 and radius 200nm
show, Magneto optical Kerr effects in non- present strong electric and magnetic dipolar
absorbing nanoparticles with magneto-optical resonances in telecom and near-infrared
24 activity arise as a consequence of radiative frequencies, (i.e. at wavelengths ≈ 1.2 – 2 μm)](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-24-320.jpg)
![nanomagma
without spectral overlap with quadrupolar and
higher order resonances. The light scattered
by these Si particles can then be perfectly
described by dipolar electric and magnetic
fields.
Fig. 3 > Scattering diagrams for a Ge nanosphere
with 240nm radius (After Ref. [4]).
Acknowledgments
We appreciate interesting discussions with J.
Aizpurua, S. Albaladejo, P. Albella, A. García-
Etxarri, M.I. Marqués and F. Scheffold. This
work has been sup- ported by the EU NMP3-
SL-2008-214107-Nanomagma, the Spanish
MICINN Consolider NanoLight (CSD2007-
00046), FIS2010-21984, FIS2009-13430-
C01-C02, and FIS2007-60158, as well as by
the Comunidad de Madrid Microseres- CM
(S2009/TIC-1476).
References
[1] S. Albaladejo,R. Gómez-Medina, L. S. Froufe-
Pérez, H. Marinchio, R. Carminati, J. F. Torrado,
G. Armelles, A. García-Martín and J.J. Sáenz,
Fig. 2 > Effective real and imaginary permittivities
Opt. Express 18 (2010) 3556.
and permeabilities for an arbitrary arrangement
of Si spheres in an otherwise homogeneous [2] A. García-Etxarri, R. Gómez-Medina, L. S.
medium with εh = μh = 1 for two different filling Froufe-Pérez, C. López, L. Chantada, F.
factors f = 0.25 (a) and f = 0.5 (b). (From Ref. [2]). Scheffold, J. Aizpurúa, M. Nieto-Vesperinas
and J. J. Sáenz, Opt. Express 19, 4815 (2011).
As we will see, the striking characteristics of [3] M. Nieto-Vesperinas, R. Gómez-Medina, and J.
the scattering diagram of small magneto- J. Sáenz, J. Opt. Soc. Am. A 28 (2011) 54.
optical and magnetodielectric particles [3,4] [4] R. Gómez-Medina, B. García-Cámara, I.
lead to a number of non-conventional effects Suárez-Lacalle, F. González, F. Moreno, M.
in the optical response of nanostructured Nieto-Vesperinas, J. J. Sáenz, J. Nanophoton.
magneto-optical structures. 5, 053512 (2011). 25](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-25-320.jpg)
![nanoresearch
Three-dimensional optical
metamaterials and nanoantennas:
Chirality, Coupling, and Sensing
Harald Giessen which are favorable for emitting and receiving
4th Physics Institute, University of Stuttgart, radiation from quantum systems [9].
D-70569 Stuttgart, Germany.
giessen@physik.uni-stuttgart.de
References
Metallic metamaterials have shown a number [1] Na Liu, Hongcang Guo, Liwei Fu, Stefan Kaiser,
of fascinating properties over the last few Heinz Schweizer, and Harald Giessen: Three-
years. A negative refractive index, negative dimensional photonic metamaterials at optical
refraction, superlenses, and optical cloaking frequencies, Nature Materials 7, 31 (2008).
are some of the ambitious applications where [2] N. Liu, H. Guo, L. Fu, S. Kaiser, H. Schweizer,
metamaterials hold great promise. and H. Giessen: Plasmon Hybridization in
Stacked Cut-Wire Metamaterials, Advanced
We are going to present fabrication methods
Materials 19, 3628 (2007).
for the manufacturing of 3D metamaterials [1].
[3] Na Liu, Liwei Fu, Stefan Kaiser, Heinz Schweizer,
We are investigating their coupling properties
and Harald Giessen: Plasmonic Building Blocks
and the resulting optical spectra. Hybridization
for Magnetic Molecules in Three-Dimensional
of the electric [2] as well as the magnetic [3]
Optical Metamaterials, Advanced Materials 20,
resonances allows us to easily understand the
3859 (2008).
complex optical properties. Lateral as well as [4] B. Lukyanchuk, N. I. Zheludev, S. A. Maier, N.
vertical coupling can result in Fano-resonances J. Halas, P. Nordlander, H. Giessen, and C.
[4] and EIT-like phenomena [5, 6]. These T. Chong: The Fano resonance in plasmonic
phenomena allow to construct novel LSPR nanostructures and metamaterials, Nature
sensors with a figure of merit as high as five [7]. Materials 9, 707 (2010).
The connection between structural symmetry [5] Na Liu, Stefan Kaiser, and Harald Giessen:
and their electric as well as magnetic dipole Magnetoinductive and Electroinductive Coupling
and higher-order multipole coupling will be in Plasmonic Metamaterial Molecules, Advanced
elucidated. It turns out that stereometamaterials Materials 20, 4521 (2008).
[8], where the spatial arrangement of the [6] Na Liu, N. Liu, L. Langguth, T. Weiss, J. Kästel,
constituents is varied (see figure), reveal a M. Fleischhauer, T. Pfau, and H. Giessen:
highly complex rotational dispersion. The chiral Plasmonic EIT analog at the Drude damping
properties are limit, Nature Materials 8, 758 (2009).
quite intriguing [7] Na Liu, T. Weiss, M. Mesch, L. Langguth, U.
and can be Eigenthaler, M. Hirschner, C. Sönnichsen, and
explained by a H. Giessen: Planar metamaterial analog of
coupled oscillator electromagnetically induced transparency for
model. plasmonic sensing, Nano Lett. 10, 1103 (2010).
Our three- [8] Na Liu, Hui Liu, Shining Zhu, and Harald Giessen:
dimensional Stereometamaterials, Nature Photonics 3, 157
stacking approach (2009).
allows also for the [9] H. Giessen and M. Lippitz: Directing light
fabrication of 3D emission from quantum dots, Science 329, 910
26 nanoantennas, (2010).](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-26-320.jpg)
![nanoresearch
Spin transfer RF nano-oscillators
for wireless communications and
microwave assisted magnetic recording
U. Ebels1, M. Quinsat1,2, D. Gusakova1, J. F. oscillations combined with the giant or tunnel
Sierra1, JP Michel1,2 , D. Houssameddine1, magnetoresistance of the stack generate
B. Delaet2, M.-C. Cyrille2, L. D. Buda- oscillations of the voltage across the stack
Prejbeanu1,3, B. Dieny1 at GHz frequencies. Moreover the frequency
1 SPINTEC, UMR(8191) CEA / CNRS / UJF / varies as a function of the current density
Grenoble INP ; INAC, 17 rue des Martyrs, 38054 flowing through the stack.
Grenoble Cedex, France.
2 CEA-LETI, MINATEC, 17 Rue des Martyrs, 38054 This phenomenon can be used to design
Grenoble, France. frequency tunable RF oscillators which could
3 Grenoble INP 46, Avenue Félix Viallet, 38031
, be quite useful in a number of devices such
Grenoble Cedex 1, France. as RF STT oscillators (STO) for wireless
communications, or as microwave generators
Slonczewski [1] and Berger [2] predicted in to assist the writing by microwaves in
1996 that a spin-polarized current flowing magnetic recording technology or as
through a magnetic nanostructure exerts magnetic field sensors taking advantage of
a torque on its magnetization due to the the shift of frequency induced by an applied
exchange interaction between the spin magnetic field.
of the conduction electrons and the spin
In this paper, our R&D efforts on STT RF
of the electrons responsible for the local
oscillators as well as current trends in this field
magnetization. This torque is called spin
are described.
transfer torque (STT). The possibility to
use the STT to switch the magnetization A significant effort has been focused on a
of a magnetic nanostructure was first particular configuration of STT RF oscillators
experimentally observed in metallic spin- in which an out-of-plane magnetized polarizer
valve nanopillars [3] and later in magnetic is used to inject out-of-plane spin polarized
tunnel junctions [4]. The spin torque acts as electrons into an in-plane magnetized
a damping or antidamping term and can free layer [5]. Indeed, it was shown that
induce very peculiar magnetization dynamics. this configuration is particularly interesting
Of particular interest is when an applied since it allows generating large angle
field and the spin transfer torque (STT) have precessional motion thereby maximizing
competing influence on the magnetization of the magnetoresistance signal associated
the free layer of a spin-valve or of a magnetic with this motion5. In this configuration, the
tunnel junction, for instance the field favoring frequency varies almost linearly with current
parallel alignment between the magnetization up to a maximum value where it saturates
of the free layer magnetization and that of because of micromagnetic distorsion of the
the reference layer whereas the STT favors magnetization.
antiparallel alignment. In such situations, the Two important characteristics must be
magnetization of the free layer is driven into carefully addressed in such oscillators before
steady state oscillations. The magnetization being able to use them in RF devices for
continuously pumps energy into the spin wireless communications. One is the output
current to compensate the dissipation due power, the other is the excitation linewidth and
to Gilbert damping. These steady state associated phase noise. 27](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-27-320.jpg)
![nanomagma
low for practical applications (300MHz-1GHz),
these results demonstrate the possibility to
increase the output powers to acceptable
value for this type of applications thanks to the
use of magnetic tunnel junctions.
Concerning the linewidth and phase noise,
several studies have aimed at understanding
the cause of the linewidth in STO oscillators
in order to try increasing the coherence of the
magnetization dynamics and thereby minimize
the excitation linewidth and phase noise.
Frequency and time-domain characterizations
were performed. As an example, Figure 2
shows time domain measurements performed
on MgO magnetic tunnel junction (MTJ) pillars
[7]. The RF voltage was measured between
the top and bottom electrodes of the MTJ
while a DC current I flows through the pillar.
Figure 2 clearly shows that the STT induced
magnetic excitations start above a current
threshold. However, the excitations first appear
in bursts (region 2). As the current is further
increased, the excitations become more and
more persistent.
Fig. 1 > STT oscillator with perpendicular
polarizer and in-plane free layer. A fixed in-
plane magnetized reference layer is added to
produce a magnetoresistance between this
reference layer and the precessing free layer.
Typical spectra obtained when measuring the
RF voltage between top and bottom electrodes.
The oscillator pillar has typical diameter between
150nm and 50nm.
Fig. 2 > Time domain measurements of the RF
By using magnetic tunnel junctions, the voltage induced by STT excitations in a MgO
output power could be increased by 2 orders based MTJ submitted to a DC current. Left: Power
of magnitude thanks to the higher impedance of excitations versus DC current amplitude. Right:
of these systems [6,7]. Recently, output real time voltage measurements for three different
powers of the order of 1 μV were reported values of the DC current flowing through the MTJ.
in STT oscillator based on magnetic tunnel
junction and exploiting a vortex configuration By performing a Fourier transform over a
of magnetization [6]. Although the frequency sliding time window of 10ns of the RF voltage
28 associated with these vortex based STO is too associated with these steady state excitations,](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-28-320.jpg)
![nanomagma
Another area of increasing interest concerning Rodmacq, I. Firastrau, F. Ponthenier, M. Brunet,
STO is the one of magnetic field sensors. The C. Thirion, J. P. Michel, L. D. Buda-Prejbeanu,
basic idea is to use the dependence of the M. C. Cyrille, O. Redon, and B. Dieny, Nature
oscillation frequency of STO on the applied Materials 6, 447 (2007).
field to measure the amplitude of the applied [6] A.Dussaux, B.Georges, J.Grollier, V.Cros,
field [8]. Figure 5 illustrates the variation f(H) A.V.Khvalkovskiy, A.Fukushima, M.Konoto,
in a spin-valve structure traversed by a DC
H.Kubota, K.Yakushiji , S.Yuasa, etal., Nature
current. The magnetization of the soft layer
Communications 1, 1(2010), ISSN2041-1723.
is driven into steady state oscillations. The
giant magnetoresistance of the stack then [7] D. Houssameddine, U. Ebels, B. Dieny, K.
produces an oscillatory voltage between Garello, J.-P. Michel, B. Delaet, B. Viala, M.-C.
top and bottom electrodes [9]. The shift of Cyrille, J. A. Katine and D. Mauri, Phys. Rev.
frequency versus applied field can be quite Lett. 102, 257202 (2009).
steep, as large as 180GHz/T [10]. With an [8] Sato et al, US7 471 491 B2 (2008).
appropriate frequency modulation detection [9] P.M.Braganca, B.A.Gurney, J.A.Katine, S.Maat,
scheme, this approach could allow the J.R.Childress, Nanotechnology 21 (2010)
realization of very small magnetic field sensors 235202.
sub-30nm*30nm which could replace TMR
[10] N.Stutzke, S.L.Burkett, S.E.Russek, Appl.
sensors in read heads of HDD.
Phys.Lett.82 (2003) 91.
[11] K.Mizushima,K.Kudo,T.Nagasawa,and R.Sato,
Journ.Appl.Phys. 107,063904(2010).
Fig. 5 > Shift of frequency versus applied field
measured in a spin-valve based STO. From Ref.[9].
References
[1] Slonczewski, J., “Currents and torques in
metallic magnetic multilayers”, J.Magn.Magn.
Mater.159, L1 (1996).
[2] Berger, L., Phys.Rev.B 54, 9353 (1996).
[3] Katine, J.A., Albert, F.J ., Buhrman, R.A., Myers,
E.B., and Ralph, D.C, Phys.Rev.Lett.84, 3149
(2000).
[4] Y.Huai et al, Appl.Phys.Lett.84 (2004), 3118.
30 [5] D. Houssameddine, U. Ebels, B. Delaët, B.](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-30-320.jpg)
![nanoresearch
Characterization of an electrostatically
actuated Pd coated MEMS resonators
J. Henriksson1, J. Arcamone2, G. Villanueva1, plate capacitance as well as the capacitance
J. Brugger1 between the electric paths on the chip and
1 Microsystems Laboratory, EPFL, Lausanne, CH- wires connecting onto the chip (see Figure 2).
1015, Switzerland.
2 CEA, LETI, MINATEC, F-38054 Grenoble, France.
jonas.henriksson@epfl.ch
Introduction
Electrostatically actuated MEMS resonators Fig. 2 > The electrical equivalent of the MEMS
feature CMOS integrability, ultra-low power device.
consumption and stable readout. To utilize
these properties for gas sensing, we The sensing principle is based on the fact
fabricated a doubly clamped free-standing that Pd expands in the presence of H2. This
beam of amorphous silicon (see Figure 1). A property has been used for H2 sensing in
functionalizing layer of palladium is patterned different configurations, such as cantilevers
to cover the beam. The Pd also serves as [3-5], chemo-mechanical switches [6, 7] and
electrical path on the beam. Below the beam, discontinuous films that form new conductive
a bottom electrode has been patterned by paths as the grains expand [8, 9].
lift-off.
Micro- and nano-electromechanical
resonators excel in the field of sensing
applications, showing high sensitivity, low
noise susceptibility and precise readout. They
are widely used in science and technology for
detection of mass [10], temperature [11] and
gas pressure [12].
With our design, the idea is to induce a
change of stress on the beam through H2
exposure, thus benefiting from the strong
natural phenomenon of H2 induced Pd
Fig. 1 > Schematic illustrating actuation, readout
expansion.
and sensing principle of the device [1]. In this report we investigate a few strategies
to improve the quality of the readout signal
in order to enhance the sensitivity and
By applying an alternating voltage between
robustness of the device.
the beam and the bottom electrode, the beam
is brought into resonance. The resonance
Background of experiments
frequency is measured by monitoring the
transmitted signal between the beam and the The electrostatic interaction between the
bottom electrode. The electrical equivalent beam and the bottom electrode can be
of this configuration is a RLC branch [2], approximately described as a parallel plate
representing the mechanical motion, in parallel capacitor interaction. In this case, the
with a capacitance, representing the parallel electrostatic force is given by 31](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-31-320.jpg)
![nanoICT
the response increases as Vdc is increased. In vacuum clearly responds stronger. At Vdc=30
addition the resonance frequency decreases V, the beam in atmospheric pressure shows
due to spring-softening. a 1.5 μV magnitude change at resonance
whereas the beam in vacuum changes by
5 μV.
Fig. 4 > Differential measurement in atmospheric
pressure, illustrating the increase in magnitude
and spring-softening effect as Vdc is increased.
The spring softening is approximated as
, where . Given
that the resonating motion originates from
the product of Vac and Vdc, the resonance
frequency is expected to decrease linearly with
respect to Vdc. Based on a linear relationship
( ), the following coefficients
were determined
∆Vdc [V] ∆fres [kHz] a [kHz/V]
20-15=5 0 0
25-20=5 -12.5 2.5
30-25=5 -12.5 2.5
Fig. 5 > Same device a) measurement in vacuum
35-30=5 -12.5 2.5 and b) measured in atmospheric pressure.
The change of resonance frequency is thus
2f measurements
approximately linear, but changes are very
close to the step size frequency increase, The result of a 2f measurement is shown in
which limits the precision. Figure 6. The most important difference is
that the background has been attenuated by
Atmospheric pressure vs. vacuum close to 2 orders of magnitude. The signal-
Figure 5a illustrates measurements made to-background ratio is much more favorable.
under vacuum. Figure 5b illustrates However, the output signal is also attenuated.
measurements on the very same device but As noise is more visible, we find that the
under atmospheric pressure. We find that for signal-to-noise ratio is worse as compared to
equivalent electrostatic forces, the beam in measurements at f. 33](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-33-320.jpg)
![nanoICT
Systems Ii-Express Briefs, 2007. 54(5): p. 377-
381.
[3] Baselt, D.R., et al., Design and performance
of a microcantilever-based hydrogen sensor.
Sensors and Actuators B-Chemical, 2003.
88(2): p. 120-131.
[4] Hu, Z.Y., T. Thundat, and R.J. Warmack,
Investigation of adsorption and absorption-
induced stresses using microcantilever
sensors. JOURNAL OF APPLIED PHYSICS,
2001. 90(1): p. 427-431.
[5] Okuyama, S., et al., Hydrogen Gas Sensing
Using a Pd-Coated Cantilever. Jpn. J. Appl.
Phys., 2000. 39: p. 3584-3590.
Fig. 6 > 2f measurement. [6] Kiefer, T., et al., A single nanotrench in a
palladium microwire for hydrogen detection.
Nanotechnology, 2008. 19(12).
Conclusions
[7] Kiefer, T., et al., Large arrays of chemo-
We have tested several different methods to mechanical nanoswitches for ultralow-power
improve the signal quality of an electrostatically hydrogen sensing. Journal of Micromechanics
actuated MEMS device. We report that the and Microengineering, 2010. 20(10).
differential measurement configuration is an [8] Xu, T., et al., Self-assembled monolayer-
effective way to improve the readout signal. A enhanced hydrogen sensing with ultrathin
weak spring-softening effect was observed. palladium films. APPLIED PHYSICS LETTERS,
Comparisons between measurements in 2005. 86(20).
vacuum and in atmospheric pressure showed [9] Kiefer, T., et al., The transition in hydrogen
that media-related damping (squeeze sensing behavior in noncontinuous palladium
damping) is dominant, but the damping films. APPLIED PHYSICS LETTERS, 2010.
caused by the clamping is also considerable. 97(12).
The 2f measurement readout scheme [10] Naik, A.K., et al., Towards single-molecule
attenuated the background very strongly, as nanomechanical mass spectrometry. Nature
expected, but it is not clear if this is helpful as Nanotechnology, 2009. 4(7): p. 445-450.
the signal-to-noise ratio is also decreased. [11] Pandey, A.K., et al., Performance of an AuPd
micromechanical resonator as a temperature
As an outlook, the next step in improving the sensor. APPLIED PHYSICS LETTERS, 2010.
readout of the sensor would be to design and 96(20).
fabricate a device with three terminals, so that [12] Huang, X.M.H., et al., Nanomechanical
driving and reading can be separated more hydrogen sensing. APPLIED PHYSICS
efficiently. LETTERS, 2005. 86(14).
References
[1] Henriksson, J., L. G. Villanueva Torrijo, and J.
Brugger. Ultra-low power palladium-coated
MEMS resonators for hydrogen detection
under ambient conditions. in Transducers ‘11.
2011. Beijing: IEEE.
[2] Arcamone, J., et al., A compact and low-
power CMOS circuit for fully integrated NEMS
34 resonators. Ieee Transactions on Circuits and](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-34-320.jpg)
![nanoICT
for this sample.
All the NWs
measured from
this samples emits
at this energy value
indicating highly
Fig. 2 > Sketch of CS NW in the top part indicating homogeneous
three parts where PL spectra are acquired and NWs. In Fig. 3(b) full width half maximum
expected emissions related to core and shell part. (FWHM) for these spectra is plotted.
Decrease in FWHM with increasing As
content in the core of these NWs gives a
2008), (Titova, 2007)]. The splitting of the direct evidence of improvement in the crystal
peaks, which is the reasons for the emission quality of these NWs.
peaks at various energy values, indicate the
presence of randomly varying ZB and WZ
sections thicknesses within the InP emission a
energy range as reported by Jancu and co
authors [(Jancu, 2010), (Pemasiri, 2009)].
Other NWs studied from the same sample
(not shown) emits in the similar wavelengths
range and spectra were always composed
of several peaks with different weights within
the energy range related to InP. Red color in
Fig. 3(a), represent the μ - PL spectra from
the sample grown with 0.38 As/P ratio. This
spectrum is comprised of a broad peak
centered at 1.447 eV along with shoulders
at higher energy corresponding to emission
from InP. PL spectra acquired on NWs from
this sample emits within this energy range
but contains many peaks. b
μ – PL spectra obtained for the NWs grown
with 0.51 As/P ratio are represented by green
color in fig 3(a). Since these NWs show fewer
stacking faults compared to previous samples
(not shown here) main peak corresponding
to InP is at 1.442 eV with a shoulder at 1.48
eV. Several NWs measured from this sample
exhibits less inhomogeneity. Recorded PL
spectra show some sharp lines distributed
between this energy ranges suggest the
insertions of certain number of monolayer
of WZ and ZB segments for all measured
NWs from this sample. Finally, sample with
Fig. 3 > μ - PL spectra acquired on single NWs for
highest As content and no stacking faults each sample. Broader PL emission (black line) for
as shown in Fig. 1(b) is measured. Blue line lowest As/P content indicates the phase mixing,
in Fig. 3(a) represent PL spectra from a NW whereas decreasing FWHM of the PL peaks with
from the sample with highest As content. increasing As content indicate an improvement in
An emission peak, at 1.43 eV, is recorded the crystal quality as presented in (b). 37](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-37-320.jpg)
![nanoICT
A red shift for the emission related to InP
shell is observed Fig. 3(a). There are two
possible explanations for the red shift in
the emission energy related to InP, with
increasing As content. Firstly, it could be
due to the two formation of InAsP shell
with parasitic As present in the chamber at
the time of the shell growth. Short growth
time and low growth temperature for shell
enhance As incorporation. The trend of this
red shift fit well with increasing As content
in the core which implies increased amount
Fig. 5 > μBand gap energy of ternary InAsxP1-x
of residual As in the chamber at the time
compound plotted against x ranging between 0
of shell growth for respective sample. and 1. Red dots with a shift close to difference of
Secondly, this shift could be related to the ZB and WZ InP band gap energy representing the
stain present in the InP shell around InAsP emission for shell above 1.42 eV and corresponds
core and increased As in the core result to core emission between 1.38 eV and 1.22 eV.
increasing strain in InP shell. Since the
diameter of the inner core is ~10 nm and
the shell around these NWs is 15 – 20 nm Conclusion
thick, such red shift only due to strain is not In conclusion we have reported the growth
envisioned. of stacking fault free InAsP/InP CS NWs
In Fig. 5, band gap energy for ZB InAs1-xPx confirmed by their μ – PL analysis. Increase
ternary is plotted in black for x between in As content in the core part results in
0 and 1 [11]. Peak PL emission energies suppression crystal imperfection and a
corresponding to InP shell plotted in blue shift in PL emission energy related to
Fig. 3 are inserted. Band gap 1.49 eV is the core emission indicate a systematically
considered for the WZ InP (Jancu, 2010), increasing As content in the core part.
(Pemasiri, 2009). InP emission from sample
with As/P ratio of 0.60 which exhibits peak
at 1.43eV is inserted with 60 meV difference References
from the WZ InP and it results with ~ 5% [1] N. Akopian. Et al, Nano Lett., 10. 1198 (2010).
content in the shell. This value, for the [2] E. D. Minot, et al, Nano Lett., 7. 367 (2007).
parasitic incorporation of As, is quite high [3] M. Tchernecheva et al., Nano Lett. 7 (2007)
so strain should have some effect for this
1500.
sample. For the samples grown with lower
[4] F. Jabeen et al., Appl. Phys. Lett. 93 (2008)
As/P ratio the shift in the emission peak
083117.
with respect to the WZ InP gives around
2% As incorporation in the shell. [5] G. Wuet al., Chem. Phys. Lett., 378 (2003) 368.
[6] F. Jabeen et al., to be published, (2011).
An increasing As incorporation trend is
[7] J.-M. Jancu et al., Appl. Phys. Lett. 97 (2010)
observed and the PL emission give 11%,
041910.
13% and 21% incorporation of As in the
ternary core for the samples grown with [8] K. Pemasiri et al., Nano Lett. 9 (2) (2009) 648.
0.38, 0.50 and 0.60 As/P ratio respectively. [9] J. Bao et al., Nano Lett. 8 (3) (2008) 836.
For the sample grown with 0.50 As/P ratio [10] L. V. Titova et al., Nano LettL 7 (11), (2007) 3383.
nominal As in the core measured by EDS [11] Brigham Young University (BYU), “Energy gap
is around 8%. So this additional shift could in III-V ternary semiconductors”, http://www.
also be related to the strain. cleanroom.byu.edu/EW_ternary.phtml. 39](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-39-320.jpg)
![nanoresearch
Brillouin light scattering measurements
in crystallographically tuned thin Co-films
O. Idigoras1, B. Obry2, B. Hillebrands2 and A. depositing different amounts of Si-oxide, we
Berger1 have obtained a set of samples with different
1 CICnanoGUNE Consolider, Tolosa Hiribidea 76, degree of crystallographic order. Moreover,
E-20018 Donostia-San Sebastian, Spain. we observe an anomalous magnetic
2 Fachbereich Physik and Landesforschungzentrum reversal near the HA in these samples above
OPTIMAS, Technische Universität Kaiserslautern, a threshold level of crystallographic disorder,
Erwin-Schrödinger-Straße 56, D-67663
while behavior in all other field orientations is
Kaiserslautern, Germany.
barely affected. This anomaly arises from a
competition of the misalignment anisotropy
We report an experimental study of spin axes with the exchange energy and it
waves in thin Co-films with in-plane uniaxial gives rise to an unusually high remanent
symmetry, which were measured by magnetization and coercive field along the
means of the Brillouin light scattering (BLS) hard axis [1].
technique. In particular, we investigated the
effect of the previously discovered hard axis The Brillouin light scattering (BLS) technique
(HA) anomalous magnetization state that [2] is a powerful method to analyze spin
occurs during magnetization reversal in waves and related magnetic properties.
partially disordered films. This work has been Primarily, it allows for the analysis of spin
performed in Prof. Dr. Burkard Hillebrands’ waves energies by means of the frequency
group where all members of team were shift of inelastically scattered photons from
really nice in supporting all the activities. a magnetic sample. Photons interact hereby
Furthermore, all have been possible thanks with spin waves, so that a spin wave is
to Phantoms Foundation’s Nano-ICT project created (Stokes process) and the photon
launches exchange visit fellowship. looses energy correspondingly or a spin
wave is annihilated (anti-Stokes process)
whereupon the photon gains that energy.
Introduction Experimentally, the frequency shift of the
In previous work we have demonstrated scattered light is detected to measure the
that is possible to tune the degree of corresponding spin wave energy. In this
crystallographic order in Co-films by partial work we have analyzed Stokes scattering
interruption of epitaxy in a well-defined processes only.
and reproducible manner [1]. Hereby, we In general, different spin wave modes
utilized an optimized growth sequence to can be present in a thin film [3]. The main
achieve good epitaxy and a high degree of distinction arises from the type of interaction
crystallographic order as a starting point. that dominates them, which is either the
Specifically, we produced epitaxial Co-films exchange or the dipolar interaction. There
with an in-plane hcp c-axis by growing the exist different dipolar dominated modes,
sequence Ag 75 nm/Cr 50 nm/Co 30 nm/ which can be bulk or surface spin wave
SiO2 10 nm onto hydrofluoric etched Si (110) modes. The surface mode, called as
substrates. For the tunable disturbance of magnetostatic surface mode (MSSM) or
the growth sequence, we introduced an Damon-Eshbach (DE) mode are excited
ultrathin Si-oxide layer of defined thickness in when the magnetization M and the wave
the order of a single monolayer on top of the vector lie both in the film plane and are
Si-substrate prior to the Ag-film growth. By perpendicular to each other. Even in this case 41](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-41-320.jpg)
![nanoICT
when the magnetization and wave vector about the spin wave damping and thus on
are perpendicular bulk modes can be the magnetization homogeneity. The peak
excited, in thin films it is not possible to at -27 GHz corresponds to the dipolar spin
distinguish between both modes and only wave mode (MSSW/DE), while the other two
MSSM/DE mode are considered. On the peaks corresponds to exchange spin wave
other hand, if both are in-plane and parallel, modes (PSSW). Measurements as the one
magnetostatic backward volume modes shown in figure 1 but for multiple applied field
(MSBVM) are excited. For the same in- strengths are shown in figure 2 as a color
plane wave vector the DE/MSSM mode map. Figure 2(a) shows the data collected
has higher energy and thus frequency than when the external field is applied along the
the MSBVM mode. The exchange type EA, and fig. 2(b) when the field is along the
perpendicular standing spin waves (PSSW) HA. The red color indicates the maximum
are formed by a superposition of two spin intensity, i.e. the spin wave positions, while
waves that are propagating in opposite the blue color indicates the minimum
directions perpendicular to the film surface. intensity, i.e. noise level. In both figures two
In this work we have analyzed all three spin spin wave modes appear, one belonging to
wave modes, DE/MSSM, MSBVM and the MSSW/DE at lower absolute values of
PSSW. frequency and one representing the PSSW
at higher absolute frequency values. As
The experimental setup that was utilized in
expected in the EA (fig. 2(a)), both spin wave
this study is equipped with a tandem Fabry- frequency vs. field curves have the same
Pérot interferometer [4]. Moreover, this type of behavior. When the field is applied
setup is able to perform an automated in- along the HA (fig. 2(b)), however, the dipolar
plane rotation of the sample. In this work, we type spin wave shows a more pronounced
used backward scattering geometry, and all frequency shift for small applied field than the
measurements have been done with an in- PSSW mode. The origin of this frequency shift
plane applied field orientation perpendicular is the in-plane rotation of the magnetization
to the wave vector. into the direction of the anisotropy axis as
the field strength decreases, causing the
spin wave to change its character from the
Influence of the HA anomaly on spin
DE mode to MSBV mode, which has a lower
waves spectra
frequency than the DE mode. This frequency
In order to analyze the effect that the HA shift for fields applied along the HA has been
anomaly has on the spin wave spectra previously reported in several works [5].
we have measured the dependence of
the spin wave frequencies as a function of
the applied field strength for (i) an epitaxial
sample, in which the anomaly is not present,
and (ii) in a sample with a 0.132 nm thick
Si-oxide underlayer, for which the anomaly
does occur. One example of such spin wave
spectra for the epitaxial sample is presented
in figure 1. Apart from the elastically
scattered light peak at zero frequency, three
peaks are clearly visible that correspond to
inelastically scattered light from three spin
wave modes. While the position of these
peaks gives the frequency of the spin wave, Fig. 1 > Spin wave spectrum for an epitaxial Co
42 the widths of the peaks contain information (1010)-film sample.](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-42-320.jpg)
![nanoICT
field orientations around the nominal hard
axis [1].
Fig. 2 > (a) and (b) show the spin wave frequency
dependence from an externally applied field
along the EA and HA, respectively, for an epitaxial
Co (1010)-film sample .
A similar set of measurements for a sample,
in which the HA anomaly is present (the
sample with Si-oxide underlayer of 0.132 nm
thickness), are shown in figure 3. Specifically,
this figure shows spin wave frequency vs.
Fig. 3 > (a), (b) and (c) show the spin wave
field dependence for the applied field along
frequency dependence from an externally applied
the EA (fig. 3(a)), along the HA (fig. 3(b)), and field along the EA, HA and 2° away from the HA,
for a field direction 2° away from the HA (fig. respectively for a uniaxial, but slightly disordered
3(c)). Although the dipolar and the exchange sample.
type spin waves are close in frequency
for measurements along the EA, so that it
If the field is applied 2° away from the HA,
is difficult to clearly see their separation,
a conventional HA spin wave behavior
both spin waves visibly follow the same
reemerges, with the dipolar type spin wave
frequency behavior as a function of the
frequency shifting towards lower frequencies
applied field, just as in the case of epitaxial
upon decreasing the applied field strength.
sample. However, the expected downward
frequency shift for the dipolar type spin For this sample, the saturation magnetization
wave upon reducing the externally applied is obtained at smaller external field strengths
field along the HA does not occur here (fig. than in the case of the epitaxial sample, so
3(b)). Only a broadening is observed that that the position shift of this mode is limited
can arise due to an inhomogeneous sample to a narrower applied field range.
magnetization. Such an inhomogeneous For the same sample, we have also
magnetization state was indeed observed performed applied field angle β dependent
earlier on the same sample by means of measurements in the vicinity of the HA
Kerr effect microscopy in a narrow range of (β=90°) in remanence. Figure 4(a) shows 43](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-43-320.jpg)
![nanoICT
spin wave spectra vs. β in remanence after I acknowledges Phantoms Foundation
saturating the sample in 1500 Oe for each for Nano-ICT project launches exchange
angle and figure 4(b) shows at the same visit fellowship, as well as, all members
type of measurment in remanence, but of Prof. Dr. Burkard Hillebrands’ group
after saturating the sample only once along at Fachbereich Physik at Technische
the EA. In the first measurement (fig. 4(a)) Universität of Kaiserslautern and specially
one can see that due to the anomalous Prof. Dr. Burkard Hillebrands for giving me
magnetization reversal in the HA, β=90°, the opportunity to work in his group and
the dipolar type spin wave is shifted towards Björn Obry with whom I have performed
all the measurements and helps me
higher frequencies. One can also see that
with everything. I also thank the Basque
a misalignment of only ±1° away from the
Government for fellowships No. BFI09.284.
HA already causes an almost complete
suppression of this anomalous behavior and
the reappearance of the typical dipolar spin References
wave at low frequencies of about 7 GHz. [1] O. Idigoras, A.K. Suszka, P. Vavassori, P.
In the second measurement series of the Landeros, J.M. Porro and A. Berger, submitted
same sample after a single EA saturarion to Phys. Rev. B.
(fig. 4(b)), no anomaly is visible, as expected [2] O. Gaier, 2009. A study of exchange interaction,
because in this case we only measure magnetic anisotropies, and ion beam induced
the EA projections of the same uniform effects in thin films on Co2-based Heusler
magnetization state along the applied field compounds. Thesis, (PhD). Technischen
direction. Universität Kaiserslautern.
[3] B. Hillebrands, Brillouin light scattering from
layered magnetic structures, in M. Cardona,
Güntherodt (Editors), Light Scattering in Solids
VII, vol. 75 of Topics in Applied Physics, Springer
Verlag, Berlin Heidelberg (2000).
[4] J.R. Sandercock, Opt. Comm. 2, 73 (1970). B.
Hillebrands, Rev. Scien. Instr. 70, 1589 (1999).
[5] R. Scheurer, R. Allenspach, P. Xhonneux and
E. Courtens, Phys. Rev. B 48, 9890 (1993). M.
Grimsditch, E.E. Fullerton and R.L. Stamps,
Phys. Rev. B 56, 2617 (1997).
Fig. 4 > Spin wave frequency as a function of
the applied angle in remanence for a slightly
disordered sample, after (a) prior saturation at
every angle and (b) after prior saturation along
44 the EA.](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-44-320.jpg)
![nanoresearch
Temperature distribution of heated
membranes for stencil lithography
application
Shenqi Xiea, Damien Ducatteaub, Bernard
Legrandb, Veronica Savua, Lionel Buchaillotb a
and Juergen Bruggera
a Microsystems Laboratory (LMIS-1), Ecole
Polytechnique Fédérale de Lausanne, Switzerland.
b Institut d’Electronique, de Microélectronique et
de Nanotechnologie, UMR CNRS 8520, IEMN,
Avenue Poincaré, B.P 69, 59652 Villeneuve d’Ascq
.
Cedex, France. b
1. Introduction
Stencil lithography (SL) is a resistless
lithography method for surface patterning
with sub-micron resolution. As the most c d
conventional application of SL, thin-film
deposition has become a reliable micro/
nano-patterning process [1, 2]. However,
the useful life time of the stencil during one
pump-down is limited by the clogging of the
aperture due to the deposited material [3].
We recently developed a novel approach to
potentially prevent and eventually eliminate
clogging by locally heating up the stencil
during metal deposition, minimizing thus
materials’ condensation on the membrane
Fig. 1 > Schematics of (a) frontside and (b)
[4]. The heatable stencil has Pt microhotplates backside of the heated stencils. (c) Optical image
embedded in two layers of SiN thin film, with of the device and (d) SEM image of the apertures
stencil apertures in between the coils, as in between the coils.
shown in figure 1.
In our previous experiments, the thickness between temperature and condensation rate,
of condensed metal film is correlated to the which can be translated into clogging rate.
temperature distribution on the membrane. In addition, the temperature coefficient of
The area with least material condensation is resistance (TCR) can be extracted from the
in the center of the heated membrane, where temperature mapping under certain input
has the highest temperature. As temperature power in ambient conditions. As the heated
drops rapidly from the center to the border stencil will be placed in the vacuum chamber
of the membrane, metal starts accumulating of an E-beam evaporator, the temperature
very quickly during deposition, which of course of the membrane can only be calculated
increases clogging rate. Therefore, it is critical from the variation of the resistance of the
to know the precise temperature distribution microhotplate based on the measured TCR
on the membrane in order to study the relation in order to monitor the process. By using the 45](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-45-320.jpg)
![nanoICT
relatively uniform temperature distribution in coils with different resistance were measured,
the centre of the membrane provides a stable which provides important information for
thermal environment for the stencil apertures. optimizing the designs in future generations.
The calibrated temperature coefficient
of resistance (TCR) is extremely useful in
monitoring and controlling the whole process.
Temperature profiles were also studied to
correlate with the clogging rate in different
areas. Extra Pt thin layers were coated on
the membrane for providing a more uniform
surface in terms of emissivity, leading to a
more accurate temperature measurement.
The temperature distribution provides us
great feasibility of studying the dependence of
temperature on clogging rate. The achieved
results from this exchange program would be
combined with other results to be considered
to submit to MEMS2012, following with other
journals.
References
[1] M. A. F. van den Boogaart, et al., “Corrugated
membranes for improved pattern definition with
micro/nanostencil lithography”, Sensors and
Actuators A, vol. 130-131, p. 568-574, 2006.
[2] V. Savu, et al., “Dynamic stencil lithography on
full wafer scale”, Journal of Vacuum Science
and Technology B, 26(6), 2008.
Fig. 3 > (a) The temperature profile of the heated [3] M. Lishchynska, et al., “Predicting Mask
stencil with and without Pt coating on the frontside Distortion, Clogging and Pattern Transfer
of the membrane. (b) The average temperature of for Stencil Lithography”, Microelectronic
the membrane versus the applied power in ambient
Engineering, 84 (2007) 42-53.
conditions.
[4] S. Xie, V. Savu, J. Brugger, “Heated
membranes prevent clogging of apertures in
nanostencil lithography”, The 16th International
Another information can be derived from
Conference on Solid-State Sensors, Actuators
the measurement is the temperature versus
and Microsystems (Transducers’11), Beijing,
power consumption, as shown in figure 3(b).
China.
The power consumption is not critical in our
preliminary stage of experiments, but it has to
be considered in future development if more
heated stencils are powered simultaneously.
3. Conclusions
We have successfully carried out the thermal
measurement on heated stencils by using the
IR microscope in IEMN. Various designs of the 47](https://image.slidesharecdn.com/enanonewsletter23finalversion-120103055900-phpapp01/85/Enano-newsletter-issue-23-47-320.jpg)
Enano newsletter issue 23
- 1. No. 23 /// October 2011 www.phantomsnet.net Nanostructured Active Magneto-Plasmonic Materials: Research Overview Characterization of an electrostatically actuated Pd coated MEMS resonators Optical analysis (study) of InAsP/InP core shell nanowires Brillouin light scattering measurements in crystallographically tuned thin Co-fi lms Temperature distribution of heated membranes for stencil lithography application
- 3. contents 05 > nanoresearch. (nanomagma EU project). Ultrafast acousto-magneto-plasmonics in hybrid metal-ferromagnet multilayer structures /// V. Temnov 10 > nanoresearch. (nanomagma EU project). Label-free ligand fishing in human plasma using surface plasmon resonance and mass spectrometry imaging /// E. Ly- Morin, W. Boireau, P. Ducouroy, S. Bellon and C. Frydman 14 > nanoresearch. (nanomagma EU project). Light localization on a gold nanodisk array probed by near-field optics /// L. Lalouat, L. Aigouy, P. Prieto, A. Vitrey, J. Anguita, A. Cebollada, M.U. González and A. García-Martín 17 > nanoresearch. (nanomagma EU project). Controlling fluorescence resonant energy transfer with a magneto-optical nanoantenna /// R. Vincent and R. Carminati 22 > nanoresearch. (nanomagma EU project). Internal electromagnetic field distribution and magneto-optical activity of metal and metal-dielectric magnetoplasmonic nanodisks /// D. Meneses-Rodríguez, E. Ferreiro-Vila, J. C. Banthí, P. Prieto, J. Anguita, A. García- Martín, M. U. González, J. M. García-Martín, A. Cebollada, and G. Armelles 24 > nanoresearch. (nanomagma EU project). Magneto-Optical properties of nanoparticles /// R. Gómez-Medina, B. García-Cámara, I. Suárez-Lacalle, L. S. Froufe- Pérez, F. González, F. Moreno, M. Nieto-Vesperinas and J. J. Sáenz 26 > nanoresearch. (nanomagma EU project). Three-dimensional optical metamaterials and nanoantennas: Chirality, Coupling, and Sensing /// H. Giessen 27 > nanoresearch. (nanomagma EU project). Spin transfer RF nano-oscillators for wireless communications and microwave assisted magnetic recording /// U. Ebels, M. Quinsat, D. Gusakova, J. F. Sierra, JP Michel, D. Houssameddine, B. Delaet, M.-C. Cyrille, L. D. Buda-Prejbeanu and B. Dieny 31 > nanoresearch. (nanoICT EU project). Characterization of an electrostatically actuated Pd coated MEMS resonators /// J. Henriksson, J. Arcamone, G. Villanueva and J. Brugger 35 > nanoresearch. (nanoICT EU project). Optical analysis (study) of InAsP/InP core shell nanowires /// F. Jabeen, B. Ketterer, G. Patriarche, A. Fontcuberta I Morral and J-C. Harmand 41 > nanoresearch. (nanoICT EU project). Brillouin light scattering measurements in crystallographically tuned thin Co-fi lms /// O. Idigoras, B. Obry, B. Hillebrands and A. Berger 45 > nanoresearch. (nanoICT EU project). Temperature distribution of heated membranes for stencil lithography application /// S. Xie, D. Ducatteau, B. Legrand, V. Savu, L. Buchaillot and J. Brugger editorial information No 23. October 2011. Published by Phantoms Foundation (Spain) editor > Dr. Antonio Correia > antonio@phantomsnet.net assistant editors > Carmen Chacón, Viviana Estêvão, Maite Fernández, Conchi Narros and José Luis Roldán. 1500 copies of this issue have been printed. Full color newsletter available at: www.phantomsnet.net/Foundation/newsletter.php For any question please contact the editor at: antonio@phantomsnet.net editorial board > Adriana Gil (Nanotec S.l., Spain), Christian Joachim (CEMES-CNRS, France), Ron Reifenberger (Purdue University, USA), Stephan Roche (ICN-CIN2, Spain), Juan José Saenz (UAM, Spain), Pedro A. Serena (ICMM-CSIC, Spain), Didier Tonneau (CNRS-CINaM Université de la Méditerranée, France) and Rainer Waser (Research Center Julich, Germany). deadline for manuscript submission depósito legal printing Issue No 25: February 29, 2012. legal deposit Gráficas Issue No 26: April 30, 2012. BI-2194/2011 Valdés, S.L. 03
- 4. dear readers, During the last decades a large effort has been invested in the development of a new discipline devoted to benefit from optical excitations in materials where metals are key element (Plasmonics). We will make an introduction on this topic below, but let’s anticipate that two application areas are sensing and information technologies. In the first case, it is the strong dependence of the plasmon resonance on the environment the factor that is used for the development of applications. In the second, it is the capability to confine the electromagnetic field beyond the diffraction limit when coupling to the plasmon what is put to work. In both cases there is way for improvement, and we have identified an element that can be used in both areas, and in turn make an interesting influence in another field of research: magneto-optics. Magneto-optics is a discipline that has been tied to the information technologies framework from long ago, mainly to endorse active (tunable) capabilities. There will be an introductory section to the topic below. Therefore the main idea behind the NANOMAGMA EU/NMP funded project is to get insight into the interplay between plasmonics and magneto-optics. The project has two main goals; the first is to prepare active magneto-plasmonic materials with tailored properties in the nanoscale and understanding the interactions of the magnetic properties with the plasmonic and optical ones, linked to electric charge oscillations. The second goal is to propose devices for applications that can benefit of this coupling. Since it is expected that the optical properties of these materials can be driven by using a magnetic field, this will allow designing and developing novel magneto-plasmonic devices. These devices will be of use in both areas mentioned above: sensing, i.e. a surface magneto-plasmon resonance (SMPR) sensor tailored on the nanoscale, and information technologies, i.e. non-reciprocal components for photonic integrated circuits based on magneto-plasmonic elements. The following height extended abstracts, presented during the one-day NANOMAGMA Symposium (Bilbao, Spain – April 13, 2011: http://www.imaginenano.com/SCIENCE/ Scienceconferences_PPM2011.php), reflects some of the latest developments on magneto-plasmonics. In 2010 and 2011, the nanoICT project (EU/ICT/FET Coordination Action) launched two calls for exchange visits for PhD students with the following main objectives: 1. To perform joint work or to be trained in the leading European industrial and academic research institutions; 2. To enhance long-term collaborations within the ERA; 3. To generate high-skilled personnel and to facilitate technology transfer; The first outcome report was published in the issue 22 (August 2011) and this edition contains four new articles providing insights in relevant fields for nanoICT. We would like to thank all the authors who contributed to this issue as well as the European Commission for the financial support (projects nanoICT No. 216165 and NANOMAGMA No. FP7-214107-2). 04 > Dr. Antonio Correia Editor - Phantoms Foundation
- 5. nanoresearch Ultrafast acousto-magneto-plasmonics in hybrid metal-ferromagnet multilayer structures Vasily Temnov, MIT Chemistry Department 77 Massachusetts Avenue, 02139 Cambridge, MA. temnov@mit.edu Nanostructured metal surfaces are presently used to effectively couple light to surface plasmons. This technology is also key to on-chip miniaturization of plasmonic sensors. Strong sub-wavelength confinement of optical surface plasmon fields combined with their macroscopic propagation distances exceeding tens of micrometers makes it possible to perform sophisticated nanoplasmonic experiments using conventional far-field optical microscopy in novel hybrid nanostructures. Combining these nano-optical experiments with femtosecond time-resolved spectroscopic pump-probe techniques opens the door to fundamental studies at the nano-scale and ultrafast characterization of nano-optical devices. Here we present a new plasmonic sensor, based on a tilted slit-groove interferometer (Fig. 1) milled by a focused ion beam into a single noble metal film [1] or into a hybrid metal-ferromagnet multilayer structure [2]. Surface plasmons excited at the groove Fig. 1 > (a) Scanning electron microscopy image propagate towards the slit, where they of a slit-groove microinterferometer in a 200 nm interfere with incident light (Fig. 1b). Due thin gold film. The width of the slit is 100 nm, the to the tilt angle the optical transmission groove is 200 nm wide and 100nm deep. (b) The through the slit shows a pronounced entire area of the microinterferometer is illuminated periodic interference pattern (Fig. 1c). A by a spatially coherent laser beam. Surface plasmon small modulation of the complex surface is excited at the groove, propagates towards the slit, plasmon wave vector is accumulated over a where it is converted into free space radiation and long propagation distance between the slit interferes with directly transmitted light. (c) Optical and the groove and results into measurable transmission shows a pronounced interference changes in the contrast and phase shift of pattern along the slit axis, see Ref. [1] for details./ the plasmonic interference pattern [1]. There exist different ways to modulate the wave In hybrid magneto-plasmonic gold-cobalt- vector of surface plasmons using multilayer gold trilayers a few nanometer thin cobalt structures. layer is sandwiched between two gold layers 05
- 6. nanomagma through the (111) gold layer at the speed of sound cS(111)=3450 m/s is converted into a tensile pulse upon reflection from the gold-air interface. The acoustic pulse creates a transient multilayer structure with higher (lower) density of free electrons for the compressive (tensile) acoustic pulses. The wave vector of femtosecond surface plasmon probe pulses propagating along the Fig. 2 > Active magneto-plasmonic interferometry gold-air interface serves as a sensitive probe in tilted slit-groove interferometers patterned in Au/ to the local perturbations of the electron Co/Au multilayer structures. The magnetic field of an electromagnet switches the magnetization in a density within the skin depth skin=13nm cobalt layer and thus changes the wave vector of a induced by the acoustic pulse. Varying the surface plasmon propagating between the slit and pump-probe delay time makes it possible to the groove, see Ref. [2] for details./ monitor the dynamics of acoustic reflection in the plasmonic pump-probe interferogram within the skin depth of surface plasmon: (Fig. 3b) and extract the pump-induced h< skin=13 nm (Fig. 2). A weak external modulation ’+i ’’ of surface dielectric magnetic field can be used to switch the function (Fig. 3c). On top of the slowly magnetization in a ferromagnetic cobalt increasing thermal background due to the layer and thus modify the wave vector of temperature rise at gold-air interface the surface plasmons [2]. Magneto-plasmonic apparent acoustic echo in ’ is observed modulation depth of up to 2% is achieved is indicating the change of surface plasmon this geometry. It can be further increased by wave vector ksp= ’/2| |2. covering the microinterferometer with high- Straightforward mathematical analysis index dielectric material [3]. delivers the exponential shape of the acoustic When combined with time-resolved optical strain pulse with the amplitude of ~10-3, see pump-probe spectroscopy, femtosecond Fig. 4. The exponential shape of acoustic surface plasmon interferometry captures pulses provides the heat penetration depth the dynamics of ultrafast electronic in cobalt within 15 nm, slightly exceeding the excitations and coherent lattice vibrations skin depth of optical pump pulses. A sharp within skin=13nm skin depth in gold with sub-picosecond back front of the acoustic femtosecond time resolution [1]. Using a pulse indicates that the bandwidth of the sapphire/cobalt/gold multilayer structure acoustic phonons exceeds 1 THz. we generate ultrashort acoustic pulses by The 600 fs temporal resolution in our thermal expansion of a cobalt film impulsively experiment is limited by 2nm (RMS) surface heated by femtosecond laser pump pulses roughness (SR) at gold-air interface, which through the sapphire substrate (Fig. 3a). acts as acoustic delay line for ultrashort 06 The compressive acoustic pulse propagates acoustic pulses and prevents observation
- 8. nanomagma (a) of higher frequencies above 1 THz. A good agreement between theory and experiment is obtained by solving linear acoustic wave equation taking into account dispersion of high-frequency phonons and surface roughness. Using higher excitation fluence we were able to generate much larger acoustic pulses with strain amplitudes exceeding 1% (corresponding uniaxial strain of ~2 GPa). These giant acoustic pulses experience significant non-linear reshaping after propagation though 120 nm and 220 nm thin gold films at room temperature [4]. Numerical solutions of the Korteveg-de Vries equation provide an accurate and nearly fit-free approximation of experimental strain pulses obtained at different excitation levels. We could not observe any signatures of ultrasonic attenuation in these measurements suggesting that THz phonons in gold propagate (b) over macroscopic distances most likely exceeding 1 micron at phonon frequency of 1 THz. Technological challenge of fabricating atomically smooth metal interfaces should be tackled in order to safely detect the acoustic phonons with frequencies exceeding 1 THz and thus provide access (c) to the mean free path of acoustic phonons over the entire Brillouin zone. Femtosecond time-resolved pump-probe measurements in such structures may lead to the new type of acoustic spectroscopy in solids with ultrahigh (μeV) spectral resolution. Given the large amplitude, short duration and eventually loss-free propagation of acoustic pulses generated in hybrid gold-cobalt multilayer structures we envision many interesting applications for the nonlinear Fig. 3 > Femtosecond ultrasonics probed with ultrashort surface plasmon pulses. (a) An ultrashort acoustics at the nano-scale. The possibility compressive acoustic pulse is generated by thermal to switch magnetization in the magneto- expansion of 35 nm thin fs-laser heated cobalt plasmonic and magneto-optical devices transducer and propagates through a 120 nm thin by giant acoustic pulses represents just gold film at the speed of sound. The dynamics one example with potentially high impact of acoustic reflection is captured in a plasmonic in the field of ultrafast telecommunication pump-probe interferogram (b) and results into the technology. pronounced modulation of the wave vector for a time-delayed femtosecond surface plasmon probe I am indebted for the invaluable 08 pulse (c)./ contributions to this research project by
- 9. (a) nanomagma my collaboration partners from IMM-CSIC Madrid, TU Berlin, TU Chemnitz, Konstanz University and Massachusetts Institute of Technology. Financial support by the German Research Foundation and European Networks of Excellence ‘Nanomagma’ and ‘Phoremost’ is deeply appreciated. References [1] V.V. Temnov, K.A. Nelson, G. Armelles, A. Cebollada, T. Thomay. A. Leitenstorfer, R. Bratschitsch, (b) Optics Express 17 (2009) 8423. [2] V.V. Temnov, G. Armelles, U. Woggon, D. Guzatov, A. Cebollada, A. Garcia-Martin, J.M. Garcia-Martin, T. Thomay, A. Leitenstorfer, R. Bratschitsch, Nature Photonics 4 (2010) 107. [3] D. Martin-Becerra, J.B. Gonzalez- Diaz, V.V. Temnov, A. Cebollada, G. Armelles, T. Thomay, A. Leitenstorfer, R. Bratschitsch, A. Garcia-Martin, M. Ujue-Gonzalez, Appl. Phys. Lett. 97 (2010) 183114. [4] V.V. Temnov, C. Klieber, K.A. Nelson, T. Thomay, A. Leitenstorfer, D. Makarov, M. Albrecht, R. Bratschitsch (to be published). Fig. 4 > (a) The exponential acoustic pulse generated by thermal expansion of fs-laser- heated cobalt transducer is modified due to the dispersive propagation through a 120 nm thin gold layer. Surface roughness (SR) at gold-air interface smears out high-frequency components in the pulse. (b) The measured acoustic pulse (extracted from curves in Fig.3c) preserves its exponential shape and sub-picosecond acoustic front demonstrating THz bandwidth of acoustic generation. Black horizontal arrows in (a) and (b) indicate the propagation direction of the acoustic pulse, see Ref. [4] for details./ 09
- 10. nanoresearch Label-free ligand fishing in human plasma using surface plasmon resonance and mass spectrometry imaging Elodie Ly-Morin1, Wilfrid Boireau2a, Patrick LAG3, a potential marker of breast cancer, in Ducouroy2b, Sophie Bellon3, Chiraz Frydman1 human plasma. chiraz.frydman@horiba.com Surface Plasmon Resonance (SPR) is an optical 1 HORIBA Scientific, Chilly-Mazarin, France. technique that offers label-free biomolecular 2 Clinical-Innovation / Proteomic Platform (CLIPP) analyses, providing information on kinetic a Institute FEMTO-ST, Université de Franche-Comté, processes (association and dissociation), CNRS, ENSMM, UTBM - F, 25044 Besançon, France. binding affinity, analyte concentration and real b Centre Hospitalier Universitaire Dijon, CGFL. time molecule detection. It has become a 1 rue du Pr Marion, 21000 Dijon, France. powerful tool for the analysis of biomolecular 3 GenOptics – HORIBA Scientific, Orsay, France. events involved in drug development, cancer research, and antibody screening... We present here the exploitation of the The phenomenon of SPR occurs when light powerful approach of Surface Plasmon interacts at the interface between a biochip Resonance imaging and Mass Spectrometry and a liquid medium. It permits to follow coupling for protein fishing in biological modifications of the refractive index (or the fluids such as human plasma at the same reflectivity) in real time. Such modifications are sensitivity. On one hand, multiplex format induced by a biomolecular interaction between SPRi analysis allows direct visualization immobilized ligands (probe molecules) and and thermodynamic analysis of molecular captured analytes (target molecules). SPR avidity, and is advantageously used for monitors theses changes of reflectivity to ligand-fishing of captured bio-molecules characterize the biomolecular events (such on multiple immobilized receptors on a as binding and dissociation) occurring at the SPRi-Biochip surface. On the other hand, surface of the biochip in real time. MALDI mass spectrometry is a powerful Surface Plasmon Resonance imaging (SPRi) tool for identification and characterization technology offered by HORIBA Scientific- of molecules captured on specific surface. GenOptics takes SPR analysis a step further. Therefore, the combination of SPRi and MS The SPRi-Plex II instrument (Figure 1) enables into one concerted procedure, using a unique visualizing the whole biochip surface in real- dedicated surface, is of a great interest for time using a video CCD camera. This design functional and structural analysis at low allows biochips to be prepared in an array femtomole level of bound molecules. format; with each spot corresponding to a specific immobilized ligand. Up to several To reach these goals, particular surface hundreds different molecules can be spotted engineering has been engaged to maintain a using an automated spotter, opening the way high level of antibody grafting and reduce non- to high throughput information for biomolecular specific adsorption. Thus, various chemistries interaction. The multiplexing capabilities of have been tested and validated towards SPRi can meet with any experimental design biological fluids such plasma, keeping in mind without concession. Meanwhile, the sensitivity the capacity of the in situ investigation by MS. of SPRi is not compromised as analyte Finally, signal to noise ratio was magnified concentrations can be detected down to the 10 leading to the characterization of protein nanomolar range.
- 11. nanomagma The open format of the HORIBA Scientific- GenOptics instruments makes MS coupling easier and faster. The possibility of direct MS analysis on the SPRi sensor was recently shown [9]. The SuPRa-MS platform (Surface Plasmon Resonance in arrays coupled with Mass Spectrometry) combines SPRi and MS in a single biochip. The biochip used for SPRi (SPRi-Slide) is directly transferred to the MS instrument. There is no need to neither elute nor re-deposit the bound analyte. The MS enzymatic digestion and the deposition of the Fig. 1 > SPRi-PlexII system MALDI matrix are performed directly on the SPRi-Slide. The latter is then directly placed The applications of SPRi are vast and include on the MS plate holder (Figure 2). for example protein:protein [1], DNA:DNA [2,3], A proof-of-concept study of SPRi-MS peptide:protein [4], polysaccharides:proteins imaging coupling was performed for the [5] or protein:cells [6,7] interactions. The detection of LAG3 recombinant protein in flexibility of the HORIBA Scientific-GenOptics plasma. The solution fraction of this protein is instruments enables complex samples such a potential biomarker for breast cancer [10]. as serum and plasma to be analyzed for For this purpose, a mouse antibody (IgG2A) clinical applications. directed against LAG3 was immobilized The coupling of SPRi biosensors and matrix- on a SPRi-Slide using a dedicated surface assisted laser desorption ionization mass chemistry compatible with MS analysis (NHS spectrometry (MALDI-MS) is an innovative chemistry). Before injecting LAG3, rat serum approach for biomarker discovery in biological albumin (RSA) was used to avoid non-specific fluids. It permits analytes captured by SPRi binding on the surface of the biochip. Then, to be identified and characterized by their the specific interaction of LAG3 (added in molecular weight and peptide sequence. plasma) and IgG2A was monitored using SPRi-MS opens a new method of detection, SPRi and images of the interaction were quantification and structural characterization of studied. Several femtomoles/mm² of LAG3 proteins of interest. In the future, it could help better discriminate between sub-species within a family of biomarkers. In this context, the complexity lies in the coupling of both techniques [8]. Most strategies require the elution of the bound analyte and its analysis by ESI- (electrospay ionisation) or MALDI-MS. This procedure has many drawbacks (analysis time, no multiplexing capabilities, decreased sensitivity, additional cross-contamination risks, etc.) which delayed the development of SPR-MS in the diagnostic field. 11
- 12. nanomagma proteins were captured by SPRi. After direct rapid and high-throughput information in real processing on the biochip surface (enzymatic time from up to several hundreds interactions digestion and matrix deposition), the SPRi- in parallel. The technology is sensitive and does Slide was analyzed using a MALDI-MS not require the use of labels. It can speed-up imager (Ultraflex, Bruker Daltonics). By showing the distribution of MS peaks specific of LAG3 and RSA respectively, it was possible to build the MS image of LAG3 spots (Figure 3) directly on the SPRi-Slide. The SuPRa-MS platform pioneers the combination of SPR imaging and MS imaging (MSi). It offers the possibility to gain spatially resolved information on the capture, sequence and molecular weight of clinical Fig. 3 > On-a-chip detection, identification and biomarkers. imaging of LAG3 protein (potential marker of Multiplexed SPRi analysis using the HORIBA breast cancer) at 10nM in human plasma through Scientific-GenOptics instruments provides the SuPRa-MS platform Impact Factor 7.333 2010 Journal Citation Reports® (Thomson Reuters, 2011) provides the very best forum for experimental For subscription details please and theoretical studies contact Wiley Customer Service: of fundamental and >> cs-journals@wiley.com applied interdisciplinary (Americas, Europe, Middle East and Africa, Asia Pacific) research at the micro- >> service@wiley-vch.de and nanoscales (Germany/Austria/Switzerland) 2011. Volume 7, 24 issues. >> cs-japan@wiley.com (Japan) Print ISSN 1613-6810 / Online ISSN 1613-6829 For more information please visit www.small-journal.com or contact us at small@wiley-vch.de 12
- 13. the workflow and reduce consumable costs the measurement of glycosaminoglycan binding during optimization processes. The coupling interactions. Anal. Chem. 80(9): 3476-3482. with MS analysis is straightforward and easier, [6] Roupioz Y. and al. (2009). “Individual Blood-Cell which makes it a valuable tool for biomarker Capture and 2D Organization on Microarrays” Small identification. 2009, 5, No. 13, 1493–1497. [7] Suraniti and al. (2007) Real-time detection of lymphocytes binding on an antibody chip using References SPR imaging. Lab Chip. 7: 1206-1208. [1] Uzun and al. (2009) Production of surface plasmon [8] Boireau and al. (2009) Revisited BIA-MS resonance based assay kit for hepatitis diagnosis. combinationb: Entire “on-a-chip” processing Biosensors and Bioelectronics. 24(9): 2878-2884. leading to the proteins identification at low [2] Spadavecchia and al. (2009) New cysteamine femtomole to sub-femtomole leveks? Biosensors based functionalization for biochip applications. and Bioelectronics 24: 1121-1127.[9] Sensors and Actuators B. 143(1):139-143. Bellon and al (2009) Hyphenation of Surface [3] Corne and al. (2008) SPR imaging for label-free Plasmon Resonance Imaging to Matrix-Assisted multiplexed analyses of DNA N-glycosylase Laser Desorption Ionization Mass Spectrometry by interactions with damaged DNA duplexes. Analyst. On-Chip Mass Spectrometry and Tandem Mass 133: 1036-1045. Spectrometry Analysis. Anal. Chem. 81: 7695– [4] Prieto and al. (2009) Synaptonemal complex 7702. assembly and H3K4Me3 demethylation determine [10] Triebel and al (2006) A soluble lymphocyte DIDO3 localization in meiosis. Chromosoma. 118: activation gene-3 (sLAG-3) protein as a prognostic 617-632. factor in human breast cancer expressing estrogen [5] Mercey and al. (2008) Polypyrrole oligosaccharide or progesterone receptors. Cancer Letters. array and surface plasmon resonance imaging for 235(1):147-53. 13
- 14. nanoresearch Light localization on a gold nanodisk array probed by near-field optics Loïc Lalouat1, Lionel Aigouy1, P Prieto2, A. . to know the position and the shape of the Vitrey2, J. Anguita2, A. Cebollada2, M.U. field enhancement zones with regard to the González2, A. García-Martín2 particles, as well as their vertical localization, 1 LPEM, UMR 8213 CNRS-ESPCI, Ecole and the influence of the incident polarization Supérieure de Physique et de Chimie Industrielle, direction. For that, theoretical simulations 10 rue Vauquelin, 75231 Paris cedex 5, France. with finite difference time domain (FDTD), 2 IMM-Instituto de Microelectrónica de Madrid finite boundary element or Green dyadic (CNM-CSIC), Isaac Newton 8, PTM, Tres Cantos, methods are often used, but the amount of E-28760 Madrid, Spain. experimental data available in the literature is quite reduced. In this work, we present an Arrays of metallic nanoparticles are artificial experimental study of the field localization on structures that can find many applications in a disk array with a scanning near-field optical physics and biology. When they are illuminated microscope (SNOM). Our experiments, by an external light source, strong evanescent which are in good agreement with numerical fields are localized in the near-field regions of simulations, show a strong localization of the the particles. These strong local fields can electromagnetic field between the particles, in be used for exciting single molecules, for the direction of the incident polarization. performing Raman scattering, for developing The experimental SNOM set-up is shown in biochemical sensors, or for performing Figure 1. In contrast to other SNOM techniques nanolithography [1-4]. The knowledge of the [5,6], our probe is a submicron size fluorescent local optical properties of these structures, particle glued at the end of a sharp tip [7]. In like the electromagnetic field distribution, is contact with the sample surface, it absorbs therefore of importance for developing such the local field at the excitation wavelength applications. For instance, it is interesting and emits light at a different one. By collecting the fluorescence as a function of the tip position on the surface, we obtain a fluorescence image which is directly related to the intensity of the electromagnetic field on the surface. The particle, which contains erbium and ytterbium ions, is excited at =975nm and its fluorescence is detected in the visible range at =550nm. Since this atypical excitation Fig. 1 > Description of the experimental set-up. The SNOM probe is a fluorescent particle 14 glued at the end of a sharp tip.
- 15. nanomagma process involves two photons, the collected fluorescence is proportional to the square of the total field intensity on the surface [7]. The sample studied is an array of gold nanodisks [diameter = 286nm, height=50nm, period=500nm, (see Figure 2)] fabricated on a glass substrate. The structure has a wide plasmon resonance peak located just below the excitation wavelength between 800 and 900nm. Fig. 3 > (a,b) SNOM images measured in a non-contact mode on the nanodisk array at =975nm. The dotted circles indicate the position of the disks. (c,d) FDTD calculations of the near-field distribution on the structures. The calculation represents the square of the intensity of the total field which is the quantity measured with the near-field fluorescent probe used in the Fig. 2 > SEM picture of the studied gold nanodisk. experiments. The calculation has been performed by taking into account of the probe size (a 160 nm large cube). The white arrows indicate the We show in Figure 3(a) and (b), the incident polarization direction. The scale bar is experimental near-field optical images of the 500 nm-long (taken from ref. 8). structure measured in a non-contact mode [8]. The incident polarization is linear and theoretical ones which exhibit the same indicated by the white arrow. The position of periodic pattern, and the same polarization the nanodisks is represented by the dotted dependence. white circles. The images show a periodic pattern, with elongated bright spots. Each We show in Figure 4(a) and (b) higher bright spot is in fact comprised of two lobes, resolution scans of the structure. Cross- located between the disks, and aligned in the sections extracted from the experimental incident polarization direction. and simulated images are shown in Figure To check the validity of the SNOM results, 4(c) and (d). The curves are in excellent we have performed an FDTD simulation agreement in terms of relative contrast. One of the measured signal. To make a realistic can clearly see that all the electromagnetic comparison, we calculated the square of field is concentrated between the disks in the the total field intensity and integrated this direction of the incident polarization and that quantity on a volume which represents the almost no light is located between the disks fluorescent particle size. Such procedure only in the direction perpendicular to the incident tends to broaden the size of the lobes but polarization. does not change the shape of the pattern. Another interesting parameter which The simulations are represented in Figure 3(c) characterizes nanodisks arrays is the vertical and (d). An excellent agreement is observed extension of the electromagnetic field above between the experimental images and the the surface. For instance, such parameter is 15
- 16. nanomagma Fig. 5 > Experimental SNOM image measured in planes perpendicular to the sample surface. The scanning planes are indicated on the drawing. The curve in the bottom shows the vertical Fig. 4 > (a) : Experimental SNOM image measured decay of the measured signal (taken from ref. 8). on the array of gold nanodisks ; (b) FDTD simulation of the near-field optical signal; (c,d) cross- sections parallel (direction A) and perpendicular References (direction B) to the incident polarization The position of the gold disks is the same than the [1] S. A. Maier, and H. A. Atwater, J. Appl. Phys. ones shown in Fig. 3 (taken from ref. 8). 98, 011101 (2005). [2] N. Fang, H. Lee, C. Sun, and X. Zhang, Science 308, 534 (2005). important if we want to use the structures for [3] S. S. Aćimović, M. P. Kreuzer, M. U. González, performing nanolithography, because it will and R. Quidant, ACS Nano 3, 1231 (2009). determine the penetration depth of the light [4] A. F. Koenderink, J. V. Hernández, F. in a photoresist situated on top of the array. Robicheaux, L. D. Noordam, and A. Polman, To have an idea of this localization, we have Nano Lett. 7, 745 (2007). performed scans in planes perpendicular [5] M. Schnell, A. Garcia-Etwarri, A. J. Huber, K. to the sample surface. We show in Figure 5 Crozier, J. Aizpurua, and R. Hillenbrand, Nat. the experimentally measured signals which Photon. 3, 287 (2009). indicate that the light is essentially confined [6] M. Salerno, N. Félidj, J. R. Krenn, A. Leitner, F. at close distance from the surface. Above R. Aussenegg, and J. C. Weeber, Phys. Rev. B 200nm, no signal is detectable anymore. 63, 165422 (2001). Such distance depends on the structure of [7] L. Aigouy, Y. De Wilde, and M. Mortier, Appl. the array, and in particular on the disks size, Phys. Lett. 83, 147 (2003). their thickness and their separation. [8] L. Aigouy, P. Prieto, A. Vitrey, J. Anguita, A. To summarize, we have performed a study of Cebollada, M.U. González, A. García-Martín, the light localization on a gold nanodisk array J. Labéguerie-Egéa, M. Mortier, J. Appl. Phys. by near-field optics. The near field has been 110, 044308 (2011). measured using a fluorescent particle glued at the end of a sharp tip. The measured near- field images, which represent the square of the total field intensity, show that the light is localized between the disks in the direction of the incident polarization direction. The results are in good agreement with numerical simulations performed by finite difference time 16 domain method.
- 17. nanoresearch Controlling fluorescence resonant energy transfer with a magneto-optical nanoantenna R. Vincent and R. Carminati In the present work, we use an established Institut Langevin, ESPCI ParisTech, CNRS, general framework for dipole-dipole energy 10 rue Vauquelin, 75231 Paris Cedex 05, France. transfer between an emitter and an absorber in a nanostructured environment [5]. The theory allows us to address FRET between a donor and an acceptor in the presence Energy transfer between a molecule in an excited of a nanoparticle with an anisotropic state (donor) and a molecule in the ground electromagnetic response. For the case of state (acceptor) underlies many significant a nanoparticle with an anisotropic dielectric photophysical and photochemical processes, response (e.g., a nanoparticle made of a from photosynthesis to fluorescence probing ferromagnetic material exhibiting a magneto- of biological systems. It is also of interest in optical response), the distance dependence, nanophotonics where efficient transfer of optical the orientation dependence and the strength excitations on subwavelength scales is a key of the FRET efficiency can be changed issue. Depending on the separation between substantially. In the case of a magneto-optical the donor (D) and the acceptor (A), the process anisotropy, it can in principle be controlled can be described accurately by various using the static magnetic field as an external theories accounting for the electromagnetic control parameter. interaction between the two species. For a D-A distance range on the order of 2-10 nm, which is relevant for photochemical studies and nanophotonics, the well-established Förster theory [1] based on quasi-static dipole- dipole interaction has been very successful. It shows that while Förster Resonant Energy Transfer (FRET) is a very useful process that can be used, for example, as a ruler for spectroscopic measurements [2], it is a rather weak process that goes down as the inverse Fig. 1 > Left panel: Schematic configuration of the D-A system in the presence of a nanoparticle. sixth power R6 of the D-A separation [3]. In The different channels for energy transfer (direct fact, one can introduce a length scale known or indirect) are indicated by dotted arrows. When as the Förster distance R0 at which FRET is the transition dipoles are orthogonal, the direct 50% efficient and it is found that R0 is on the Förster transfer is disabled. Right panel: Energy- order of a few nanometers in most practical level diagram of the FRET process between a do- situations. For even smaller distances, Dexter nor and acceptor molecules. [4] recognized that electronic exchange and multipolar interactions become important and In principle, the presence of a nanostructure a full quantum mechanical treatment must be close to a D-A couple will modify the emission implemented. On the other hand, in the large and absorption by the transition dipoles; here distance regime (non-negligible compared to we use the formalism to express explicitly the wavelength), full electrodynamics is needed the FRET rate of a D-A couple interacting to account for retardation effects. with a spherical nanoparticle exhibiting a 17
- 18. nanomagma purely metallic response or a magneto-optical where RNP is the radius of the nanoparticle. response. We have shown previously [5], that This simple expression shows that the ratio the FRET rate mediated by the nanoparticle Rp/RNP is the crucial parameter that describes can be expressed simply as follows the influence of the nanoparticle on the FRET rate. For Rp>RNP, the nanoparticle enhances the FRET transfer, while for Rp<<RNP, the FRET becomes exclusively driven by the direct transfer. Moreover, in previous work In this expression is the energy transfer [5], we have shown that in the condition rate from donor to acceptor mediated by the that the polarizability of the nanoparticle α(ω) nanoparticle, 0 is the decay rate of the donor varies smoothly on the frequency range of the in free space, R0 is the Förster radius for an spectral overlap between absorption cross orientational factor equal to one, RA is the section of the acceptor and the normalized distance of the acceptor to the nanoparticle, RD emission spectrum of the donor, the the distance of the donor to the nanoparticle, polarization coupling radius Rp depends only and Rp is the polarization coupling radius [5] on the polarizability tensor of the nanoparticle. which describes the influence of the nanoparticle on the FRET rate meditated by the nanoparticle. This distance Rp defines an influence radius of the nanoparticle, it allows to compare the indirect FRET rate (i.e., mediated by the nanoparticle) and the standard free- space FRET rate . Fig. 3 > Ratio Rp/RNP for Iron (blue solid line), Nickel (gold dash-dotted line), and Cobalt (red dashed line) as a function of the emission wavelength lof the donor. RNP=10 nm. The configuration is illustrated in the inset, showing that the dipole are collinear and the couple Donor-Acceptor and nanoparticle are aligned. Fig. 2 > Two canonical configurations of the tree We illustrate the formalism for the well-known body system Donor-NP-Acceptor studied in the metallic nanoparticle. Noble metals are known present work. (a) Left panel: Aligned configuration. to hold plasmon resonances that enhance, for (b) Right panel: Orthogonal configuration. The example, the polarizability of a nanoparticle. arrows illustrate the molecular dipole orientations. Since the polarization coupling radius Rp directly depends on the polarizability, one can For the sake of illustration, let us consider the expect a substantial influence of the plasmon situation in which the three bodies are aligned resonance on the FRET rate mediated by with RD=RA=2RNP, and the transition dipole are the nanoparticle. This is indeed what we aligned in the same direction [see Fig. 2(a)], in observe in Fig. 3, in which we have plotted this case, we obtain the ratio Rp/RNP (with RNP=10 nm) versus the emission wavelength of the donor for gold and silver that are common materials in studies of fluorescence enhancement or quenching. The 18 plasmon resonance is visible in both cases,
- 19. nanomagma leading to an enhancement of Rp/RNP. For instance in the case of silver, one reaches Rp/ RNP 3; for gold one has Rp/RNP 1.9. In the particular conditions RD = RA = 2 RNP and R = 4 RNP, we obtain an enhancement factor of the FRET rate on the order of 180 for silver and 10 for gold. For a D-A couple working at plasmon resonance with these materials, we conclude that FRET is mainly driven by the nanoparticle. Incidentally, Fig. 5 > Ratio Rp / RNP for Iron (blue solid line), Nickel any change of the dielectric property of the (yellow dash-dotted line), and Cobalt (red dashed nanoparticle will be reflected in a modulation line) as a function of the emission wavelength of the FRET rate. Modulation of the dielectric of the donor in the presence of an external response can be achieved, for example, magnetic field inducing a magnetization in the through the magneto-optical effect that we direction orthogonal of the plane containing the consider in the following. tree body D-A-NP. RNP =10 nm. The configuration is illustrated in the inset. with a change of the optical dielectric response (anisotropic response). Therefore ferromagnetic particles own an anisotropic dielectric response controlled by an external magnetic field. Fig. 4 is an illustration of the ratio Rp/RNP (with RNP =10 nm) versus the emission wavelength of the donor for different standard magneto- optical materials: Nickel, Iron and Cobalt in an Fig. 4 > Ratio Rp/RNP for Iron (blue solid line), Nickel aligned configuration (see Fig. 2(a) for a sketch (gold dash-dotted line), and Cobalt (red dashed of the geometry). We observe a smoother line) as a function of the emission wavelength behavior than in the case of noble metals. Its of the donor. RNP=10 nm. The configuration is origin lie in the stronger plasmon damping illustrated in the inset, showing that the dipole of magneto-optical materials comparing to are collinear and the couple Donor-Acceptor and nanoparticle are aligned. (b) Right panel: metallic materials. For these materials, the Canonical orthogonal configuration. amplification factor is around twenty, therefore in this configuration, the FRET rate is still govern by the nanoparticle. Using experimental data for the dielectric Figure 5 shows a computation of the ratio Rp/ function of different magneto-optical materials RNP with the same materials as in Fig. 4, but [6], we compute the polarization coupling in the case of an orthogonal configuration radius Rp normalized by the nanoparticle radius [see Fig. 2(b) for a sketch of the orthogonal RNP as a function of several parameters: The geometry]. The magnetization is orthogonal emission wavelength of the donor, the radius of to the plane containing the D-A couple and the nanoparticle, and the material properties. the nanoparticle. Let us stress that in this Ferromagnetic materials are materials with configuration the FRET rate vanishes in magnetic anisotropy. Magnetic anisotropy is absence of an external static magnetic field a consequence of the different directions of due to the orthogonality of the donor and magnetization of the different magnetization acceptor transition dipoles. Although one domains. At saturation, small nanoparticles observes that Rp/RNP remains smaller than are customarily considered owning a single one, the possibility of inducing a FRET rate domain. This change in magnetization comes driven only by the polarization anisotropy of the 19
- 20. nanomagma nanoparticle is an interesting result, showing anisotropy can be controlled by an external the potential of magneto-optical nanoparticles static magnetic field, and we have discussed for FRET. On the one hand, the anisotropic potential application for FRET tuning and response allows us to couple molecules for modulation. Here, we have presented a proof which standard FRET gives a vanishing signal of concept. Further work should focus on due to orientational mismatch (orthogonal enhancing the (weak) magneto-optical FRET transition dipole). On the other hand, the signal. We have illustrated the effect also for the possibility of controlling the magneto-optical well known metallic nanoparticle, showing that response with a static magnetic field as an it furnishes insight in the understanding of the external parameter could allow us to tune good quantities controlling this process. or modulate the FRET rate, which can be an advantage, for example, to increase the References sensitivity of the detection process. [1] T. Förster, Ann. Phys. 437, 55 (1948); Discuss. We have elucidated the Förster energy transfer Faraday Soc. 27, 7 (1959). problem in a three body configuration, involving [2] L. Stryer, Annu. Rev. Biochem. 47, 819 (1978). two fluorophores close to a nanoparticle [3] L. Novotny, B. Hecht, Principles of Nano-optics, with an anisotropic dielectric response. We Cambridge University Press, (2006). have shown that the distance dependence [4] D. L. Dexter, J. Chem. Phys. 21, 836 (1953). is controlled by the Förster radius and a new [5] R. Vincent, and R. Carminati, Magneto-optical distance that depends of the polarization control of Förster energy transfer, Phys. Rev. B 83, properties of the nanoparticle. We have 165426 (2011). illustrated the effects in the case of a magneto- [6] E. D. Palik, Handbook of Optical Constants of optical nanoparticle for which the degree of Solids (Academic, New York, 1985). 20
- 21. www.nanociencia.imdea.org RESEARCH PROGRAMMES • Molecular nanoscience IMDEA-Nanociencia is a private Foundation created by joint initia- tive of the Comunidad de Madrid and the Ministry of Education of • Scanning probe microscopies the Government of Spain in February 2007 to manage a new and surfaces research Institute in Nanoscience and Nanotechnology (IMDEA- Nanociencia). The Institute is located at the campus of the Univer- • Nanomagnetism sidad Autónoma de Madrid in Cantoblanco. The Institute aims at performing research of excellence in selected • Nanobiosystems: biomachines and manipulation of macromolecules areas and offers attractive opportunities to develop a career in sci- ence at various levels from Ph.D. students to senior staff positions. • Nanoelectronics and superconductivity The Madrid Institute for Advanced Studies in Nanoscience also develops an important program of technology transfer and creation of spin-off companies. • Semiconducting nanostructures and nanophotonics E-mail contacto.nanociencia@imdea.org Phone 34 91 497 68 49 / 68 51 Fax 34 91 497 68 55 • Nanofabrication and advanced instrumentation [Nanociencia y Nanotecnología: lo pequeño es diferente small is different Nanoscience and Nanotechnology: ]
- 22. nanoresearch Internal electromagnetic field distribution and magneto-optical activity of metal and metal-dielectric magnetoplasmonic nanodisks D. Meneses-Rodríguez, E. Ferreiro-Vila, J.C. structures respectively, with total heights Banthí, P Prieto, J. Anguita, A. García-Martín, . between 50 and 70 nm and diameters M. U. González, J. M. García-Martín, A. between 110 and 140nm (Figure 1(a)). For Cebollada and G. Armelles the sake of comparison, continuous thin films IMM-Instituto de Microelectrónica de Madrid with identical composition have been also (CNM-CSIC), Isaac Newton 8, PTM, E-28760 prepared. Tres Cantos (Madrid), Spain. The MO activity ( ) has been obtained david.meneses@imm.cnm.csic.es by measuring the MO Kerr effect in polar configuration upon normal incidence illumination, previously identifying the optical Localized surface plasmon resonances resonances through extinction spectra. In (LSPRs) greatly influence the optical [1-4] the fully metallic nanostructures, we find and magneto-optical (MO) [5-10] properties a distinctive evolution as a function of Co of fully metallic and metal-dielectric position of the MO activity in the nanodiscs nanostructures. The observed enhancement compared with that of the continuous layers, in the MO activity when these LSPRs are with maximum values when the Co layer excited is attributed to the high intensity of is located near the top or the bottom of the electromagnetic (EM) field inside the the disks and minimum values in-between global nanostructure when the LSPR occurs due to the LSPR excitation (Figure 1(b)). [5,11]. Unfortunately, it is not straightforward This behavior is in contrast with the MO to experimentally determine the intensity of activity exhibited by the continuous films, the EM field inside a nanostructure. Here which increases monotonously as the Co we show how the EM profile related to layer becomes closer to the top surface. the LSPR can be probed locally inside the This indicates that the EM field inside the nanostructure by measuring the MO activity nanodisks exhibits a nonuniform distribution of the system as a function of the position a in plasmon resonance conditions. In fact, MO active probe (a Co nanolayer) [12]. This the Co layer acts as a probe sensing the will be done in full detail in metallic systems, EM field within the nanodisk, since the MO and preliminary results will also be presented activity depends on the intensity of such field. in more complex metal-dielectric magneto- Preliminary results on the possible influence plasmonic nanodiscs. of multiple resonances in metal-dielectric The magnetoplasmonic nanodisk arrays magnetoplasmonic nanodiscs will be also have been fabricated in large area onto presented. glass substrates by combining colloidal This information could be very relevant for lithography with sputter, thermal and the design of magnetoplasmonic systems electron beam deposition and lift-off offering optimum MO enhancement, for techniques. Typical nanodisk structures are instance for sensing applications where Au/Co/Au/Cr and Au/SiO2/Co/SiO2/Au/Ti, maximum sensitivity is expected in the areas 22 for the fully metallic and the metal-dielectric with higher EM field.
- 23. nanomagma (a) [4] T. Pakizeh, A. Dimitriev, M. S. Abrishamian, N. Granpayeh, and M. Häll, J. Opt. Soc. Am. B 25 (2008) 659. [5] J. B. González-Díaz, A. García-Martín, J. M. García-Martín, A. Cebollada, G. Armelles, B. Sepúlveda, Y. Alaverdyan and M. Käll, Small 4 (2008) 202. [6] G. A. Wurtz, W. Hendren, R. Pollard, R. Atkinson, L. Le Guyader, A. Kirilyuk, Th. Rasing, I. I. Smolyaninov and A. V. Zayats, New J. of Phys. 10 (2008) 105012. (b) [7] P. K. Jain, Y. Xiao, R. Walsworth, and A. E. Cohen, Nanolett. 9 (2009) 1644. [8] G. X. Du, T. Mori, M. Suzuki, S. Saito, H. Fukuda, and M. Takahashi, Appl. Phys. Lett. 96 (2010) 081915. [9] L. Wang, K. Yang, C. Clavero, A. J. Nelson, K. J. Karroll, E. E. Carpenter, and R. A. Lukaszew, J. Appl. Phys. 107 (2010) 09B303. [10] G. X. Du, T. Mori, M. Suzuki, S. Saito, H. Fukuda, and M. Takahashi, J. Appl. Phys. 107 (2010) 09A928. [11] G. Armelles, A. Cebollada, A. García-Martín, J. M. García-Martín, M. U. González, J. B. González-Díaz, E. Ferreiro-Vila and J. F. (c) Torrado, J. Opt. A: Pure Appl. Opt. 11 (2009) 114023. [12] D. Meneses-Rodríguez, E. Ferreiro-Vila, P. Prieto, J. Anguita, M. U. González, J. M. García- Martín, A. Cebollada, A. García-Martín and G. Armelles, Small, DOI: 10.1002/smll.201101060 (2011). Fig. 1 > (a) Sketch of the fully metallic nanodiscs (b) Maximum magneto-optical activity as a function of the Co position for fully metallic nanodiscs (c) SEM image of an array of metallic nanodisc (Inset: extinction spectrum). References [1] S. A. Maier, Plasmonics: Fundamentals and Applications (Springer, Berlin, 2007). [2] S. A. Maier and H. A. Atwater, J. Appl. Phys. 98, (2005) 011101. [3] K. H. Su, Q. H. Wei, and X. Zhang, Appl. Phys. Lett. 88 (2006) 063118. 23
- 24. nanoresearch Magneto-Optical properties of nanoparticles R. Gómez-Medina1, B. García-Cámara2, corrections to the electrostatic polarizability I. Suárez-Lacalle1, L. S. Froufe-Pérez1, F. tensor. González2, F. Moreno2, M. Nieto-Vesperinas3 and J. J. Sáenz1 1 Departamento de Física de la Materia Condensada and Instituto “Nicolás Cabrera”, Universidad Autónoma de Madrid, 28049 Madrid, Spain. 2 Grupo de Óptica, Departamento de Física Aplicada, Universidad Cantabria, 39005 Santander, Spain. 3 Instituto de Ciencia de Materiales de Madrid, C.S.I.C., Campus de Cantoblanco, 28049 Madrid, Spain. juanjo.saenz@uam.es Fig. 1 > Scattering cross section map of a non- Electromagnetic scattering from nanometer- absorbing Mie sphere as a function of the scale objects has long been a topic of refractive index m and the y parameter, y = mka = large interest and relevance to fields from m(2 a/ ). Green areas correspond to parameter astrophysics or meteorology to biophysics, ranges where the magnetic dipole contribution medicine and material science [1-5]. In the dominates the total scattering cross section, while red areas represent regions where the last few years, small particles with resonant electric dipole contribution is dominating. Higher magnetic properties are being explored as order multipoles dominate the remaining blue- constitutive elements of new metamaterials saturated areas. (Adapted from Ref. [2]). and devices. The studies in the field often involve randomly distributed small elements or particles where the dipole approximation We will also explore the properties of high- permittivity dielectric particles with resonant may be sufficient to describe the optical magnetic properties as constitutive elements response. We will discuss the optical of new metamaterials and devices [2]. response of disordered nano-materials where Magnetic properties of low-loss dielectric the constitutive nanoparticles can have a non- nanoparticles in the visible or infrared are not negligible response to static (Magneto-Optical expected due to intrinsic low refractive index active nanoparticles) or dynamic (Magneto- of optical media in these regimes. Here we dielectric nanoparticles) magnetic fields. analyze the dipolar electric and magnetic We will first analyze the peculiar scattering response of lossless dielectric spheres properties of single nanoparticles. In made of moderate permittivity materials. particular, we derive the radiative corrections For low material refractive index there are no to the polarizability tensor of anisotropic sharp resonances due to strong overlapping particles, a fundamental issue to understand between different multipole contributions. the energy balance between absorption However, we find that Silicon particles with and scattering processes [1]. As we will index of refraction 3.5 and radius 200nm show, Magneto optical Kerr effects in non- present strong electric and magnetic dipolar absorbing nanoparticles with magneto-optical resonances in telecom and near-infrared 24 activity arise as a consequence of radiative frequencies, (i.e. at wavelengths ≈ 1.2 – 2 μm)
- 25. nanomagma without spectral overlap with quadrupolar and higher order resonances. The light scattered by these Si particles can then be perfectly described by dipolar electric and magnetic fields. Fig. 3 > Scattering diagrams for a Ge nanosphere with 240nm radius (After Ref. [4]). Acknowledgments We appreciate interesting discussions with J. Aizpurua, S. Albaladejo, P. Albella, A. García- Etxarri, M.I. Marqués and F. Scheffold. This work has been sup- ported by the EU NMP3- SL-2008-214107-Nanomagma, the Spanish MICINN Consolider NanoLight (CSD2007- 00046), FIS2010-21984, FIS2009-13430- C01-C02, and FIS2007-60158, as well as by the Comunidad de Madrid Microseres- CM (S2009/TIC-1476). References [1] S. Albaladejo,R. Gómez-Medina, L. S. Froufe- Pérez, H. Marinchio, R. Carminati, J. F. Torrado, G. Armelles, A. García-Martín and J.J. Sáenz, Fig. 2 > Effective real and imaginary permittivities Opt. Express 18 (2010) 3556. and permeabilities for an arbitrary arrangement of Si spheres in an otherwise homogeneous [2] A. García-Etxarri, R. Gómez-Medina, L. S. medium with εh = μh = 1 for two different filling Froufe-Pérez, C. López, L. Chantada, F. factors f = 0.25 (a) and f = 0.5 (b). (From Ref. [2]). Scheffold, J. Aizpurúa, M. Nieto-Vesperinas and J. J. Sáenz, Opt. Express 19, 4815 (2011). As we will see, the striking characteristics of [3] M. Nieto-Vesperinas, R. Gómez-Medina, and J. the scattering diagram of small magneto- J. Sáenz, J. Opt. Soc. Am. A 28 (2011) 54. optical and magnetodielectric particles [3,4] [4] R. Gómez-Medina, B. García-Cámara, I. lead to a number of non-conventional effects Suárez-Lacalle, F. González, F. Moreno, M. in the optical response of nanostructured Nieto-Vesperinas, J. J. Sáenz, J. Nanophoton. magneto-optical structures. 5, 053512 (2011). 25
- 26. nanoresearch Three-dimensional optical metamaterials and nanoantennas: Chirality, Coupling, and Sensing Harald Giessen which are favorable for emitting and receiving 4th Physics Institute, University of Stuttgart, radiation from quantum systems [9]. D-70569 Stuttgart, Germany. giessen@physik.uni-stuttgart.de References Metallic metamaterials have shown a number [1] Na Liu, Hongcang Guo, Liwei Fu, Stefan Kaiser, of fascinating properties over the last few Heinz Schweizer, and Harald Giessen: Three- years. A negative refractive index, negative dimensional photonic metamaterials at optical refraction, superlenses, and optical cloaking frequencies, Nature Materials 7, 31 (2008). are some of the ambitious applications where [2] N. Liu, H. Guo, L. Fu, S. Kaiser, H. Schweizer, metamaterials hold great promise. and H. Giessen: Plasmon Hybridization in Stacked Cut-Wire Metamaterials, Advanced We are going to present fabrication methods Materials 19, 3628 (2007). for the manufacturing of 3D metamaterials [1]. [3] Na Liu, Liwei Fu, Stefan Kaiser, Heinz Schweizer, We are investigating their coupling properties and Harald Giessen: Plasmonic Building Blocks and the resulting optical spectra. Hybridization for Magnetic Molecules in Three-Dimensional of the electric [2] as well as the magnetic [3] Optical Metamaterials, Advanced Materials 20, resonances allows us to easily understand the 3859 (2008). complex optical properties. Lateral as well as [4] B. Lukyanchuk, N. I. Zheludev, S. A. Maier, N. vertical coupling can result in Fano-resonances J. Halas, P. Nordlander, H. Giessen, and C. [4] and EIT-like phenomena [5, 6]. These T. Chong: The Fano resonance in plasmonic phenomena allow to construct novel LSPR nanostructures and metamaterials, Nature sensors with a figure of merit as high as five [7]. Materials 9, 707 (2010). The connection between structural symmetry [5] Na Liu, Stefan Kaiser, and Harald Giessen: and their electric as well as magnetic dipole Magnetoinductive and Electroinductive Coupling and higher-order multipole coupling will be in Plasmonic Metamaterial Molecules, Advanced elucidated. It turns out that stereometamaterials Materials 20, 4521 (2008). [8], where the spatial arrangement of the [6] Na Liu, N. Liu, L. Langguth, T. Weiss, J. Kästel, constituents is varied (see figure), reveal a M. Fleischhauer, T. Pfau, and H. Giessen: highly complex rotational dispersion. The chiral Plasmonic EIT analog at the Drude damping properties are limit, Nature Materials 8, 758 (2009). quite intriguing [7] Na Liu, T. Weiss, M. Mesch, L. Langguth, U. and can be Eigenthaler, M. Hirschner, C. Sönnichsen, and explained by a H. Giessen: Planar metamaterial analog of coupled oscillator electromagnetically induced transparency for model. plasmonic sensing, Nano Lett. 10, 1103 (2010). Our three- [8] Na Liu, Hui Liu, Shining Zhu, and Harald Giessen: dimensional Stereometamaterials, Nature Photonics 3, 157 stacking approach (2009). allows also for the [9] H. Giessen and M. Lippitz: Directing light fabrication of 3D emission from quantum dots, Science 329, 910 26 nanoantennas, (2010).
- 27. nanoresearch Spin transfer RF nano-oscillators for wireless communications and microwave assisted magnetic recording U. Ebels1, M. Quinsat1,2, D. Gusakova1, J. F. oscillations combined with the giant or tunnel Sierra1, JP Michel1,2 , D. Houssameddine1, magnetoresistance of the stack generate B. Delaet2, M.-C. Cyrille2, L. D. Buda- oscillations of the voltage across the stack Prejbeanu1,3, B. Dieny1 at GHz frequencies. Moreover the frequency 1 SPINTEC, UMR(8191) CEA / CNRS / UJF / varies as a function of the current density Grenoble INP ; INAC, 17 rue des Martyrs, 38054 flowing through the stack. Grenoble Cedex, France. 2 CEA-LETI, MINATEC, 17 Rue des Martyrs, 38054 This phenomenon can be used to design Grenoble, France. frequency tunable RF oscillators which could 3 Grenoble INP 46, Avenue Félix Viallet, 38031 , be quite useful in a number of devices such Grenoble Cedex 1, France. as RF STT oscillators (STO) for wireless communications, or as microwave generators Slonczewski [1] and Berger [2] predicted in to assist the writing by microwaves in 1996 that a spin-polarized current flowing magnetic recording technology or as through a magnetic nanostructure exerts magnetic field sensors taking advantage of a torque on its magnetization due to the the shift of frequency induced by an applied exchange interaction between the spin magnetic field. of the conduction electrons and the spin In this paper, our R&D efforts on STT RF of the electrons responsible for the local oscillators as well as current trends in this field magnetization. This torque is called spin are described. transfer torque (STT). The possibility to use the STT to switch the magnetization A significant effort has been focused on a of a magnetic nanostructure was first particular configuration of STT RF oscillators experimentally observed in metallic spin- in which an out-of-plane magnetized polarizer valve nanopillars [3] and later in magnetic is used to inject out-of-plane spin polarized tunnel junctions [4]. The spin torque acts as electrons into an in-plane magnetized a damping or antidamping term and can free layer [5]. Indeed, it was shown that induce very peculiar magnetization dynamics. this configuration is particularly interesting Of particular interest is when an applied since it allows generating large angle field and the spin transfer torque (STT) have precessional motion thereby maximizing competing influence on the magnetization of the magnetoresistance signal associated the free layer of a spin-valve or of a magnetic with this motion5. In this configuration, the tunnel junction, for instance the field favoring frequency varies almost linearly with current parallel alignment between the magnetization up to a maximum value where it saturates of the free layer magnetization and that of because of micromagnetic distorsion of the the reference layer whereas the STT favors magnetization. antiparallel alignment. In such situations, the Two important characteristics must be magnetization of the free layer is driven into carefully addressed in such oscillators before steady state oscillations. The magnetization being able to use them in RF devices for continuously pumps energy into the spin wireless communications. One is the output current to compensate the dissipation due power, the other is the excitation linewidth and to Gilbert damping. These steady state associated phase noise. 27
- 28. nanomagma low for practical applications (300MHz-1GHz), these results demonstrate the possibility to increase the output powers to acceptable value for this type of applications thanks to the use of magnetic tunnel junctions. Concerning the linewidth and phase noise, several studies have aimed at understanding the cause of the linewidth in STO oscillators in order to try increasing the coherence of the magnetization dynamics and thereby minimize the excitation linewidth and phase noise. Frequency and time-domain characterizations were performed. As an example, Figure 2 shows time domain measurements performed on MgO magnetic tunnel junction (MTJ) pillars [7]. The RF voltage was measured between the top and bottom electrodes of the MTJ while a DC current I flows through the pillar. Figure 2 clearly shows that the STT induced magnetic excitations start above a current threshold. However, the excitations first appear in bursts (region 2). As the current is further increased, the excitations become more and more persistent. Fig. 1 > STT oscillator with perpendicular polarizer and in-plane free layer. A fixed in- plane magnetized reference layer is added to produce a magnetoresistance between this reference layer and the precessing free layer. Typical spectra obtained when measuring the RF voltage between top and bottom electrodes. The oscillator pillar has typical diameter between 150nm and 50nm. Fig. 2 > Time domain measurements of the RF By using magnetic tunnel junctions, the voltage induced by STT excitations in a MgO output power could be increased by 2 orders based MTJ submitted to a DC current. Left: Power of magnitude thanks to the higher impedance of excitations versus DC current amplitude. Right: of these systems [6,7]. Recently, output real time voltage measurements for three different powers of the order of 1 μV were reported values of the DC current flowing through the MTJ. in STT oscillator based on magnetic tunnel junction and exploiting a vortex configuration By performing a Fourier transform over a of magnetization [6]. Although the frequency sliding time window of 10ns of the RF voltage 28 associated with these vortex based STO is too associated with these steady state excitations,
- 29. nanomagma it appears that the excitation frequency is not and iii) the ability to write on the media with stable but fluctuates (see the spectrogram in magnetic field which can be produced by the Fig. 3). write head (maximum produced field by the write head of the order of 2T). It was therefore proposed to assist the writing either with a temporary heating of the media (Heat Assisted Magnetic Recording: HAMR) or by microwave (Microwave Assisted Magnetic Recording: MAMR). In MAMR a spin-transfer oscillator is inserted in the write gap of the head. This oscillator has also a perpendicular to plane polarizer Fig. 3 > Spectrogram obtained by a Fourier combined with and in-plane magnetized transform over a sliding time window of 10ns of the real-time RF voltage shown in Fig. 2, regime 3. free layer. However, it has no reference layer since its purpose is to generate a RF field outside the pillar (i.e. on the media, where the These frequency fluctuations can have different bit has to be written) and not a RF voltage origins. One is the influence of the thermal across the pillar. The precession of the free activation kBT. Random thermal fluctuations layer generates a rotating dipolar RF field perturb the modes which are excited by outside the nano-oscillator pillar. This RF STT, temporarily changing the magnetization field penetrates into the media and transfer dynamics. Furthermore, the magnetization dynamics under STT is highly non-linear. The energy to the magnetization of the grains. non-linearity, as in systems prone to chaotic This additional energy combined with the behavior, can lead to unstable magnetization field from the write pole of the head allows dynamics. Furthermore the non-linearity also the switching of the magnetization of the causes the frequency to depend on the media. This technology is under development amplitude of the excitations which may also in most major Hard Disk Drive companies. participate to the fluctuations seen in Fig. 3. The difficulties are too produce enough RF power at a frequency of the order of 30 to By optimizing the structure of the stack and 40GHz close to the FMR frequencies of for instance using synthetic antiferromagnetic high anisotropy magnetic media and ensure free layer, the linewidth could be also that this power is primarily absorbed in the significantly reduced. media and not in the surrounding magnetic Another area where these STT oscillators environment, particularly the writing pole and can be quite useful is the one of magnetic trailing shield. recording. The present technology of recording which consists in storing the information on granular media and switching the magnetization of the grains with a write head which is a tiny electromagnet, is reaching a physical limit called the magnetic trilemma. This trilemma is caused by the impossibility to satisfy simultaneously i) a sufficient stability of the magnetization of the grains in the media against thermal fluctuations, ii) Fig. 4 > Schematic representation of the a sufficient media signal to noise ratio operation principle of MAMR. 29
- 30. nanomagma Another area of increasing interest concerning Rodmacq, I. Firastrau, F. Ponthenier, M. Brunet, STO is the one of magnetic field sensors. The C. Thirion, J. P. Michel, L. D. Buda-Prejbeanu, basic idea is to use the dependence of the M. C. Cyrille, O. Redon, and B. Dieny, Nature oscillation frequency of STO on the applied Materials 6, 447 (2007). field to measure the amplitude of the applied [6] A.Dussaux, B.Georges, J.Grollier, V.Cros, field [8]. Figure 5 illustrates the variation f(H) A.V.Khvalkovskiy, A.Fukushima, M.Konoto, in a spin-valve structure traversed by a DC H.Kubota, K.Yakushiji , S.Yuasa, etal., Nature current. The magnetization of the soft layer Communications 1, 1(2010), ISSN2041-1723. is driven into steady state oscillations. The giant magnetoresistance of the stack then [7] D. Houssameddine, U. Ebels, B. Dieny, K. produces an oscillatory voltage between Garello, J.-P. Michel, B. Delaet, B. Viala, M.-C. top and bottom electrodes [9]. The shift of Cyrille, J. A. Katine and D. Mauri, Phys. Rev. frequency versus applied field can be quite Lett. 102, 257202 (2009). steep, as large as 180GHz/T [10]. With an [8] Sato et al, US7 471 491 B2 (2008). appropriate frequency modulation detection [9] P.M.Braganca, B.A.Gurney, J.A.Katine, S.Maat, scheme, this approach could allow the J.R.Childress, Nanotechnology 21 (2010) realization of very small magnetic field sensors 235202. sub-30nm*30nm which could replace TMR [10] N.Stutzke, S.L.Burkett, S.E.Russek, Appl. sensors in read heads of HDD. Phys.Lett.82 (2003) 91. [11] K.Mizushima,K.Kudo,T.Nagasawa,and R.Sato, Journ.Appl.Phys. 107,063904(2010). Fig. 5 > Shift of frequency versus applied field measured in a spin-valve based STO. From Ref.[9]. References [1] Slonczewski, J., “Currents and torques in metallic magnetic multilayers”, J.Magn.Magn. Mater.159, L1 (1996). [2] Berger, L., Phys.Rev.B 54, 9353 (1996). [3] Katine, J.A., Albert, F.J ., Buhrman, R.A., Myers, E.B., and Ralph, D.C, Phys.Rev.Lett.84, 3149 (2000). [4] Y.Huai et al, Appl.Phys.Lett.84 (2004), 3118. 30 [5] D. Houssameddine, U. Ebels, B. Delaët, B.
- 31. nanoresearch Characterization of an electrostatically actuated Pd coated MEMS resonators J. Henriksson1, J. Arcamone2, G. Villanueva1, plate capacitance as well as the capacitance J. Brugger1 between the electric paths on the chip and 1 Microsystems Laboratory, EPFL, Lausanne, CH- wires connecting onto the chip (see Figure 2). 1015, Switzerland. 2 CEA, LETI, MINATEC, F-38054 Grenoble, France. jonas.henriksson@epfl.ch Introduction Electrostatically actuated MEMS resonators Fig. 2 > The electrical equivalent of the MEMS feature CMOS integrability, ultra-low power device. consumption and stable readout. To utilize these properties for gas sensing, we The sensing principle is based on the fact fabricated a doubly clamped free-standing that Pd expands in the presence of H2. This beam of amorphous silicon (see Figure 1). A property has been used for H2 sensing in functionalizing layer of palladium is patterned different configurations, such as cantilevers to cover the beam. The Pd also serves as [3-5], chemo-mechanical switches [6, 7] and electrical path on the beam. Below the beam, discontinuous films that form new conductive a bottom electrode has been patterned by paths as the grains expand [8, 9]. lift-off. Micro- and nano-electromechanical resonators excel in the field of sensing applications, showing high sensitivity, low noise susceptibility and precise readout. They are widely used in science and technology for detection of mass [10], temperature [11] and gas pressure [12]. With our design, the idea is to induce a change of stress on the beam through H2 exposure, thus benefiting from the strong natural phenomenon of H2 induced Pd Fig. 1 > Schematic illustrating actuation, readout expansion. and sensing principle of the device [1]. In this report we investigate a few strategies to improve the quality of the readout signal in order to enhance the sensitivity and By applying an alternating voltage between robustness of the device. the beam and the bottom electrode, the beam is brought into resonance. The resonance Background of experiments frequency is measured by monitoring the transmitted signal between the beam and the The electrostatic interaction between the bottom electrode. The electrical equivalent beam and the bottom electrode can be of this configuration is a RLC branch [2], approximately described as a parallel plate representing the mechanical motion, in parallel capacitor interaction. In this case, the with a capacitance, representing the parallel electrostatic force is given by 31
- 32. nanoICT energy losses due to drag forces are higher. To find out how critical this loss mechanism where is the permittivity, A is the area, is compared to others, such as clamping d is the distance and V is the voltage. In losses, we made a comparative experiment. our configuration an alternating voltage is In addition, we studied the spring softening ( ) super-positioned effect, which is caused by the fact that the with a direct voltage ( ). The voltage term electrical force becomes stronger as the can be developed inter-electrode distance is decreased during actuation. It is approximated as , leading to a decrease of resonance frequency according to . Given that the resonating motion originates from the product Not surprisingly, the direct voltage gives a of Vac and Vdc, the resonance frequency is component which does not vary in time and is expected to decrease linearly with respect to redundant with respect to dynamic actuation. Vdc, giving . The alternating voltage renders variation of force that varies at 2f, twice the frequency of Results the electric signal frequency. The combined Single vs. differential measurements term gives a force component that varies at In Figure 3, the difference in feed-through f, the same frequency as the electric signal between single and differential measurement frequency. We can thus either actuate at f configuration is illustrated. We found that while simultaneously measuring the response in this case, the magnitude of the feed- at f, using the latter term, or actuate at f and through decreased by more than an order of measure the response at 2f, using the prior magnitude. term. Which of these strategies is better depends on the characteristics of the device. A common problem is that the feed-through signal (the signal going through Cp) is so large that the resonance peak is impossible to measure. In this case, that feed-through signal will be substantially decreased by applying an actuation voltage at fres/2 while measuring at the doubled frequency. Another approach to reduce feed-through is to connect two identical devices in parallel, while only applying on one of them. Equivalent alternating voltages are applied to Fig. 3 > Comparison between single and the two devices, however 180° phase shifted. differential configuration (amplitude Immediately after the devices, the paths measurement). reconnect and the feed-through signals cancel out each other due to the difference in phase. As the differential configuration turned out to Any remaining signal close to resonance be very successful, we used it also for the frequency is caused by the mechanical motion most experiments. of the resonator. This approach is called a differential measurement setup. Spring-softening effect Measurements at atmospheric pressure Figure 4 illustrates what happens when we 32 differ from those made in vacuum in that the increase the Vdc. We see that the magnitude of
- 33. nanoICT the response increases as Vdc is increased. In vacuum clearly responds stronger. At Vdc=30 addition the resonance frequency decreases V, the beam in atmospheric pressure shows due to spring-softening. a 1.5 μV magnitude change at resonance whereas the beam in vacuum changes by 5 μV. Fig. 4 > Differential measurement in atmospheric pressure, illustrating the increase in magnitude and spring-softening effect as Vdc is increased. The spring softening is approximated as , where . Given that the resonating motion originates from the product of Vac and Vdc, the resonance frequency is expected to decrease linearly with respect to Vdc. Based on a linear relationship ( ), the following coefficients were determined ∆Vdc [V] ∆fres [kHz] a [kHz/V] 20-15=5 0 0 25-20=5 -12.5 2.5 30-25=5 -12.5 2.5 Fig. 5 > Same device a) measurement in vacuum 35-30=5 -12.5 2.5 and b) measured in atmospheric pressure. The change of resonance frequency is thus 2f measurements approximately linear, but changes are very close to the step size frequency increase, The result of a 2f measurement is shown in which limits the precision. Figure 6. The most important difference is that the background has been attenuated by Atmospheric pressure vs. vacuum close to 2 orders of magnitude. The signal- Figure 5a illustrates measurements made to-background ratio is much more favorable. under vacuum. Figure 5b illustrates However, the output signal is also attenuated. measurements on the very same device but As noise is more visible, we find that the under atmospheric pressure. We find that for signal-to-noise ratio is worse as compared to equivalent electrostatic forces, the beam in measurements at f. 33
- 34. nanoICT Systems Ii-Express Briefs, 2007. 54(5): p. 377- 381. [3] Baselt, D.R., et al., Design and performance of a microcantilever-based hydrogen sensor. Sensors and Actuators B-Chemical, 2003. 88(2): p. 120-131. [4] Hu, Z.Y., T. Thundat, and R.J. Warmack, Investigation of adsorption and absorption- induced stresses using microcantilever sensors. JOURNAL OF APPLIED PHYSICS, 2001. 90(1): p. 427-431. [5] Okuyama, S., et al., Hydrogen Gas Sensing Using a Pd-Coated Cantilever. Jpn. J. Appl. Phys., 2000. 39: p. 3584-3590. Fig. 6 > 2f measurement. [6] Kiefer, T., et al., A single nanotrench in a palladium microwire for hydrogen detection. Nanotechnology, 2008. 19(12). Conclusions [7] Kiefer, T., et al., Large arrays of chemo- We have tested several different methods to mechanical nanoswitches for ultralow-power improve the signal quality of an electrostatically hydrogen sensing. Journal of Micromechanics actuated MEMS device. We report that the and Microengineering, 2010. 20(10). differential measurement configuration is an [8] Xu, T., et al., Self-assembled monolayer- effective way to improve the readout signal. A enhanced hydrogen sensing with ultrathin weak spring-softening effect was observed. palladium films. APPLIED PHYSICS LETTERS, Comparisons between measurements in 2005. 86(20). vacuum and in atmospheric pressure showed [9] Kiefer, T., et al., The transition in hydrogen that media-related damping (squeeze sensing behavior in noncontinuous palladium damping) is dominant, but the damping films. APPLIED PHYSICS LETTERS, 2010. caused by the clamping is also considerable. 97(12). The 2f measurement readout scheme [10] Naik, A.K., et al., Towards single-molecule attenuated the background very strongly, as nanomechanical mass spectrometry. Nature expected, but it is not clear if this is helpful as Nanotechnology, 2009. 4(7): p. 445-450. the signal-to-noise ratio is also decreased. [11] Pandey, A.K., et al., Performance of an AuPd micromechanical resonator as a temperature As an outlook, the next step in improving the sensor. APPLIED PHYSICS LETTERS, 2010. readout of the sensor would be to design and 96(20). fabricate a device with three terminals, so that [12] Huang, X.M.H., et al., Nanomechanical driving and reading can be separated more hydrogen sensing. APPLIED PHYSICS efficiently. LETTERS, 2005. 86(14). References [1] Henriksson, J., L. G. Villanueva Torrijo, and J. Brugger. Ultra-low power palladium-coated MEMS resonators for hydrogen detection under ambient conditions. in Transducers ‘11. 2011. Beijing: IEEE. [2] Arcamone, J., et al., A compact and low- power CMOS circuit for fully integrated NEMS 34 resonators. Ieee Transactions on Circuits and
- 35. nanoresearch Optical analysis (study) of InAsP/InP core shell nanowires Fauzia Jabeen1, Bernt Ketterer2, Gilles optimization of the growth parameters is Patriarche1, Anna Fontcuberta I Morral2, required. In a previous study, we observed Jean-Christophe Harmand1 the absence of stacking faults in the 1 CNRS - Laboratoire de Photonique et de nanowire section where InAsP was inserted. Nanostructures, Route de Nozay, 91460 This observation gives a hint of a positive Marcoussis, France. role of the arsenic flux which was supplied 2 Laboratoire des Matériaux Semiconducteurs, for InAsP growth in suppressing such crystal Institut des Matériaux, Ecole Polytechnique Fédérale de Lausanne, CH-1015 Lausanne, phase mixing. Switzerland. Here we present a systematic study on fauzia.jabeen@lpn.cnrs.fr MBE-grown InAsxP1-x NWs. We examine the effects due to variations of As content on Abstract > InAsP/InP core-shell their crystal structure and optical properties. nanowires (NWs) with a systematically In order to get good photoluminescence increasing As content are grown by (PL) efficiency the surface states need to molecular beam epitaxy. The As content be passivated. This can be obtained by in the ternary core part of these NWs is growing InP shells around the initial InAsxP1-x varied in pursuit of obtaining stacking cores. It is well-known that the crystal fault free InAsP/InP core-shell NWs and structure of the shell follows that of the core; μ – PL analysis is carried out to make a i. e. crystal phase of the core-shell (CS) correlation between the improvement NW is determined during core growth and in the crystal structure of NWs and their remains unchanged during shell formation. optical properties. Therefore growth parameters are optimized for stacking faults free InAsxP1-x core NWs which are subsequently wrapped with InP. Introduction These CS NWs will have better PL efficiency Most of the III-V materials exhibit cubic as reported for other NWs systems (Jabeen, zinc blende (ZB) crystal structure in bulk, 2008), (Wu, 2003). whereas one dimensional (1D) structures of the same materials (NWs) often turn out to be hexagonal wurtzite (WZ) crystals with Experimental details stacking faults, twins and intermixing of ZB InAsP/InP CS NWs were grown by molecular and WZ sections. These crystal defects, beam epitaxy on InP substrates by vapor- randomly occurring along the length of the liquid-solid mechanism with Au used as NWs, undermine the optical and electrical catalyst. A thin InP buffer layer was grown on properties of NWs (Akopian, 2010), (Minot, the epi-ready InP substrate for an atomically 2007). Therefore, for the realization of smooth surface. The growth chamber is efficient NWs based devices, it is essential equipped with an in-situ Au effusion cell. to grow either single crystal structure NWs This permits to deposit a controlled amount or controlled heterostructures of ZB and of catalyst on the as-grown buffer layer, WZ sections. Presence of stacking faults in with no risk of surface contamination by ex molecular beam epitaxy grown InP NWs has situ manipulation. Details about the growth already been reported (Tchernycheva, 2007). parameters, morphology and structural To suppress the stacking faults, a thorough analysis can be found elsewhere (Jabeen, 35
- 36. nanoICT 2011) here we will focus only on the optical a studies which were performed on these NWs with a brief description of the NWs structure for two extremes i. e. grown with lowest and highest As in the core. For the low-temperature micro photoluminescence (μ - PL) experiments NWs were transferred mechanically on a Si substrate with markers defined by lithography for easy spotting of individual NWs. This experiment is performed at 10K in a liquid helium flow cryostat by illuminating b the sample with Ar+Kr+ or HeNe lasers with wavelengths respectively 647.1nm and 632.8nm and collecting the signal in a Si detector. Laser spot size is <600 nm. In Fig. 1 scanning transmission electron microscopy (STEM) images acquired by SEM are presented for two extremes. These images are acquired by transferring the NWs on the TEM grids. Fig. 1(a) is InAsP/InP CS NWs with lowest As content in the core and Fig. 1 > (a) Scanning transmission, electron these NWs exhibits stacking faults all along microscopy (STEM) images of the InAsP/InP CS the NW. For this low As content no CS NWs. In (a) CS NWs grown with lowest As/P ratio formation is observed. EDS performed on exhibiting stacking faults all along the NW length. these NWs results with no As incorporation In (b) NW from the sample grown with highest As/P ratio and resulting NW with a clear core of in these NWs or it is the below the detection 10 – 15 nm diameter with a shell around but no limit of system used for this experiment. In stacking faults arte observed. Fig. 1(b), bottom section of a single NW grown with highest As content is shown. CS formation is clearly visible in this image related to shell for all the NWs and at the where contrast in the middle part of the tip, where only shell related emission is NW indicates core part with a diameter of expected. Hence for all the analyzed NWs, ~10 – 15 nm. No contrast along the NW for comparison, spectra acquired at the base observed for the whole length indicating the are taken into account. suppression of the stacking faults in these NWs. μ – Photoluminescence In Fig. 2 a sketch of the CS NW is presented to describe how the PL experiment was μ – PL spectra were acquired on several performed. In order to distinguish that the single NWs for each sample grown with acquired spectra is from the base or the tip of different As content in the InAsP core part. the NW, NWs are measured at three different First we focus on the PL spectra related to the points. At the base, it is expected to have a InP emission energy range. μ – PL spectrum contribution from the core as well as the shell, recorded for a single NW with lowest As is at the middle part, where depending on the shown in black color in Fig. 3(a). Broad PL length of the NW and point of detachment spectrum lies within the range of emission from the substrate, the spectra can have the energy for InP with mixed crystal structure 36 contribution from core or not and emission [(Jancu, 2010), (Pemasiri, 2009), (Bao,
- 37. nanoICT for this sample. All the NWs measured from this samples emits at this energy value indicating highly Fig. 2 > Sketch of CS NW in the top part indicating homogeneous three parts where PL spectra are acquired and NWs. In Fig. 3(b) full width half maximum expected emissions related to core and shell part. (FWHM) for these spectra is plotted. Decrease in FWHM with increasing As content in the core of these NWs gives a 2008), (Titova, 2007)]. The splitting of the direct evidence of improvement in the crystal peaks, which is the reasons for the emission quality of these NWs. peaks at various energy values, indicate the presence of randomly varying ZB and WZ sections thicknesses within the InP emission a energy range as reported by Jancu and co authors [(Jancu, 2010), (Pemasiri, 2009)]. Other NWs studied from the same sample (not shown) emits in the similar wavelengths range and spectra were always composed of several peaks with different weights within the energy range related to InP. Red color in Fig. 3(a), represent the μ - PL spectra from the sample grown with 0.38 As/P ratio. This spectrum is comprised of a broad peak centered at 1.447 eV along with shoulders at higher energy corresponding to emission from InP. PL spectra acquired on NWs from this sample emits within this energy range but contains many peaks. b μ – PL spectra obtained for the NWs grown with 0.51 As/P ratio are represented by green color in fig 3(a). Since these NWs show fewer stacking faults compared to previous samples (not shown here) main peak corresponding to InP is at 1.442 eV with a shoulder at 1.48 eV. Several NWs measured from this sample exhibits less inhomogeneity. Recorded PL spectra show some sharp lines distributed between this energy ranges suggest the insertions of certain number of monolayer of WZ and ZB segments for all measured NWs from this sample. Finally, sample with Fig. 3 > μ - PL spectra acquired on single NWs for highest As content and no stacking faults each sample. Broader PL emission (black line) for as shown in Fig. 1(b) is measured. Blue line lowest As/P content indicates the phase mixing, in Fig. 3(a) represent PL spectra from a NW whereas decreasing FWHM of the PL peaks with from the sample with highest As content. increasing As content indicate an improvement in An emission peak, at 1.43 eV, is recorded the crystal quality as presented in (b). 37
- 38. nanoICT Concerning the PL emission related to the core of the NWs, acquired spectra at the base of all the NWs are plotted in Fig. 4. The spread of the emission energy is between 1.35eV – 1.221 eV which indicate an increase in the As content in the core of these NWs. For the emission from NWs grown with 0.38 As/P ratio, shown in Fig. 4 in red line, a broad peak is observed. The possible reason of this broad peak could be the variations in the thicknesses of WZ and ZB segments which results in the emission at several energies value within a Fig. 4 > μ - PL spectrum corresponding to the InAsP core with increasing As/P ratio. Blue shift in certain range. From one NW to other NW, energy indicate an increase in the As content in shift in observe peak position shift within the InAsP core. In the top left an inset regarding the same energy range. For few NWs, the FWHM of these emission peaks is plotted as sharp peaks are observed, indicating either a function of the As/P ratio. Decreasing FWHM repetition of ZB and WZ section lengths or indicate improvement in overall crystal quality of observed NWs. non homogeneous As incorporation in the core. This splitting is not a general trend for all the measured NWs for this sample and Discussion it could be due to the presence of a closely This series of InAsP/InP CS NWs indicate attached another NW. In Fig. 4, green line that by systematically increasing As content represents the PL emission for NWs grown in the InAsP core, phase mixing can be with 0.50 As/P ratio. Here again a broad suppressed completely. In this series for peak is recorded with an emission between As/P ratio of 0.50 NWs with fewer stacking 1.35 eV – 1.25 eV. For all the measured faults are obtained whereas further increase NWs from this sample, as in case of InP in the As/P ratio give stacking faults free shell for this sample, emission peaks are NWs. This suppression of stacking faults at the similar energies. This indicates a is demonstrated by μ – PL analysis carried step towards homogeneity in NWs but out for each of the sample and several the broad spectra indicate presence of NWs from each sample are studied. The stacking faults. NWs sample with highest NWs rich in stacking faults (Fig. 1a) show a As content, plotted in blue line in Fig. 4, broad μ – PL spectra showing large degree emit at 1.217 eV with all the measured of inhomogeneity, while NWs grown with NWs emitting around this energy. This As/P ration of 0.60 exhibit greater degree of homogeneity in the emission energy has homogeneity (Fig. 1b) with a considerable also been observed for the shell of these decrease in the width of PL peak. This NWs and it gives a further weight to the strongly suggest that increasing As content fact that these NWs are of good crystal modifies the crystal structure and yields quality. Another indication is the decrease defect free NWs. Red shift in the emission in FWHM of the emission spectra with energy with increasing As content indicate increasing As content in the core. In the set the higher As incorporation in the InAsP at the top left of the Fig. 4, FWHM for these core whereas the decrease in the FWHM samples related to the core emission PL is the evidence of the good crystal quality 38 peaks is plotted. of these NWs.
- 39. nanoICT A red shift for the emission related to InP shell is observed Fig. 3(a). There are two possible explanations for the red shift in the emission energy related to InP, with increasing As content. Firstly, it could be due to the two formation of InAsP shell with parasitic As present in the chamber at the time of the shell growth. Short growth time and low growth temperature for shell enhance As incorporation. The trend of this red shift fit well with increasing As content in the core which implies increased amount Fig. 5 > μBand gap energy of ternary InAsxP1-x of residual As in the chamber at the time compound plotted against x ranging between 0 of shell growth for respective sample. and 1. Red dots with a shift close to difference of Secondly, this shift could be related to the ZB and WZ InP band gap energy representing the stain present in the InP shell around InAsP emission for shell above 1.42 eV and corresponds core and increased As in the core result to core emission between 1.38 eV and 1.22 eV. increasing strain in InP shell. Since the diameter of the inner core is ~10 nm and the shell around these NWs is 15 – 20 nm Conclusion thick, such red shift only due to strain is not In conclusion we have reported the growth envisioned. of stacking fault free InAsP/InP CS NWs In Fig. 5, band gap energy for ZB InAs1-xPx confirmed by their μ – PL analysis. Increase ternary is plotted in black for x between in As content in the core part results in 0 and 1 [11]. Peak PL emission energies suppression crystal imperfection and a corresponding to InP shell plotted in blue shift in PL emission energy related to Fig. 3 are inserted. Band gap 1.49 eV is the core emission indicate a systematically considered for the WZ InP (Jancu, 2010), increasing As content in the core part. (Pemasiri, 2009). InP emission from sample with As/P ratio of 0.60 which exhibits peak at 1.43eV is inserted with 60 meV difference References from the WZ InP and it results with ~ 5% [1] N. Akopian. Et al, Nano Lett., 10. 1198 (2010). content in the shell. This value, for the [2] E. D. Minot, et al, Nano Lett., 7. 367 (2007). parasitic incorporation of As, is quite high [3] M. Tchernecheva et al., Nano Lett. 7 (2007) so strain should have some effect for this 1500. sample. For the samples grown with lower [4] F. Jabeen et al., Appl. Phys. Lett. 93 (2008) As/P ratio the shift in the emission peak 083117. with respect to the WZ InP gives around 2% As incorporation in the shell. [5] G. Wuet al., Chem. Phys. Lett., 378 (2003) 368. [6] F. Jabeen et al., to be published, (2011). An increasing As incorporation trend is [7] J.-M. Jancu et al., Appl. Phys. Lett. 97 (2010) observed and the PL emission give 11%, 041910. 13% and 21% incorporation of As in the ternary core for the samples grown with [8] K. Pemasiri et al., Nano Lett. 9 (2) (2009) 648. 0.38, 0.50 and 0.60 As/P ratio respectively. [9] J. Bao et al., Nano Lett. 8 (3) (2008) 836. For the sample grown with 0.50 As/P ratio [10] L. V. Titova et al., Nano LettL 7 (11), (2007) 3383. nominal As in the core measured by EDS [11] Brigham Young University (BYU), “Energy gap is around 8%. So this additional shift could in III-V ternary semiconductors”, http://www. also be related to the strain. cleanroom.byu.edu/EW_ternary.phtml. 39
- 40. nanojobs nanoconfs • Master + PhD Position (CEA, Leti, • Architecture & Design of Molecule Logic Grenoble, France): “Development of XRR and Gates and Atom Circuits (AtMol conference GIXRF combined analysis” series) The micro and nano electronic world is January 12-13, 2012. Barcelona (Spain) experiencing a revolution in order to tackle challenges of miniaturization, power consumption, http://atmol.phantomsnet.net/ power density and processing speed of CMOS Barcelona2012_index.php?project=7 devices. There is now a critical need for metrology Atomic Scale Technology to give quantitative chemical composition measurement of new materials with buried • Nanospain2012 interfaces and with nanometre depth resolution. February 27 - March 01, 2012. Santander The deadline for submitting applications is (Spain) January 14, 2012 www.nanospainconf.org/2012/ For further information about the position, please contact: NanoBiotechnology, NanoChemistry, Nanotechnologies Emmanuel Nolot (emmanuel.nolot@cea.fr) • PhD or PostDoctoral Position (RWTH Aachen University, Germany): “Semiclassical Simulation of CNT- and Graphene FETs” The deadline for submitting applications is January 15, 2012 For further information about the position, please contact: • nanotech 2012 Christoph Jungemann (christoph.jungemann@ithe.rwth-aachen.de) February 15-17, 2012. Tokyo (Japan) www.nanotechexpo.jp/en/ • Master + PhD Position (CEA Grenoble, Nanotechnologies, Nanotechnology Business INAC, SPSMS, LaTEQS, France): “Non- adiabatic silicon electron pumps” • Bionanotechnology III: from An experimental investigation of a new kind of biomolecular assembly to applications electron pumps is proposed. Electron pumps 04-06 January 2012. Robinson College, are two-terminal devices transferring electrons Cambridge (UK) one by one, hence producing a quantized current I=ef for a driving frequency f. These devices are at w w w. b i o c h e m i s t r y.o r g / t a b i d / 379 / the heart of a future redefinition of the S.I. unit for MeetingNo/SA121/view/Conference/ electrical current, the ampère, which would, like default.aspx other units, be related to fundamental constants Nanomaterials, Nanobiotechnology (here the charge e of an electron). The deadline for submitting applications is • BioNanoMed 2012: Nanotechnology in January 18, 2012 Medicine & Biology For further information about the position, March 01-02, 2012 ,Austria (Spain) please contact: www.bionanomed.at/ 40 Xavier Jehl (xavier.jehl@cea.fr) Nanobiotechnology, Nanomedicine
- 41. nanoresearch Brillouin light scattering measurements in crystallographically tuned thin Co-films O. Idigoras1, B. Obry2, B. Hillebrands2 and A. depositing different amounts of Si-oxide, we Berger1 have obtained a set of samples with different 1 CICnanoGUNE Consolider, Tolosa Hiribidea 76, degree of crystallographic order. Moreover, E-20018 Donostia-San Sebastian, Spain. we observe an anomalous magnetic 2 Fachbereich Physik and Landesforschungzentrum reversal near the HA in these samples above OPTIMAS, Technische Universität Kaiserslautern, a threshold level of crystallographic disorder, Erwin-Schrödinger-Straße 56, D-67663 while behavior in all other field orientations is Kaiserslautern, Germany. barely affected. This anomaly arises from a competition of the misalignment anisotropy We report an experimental study of spin axes with the exchange energy and it waves in thin Co-films with in-plane uniaxial gives rise to an unusually high remanent symmetry, which were measured by magnetization and coercive field along the means of the Brillouin light scattering (BLS) hard axis [1]. technique. In particular, we investigated the effect of the previously discovered hard axis The Brillouin light scattering (BLS) technique (HA) anomalous magnetization state that [2] is a powerful method to analyze spin occurs during magnetization reversal in waves and related magnetic properties. partially disordered films. This work has been Primarily, it allows for the analysis of spin performed in Prof. Dr. Burkard Hillebrands’ waves energies by means of the frequency group where all members of team were shift of inelastically scattered photons from really nice in supporting all the activities. a magnetic sample. Photons interact hereby Furthermore, all have been possible thanks with spin waves, so that a spin wave is to Phantoms Foundation’s Nano-ICT project created (Stokes process) and the photon launches exchange visit fellowship. looses energy correspondingly or a spin wave is annihilated (anti-Stokes process) whereupon the photon gains that energy. Introduction Experimentally, the frequency shift of the In previous work we have demonstrated scattered light is detected to measure the that is possible to tune the degree of corresponding spin wave energy. In this crystallographic order in Co-films by partial work we have analyzed Stokes scattering interruption of epitaxy in a well-defined processes only. and reproducible manner [1]. Hereby, we In general, different spin wave modes utilized an optimized growth sequence to can be present in a thin film [3]. The main achieve good epitaxy and a high degree of distinction arises from the type of interaction crystallographic order as a starting point. that dominates them, which is either the Specifically, we produced epitaxial Co-films exchange or the dipolar interaction. There with an in-plane hcp c-axis by growing the exist different dipolar dominated modes, sequence Ag 75 nm/Cr 50 nm/Co 30 nm/ which can be bulk or surface spin wave SiO2 10 nm onto hydrofluoric etched Si (110) modes. The surface mode, called as substrates. For the tunable disturbance of magnetostatic surface mode (MSSM) or the growth sequence, we introduced an Damon-Eshbach (DE) mode are excited ultrathin Si-oxide layer of defined thickness in when the magnetization M and the wave the order of a single monolayer on top of the vector lie both in the film plane and are Si-substrate prior to the Ag-film growth. By perpendicular to each other. Even in this case 41
- 42. nanoICT when the magnetization and wave vector about the spin wave damping and thus on are perpendicular bulk modes can be the magnetization homogeneity. The peak excited, in thin films it is not possible to at -27 GHz corresponds to the dipolar spin distinguish between both modes and only wave mode (MSSW/DE), while the other two MSSM/DE mode are considered. On the peaks corresponds to exchange spin wave other hand, if both are in-plane and parallel, modes (PSSW). Measurements as the one magnetostatic backward volume modes shown in figure 1 but for multiple applied field (MSBVM) are excited. For the same in- strengths are shown in figure 2 as a color plane wave vector the DE/MSSM mode map. Figure 2(a) shows the data collected has higher energy and thus frequency than when the external field is applied along the the MSBVM mode. The exchange type EA, and fig. 2(b) when the field is along the perpendicular standing spin waves (PSSW) HA. The red color indicates the maximum are formed by a superposition of two spin intensity, i.e. the spin wave positions, while waves that are propagating in opposite the blue color indicates the minimum directions perpendicular to the film surface. intensity, i.e. noise level. In both figures two In this work we have analyzed all three spin spin wave modes appear, one belonging to wave modes, DE/MSSM, MSBVM and the MSSW/DE at lower absolute values of PSSW. frequency and one representing the PSSW at higher absolute frequency values. As The experimental setup that was utilized in expected in the EA (fig. 2(a)), both spin wave this study is equipped with a tandem Fabry- frequency vs. field curves have the same Pérot interferometer [4]. Moreover, this type of behavior. When the field is applied setup is able to perform an automated in- along the HA (fig. 2(b)), however, the dipolar plane rotation of the sample. In this work, we type spin wave shows a more pronounced used backward scattering geometry, and all frequency shift for small applied field than the measurements have been done with an in- PSSW mode. The origin of this frequency shift plane applied field orientation perpendicular is the in-plane rotation of the magnetization to the wave vector. into the direction of the anisotropy axis as the field strength decreases, causing the spin wave to change its character from the Influence of the HA anomaly on spin DE mode to MSBV mode, which has a lower waves spectra frequency than the DE mode. This frequency In order to analyze the effect that the HA shift for fields applied along the HA has been anomaly has on the spin wave spectra previously reported in several works [5]. we have measured the dependence of the spin wave frequencies as a function of the applied field strength for (i) an epitaxial sample, in which the anomaly is not present, and (ii) in a sample with a 0.132 nm thick Si-oxide underlayer, for which the anomaly does occur. One example of such spin wave spectra for the epitaxial sample is presented in figure 1. Apart from the elastically scattered light peak at zero frequency, three peaks are clearly visible that correspond to inelastically scattered light from three spin wave modes. While the position of these peaks gives the frequency of the spin wave, Fig. 1 > Spin wave spectrum for an epitaxial Co 42 the widths of the peaks contain information (1010)-film sample.
- 43. nanoICT field orientations around the nominal hard axis [1]. Fig. 2 > (a) and (b) show the spin wave frequency dependence from an externally applied field along the EA and HA, respectively, for an epitaxial Co (1010)-film sample . A similar set of measurements for a sample, in which the HA anomaly is present (the sample with Si-oxide underlayer of 0.132 nm thickness), are shown in figure 3. Specifically, this figure shows spin wave frequency vs. Fig. 3 > (a), (b) and (c) show the spin wave field dependence for the applied field along frequency dependence from an externally applied the EA (fig. 3(a)), along the HA (fig. 3(b)), and field along the EA, HA and 2° away from the HA, for a field direction 2° away from the HA (fig. respectively for a uniaxial, but slightly disordered 3(c)). Although the dipolar and the exchange sample. type spin waves are close in frequency for measurements along the EA, so that it If the field is applied 2° away from the HA, is difficult to clearly see their separation, a conventional HA spin wave behavior both spin waves visibly follow the same reemerges, with the dipolar type spin wave frequency behavior as a function of the frequency shifting towards lower frequencies applied field, just as in the case of epitaxial upon decreasing the applied field strength. sample. However, the expected downward frequency shift for the dipolar type spin For this sample, the saturation magnetization wave upon reducing the externally applied is obtained at smaller external field strengths field along the HA does not occur here (fig. than in the case of the epitaxial sample, so 3(b)). Only a broadening is observed that that the position shift of this mode is limited can arise due to an inhomogeneous sample to a narrower applied field range. magnetization. Such an inhomogeneous For the same sample, we have also magnetization state was indeed observed performed applied field angle β dependent earlier on the same sample by means of measurements in the vicinity of the HA Kerr effect microscopy in a narrow range of (β=90°) in remanence. Figure 4(a) shows 43
- 44. nanoICT spin wave spectra vs. β in remanence after I acknowledges Phantoms Foundation saturating the sample in 1500 Oe for each for Nano-ICT project launches exchange angle and figure 4(b) shows at the same visit fellowship, as well as, all members type of measurment in remanence, but of Prof. Dr. Burkard Hillebrands’ group after saturating the sample only once along at Fachbereich Physik at Technische the EA. In the first measurement (fig. 4(a)) Universität of Kaiserslautern and specially one can see that due to the anomalous Prof. Dr. Burkard Hillebrands for giving me magnetization reversal in the HA, β=90°, the opportunity to work in his group and the dipolar type spin wave is shifted towards Björn Obry with whom I have performed all the measurements and helps me higher frequencies. One can also see that with everything. I also thank the Basque a misalignment of only ±1° away from the Government for fellowships No. BFI09.284. HA already causes an almost complete suppression of this anomalous behavior and the reappearance of the typical dipolar spin References wave at low frequencies of about 7 GHz. [1] O. Idigoras, A.K. Suszka, P. Vavassori, P. In the second measurement series of the Landeros, J.M. Porro and A. Berger, submitted same sample after a single EA saturarion to Phys. Rev. B. (fig. 4(b)), no anomaly is visible, as expected [2] O. Gaier, 2009. A study of exchange interaction, because in this case we only measure magnetic anisotropies, and ion beam induced the EA projections of the same uniform effects in thin films on Co2-based Heusler magnetization state along the applied field compounds. Thesis, (PhD). Technischen direction. Universität Kaiserslautern. [3] B. Hillebrands, Brillouin light scattering from layered magnetic structures, in M. Cardona, Güntherodt (Editors), Light Scattering in Solids VII, vol. 75 of Topics in Applied Physics, Springer Verlag, Berlin Heidelberg (2000). [4] J.R. Sandercock, Opt. Comm. 2, 73 (1970). B. Hillebrands, Rev. Scien. Instr. 70, 1589 (1999). [5] R. Scheurer, R. Allenspach, P. Xhonneux and E. Courtens, Phys. Rev. B 48, 9890 (1993). M. Grimsditch, E.E. Fullerton and R.L. Stamps, Phys. Rev. B 56, 2617 (1997). Fig. 4 > Spin wave frequency as a function of the applied angle in remanence for a slightly disordered sample, after (a) prior saturation at every angle and (b) after prior saturation along 44 the EA.
- 45. nanoresearch Temperature distribution of heated membranes for stencil lithography application Shenqi Xiea, Damien Ducatteaub, Bernard Legrandb, Veronica Savua, Lionel Buchaillotb a and Juergen Bruggera a Microsystems Laboratory (LMIS-1), Ecole Polytechnique Fédérale de Lausanne, Switzerland. b Institut d’Electronique, de Microélectronique et de Nanotechnologie, UMR CNRS 8520, IEMN, Avenue Poincaré, B.P 69, 59652 Villeneuve d’Ascq . Cedex, France. b 1. Introduction Stencil lithography (SL) is a resistless lithography method for surface patterning with sub-micron resolution. As the most c d conventional application of SL, thin-film deposition has become a reliable micro/ nano-patterning process [1, 2]. However, the useful life time of the stencil during one pump-down is limited by the clogging of the aperture due to the deposited material [3]. We recently developed a novel approach to potentially prevent and eventually eliminate clogging by locally heating up the stencil during metal deposition, minimizing thus materials’ condensation on the membrane Fig. 1 > Schematics of (a) frontside and (b) [4]. The heatable stencil has Pt microhotplates backside of the heated stencils. (c) Optical image embedded in two layers of SiN thin film, with of the device and (d) SEM image of the apertures stencil apertures in between the coils, as in between the coils. shown in figure 1. In our previous experiments, the thickness between temperature and condensation rate, of condensed metal film is correlated to the which can be translated into clogging rate. temperature distribution on the membrane. In addition, the temperature coefficient of The area with least material condensation is resistance (TCR) can be extracted from the in the center of the heated membrane, where temperature mapping under certain input has the highest temperature. As temperature power in ambient conditions. As the heated drops rapidly from the center to the border stencil will be placed in the vacuum chamber of the membrane, metal starts accumulating of an E-beam evaporator, the temperature very quickly during deposition, which of course of the membrane can only be calculated increases clogging rate. Therefore, it is critical from the variation of the resistance of the to know the precise temperature distribution microhotplate based on the measured TCR on the membrane in order to study the relation in order to monitor the process. By using the 45
- 46. nanoICT high resolution (3 μm) InSb IR microscope in IEMN, we are able to measure the precise temperature in the area close to the stencil apertures, which provides us the feasibility of studying the dependence of temperature on clogging rate. 2. Infrared measurement The microhotplate was powered below the IR microscope for recording thermal images. Various designs of the coils with different resistance were measured. Figure 2(a) shows the thermal image of one of the designs taken under 1.0 V bias. The highest temperature appears in the centre of the microhotplate, which agrees well with the results from simulation. Several measurements were performed under different input power, and the temperature extracted from each image was used to calculate the TCR. Since the TCR for thin Pt film is different from its bulk form, this calibration is necessary. Figure 2(b) shows the measured temperature with corresponding resistance. Due to the limitation of the detection range, the highest measurable temperature is around 430 ºC. The TCR acquired from the linear fit is 1.947e-3/K. This calculated value will be used to monitor the temperature during metal deposition in vacuum. Fig. 2 > (a) Thermal image of one of the coil Further analysis of the thermal image provides designs powered under 1.0 V bias. (b) The measured temperature with corresponding us the temperature profile of the membrane, resistance. The TCR extracted from the curve is which is not smooth as shown in the black 1.947e-3/K. curve in figure 3(a). The temperature on the SiN area in between the Pt coils drops dramatically. In fact, it is due to the difference However, the drawback is that the coated of emissivity on different materials. Therefore, Pt thin film increases the thermal conductivity SiN looks more transparent than Pt in the on the membrane. Thus, more heat was infrared range, which gives us a much lower dissipated from the membrane, which of temperature. In order to have a more accurate course decreases the average temperature. thermal mapping, thin layers of Pt were coated But still, it offers us the possibility of estimating on the frontside of the membrane, offering a the temperature around the stencil apertures more uniform surface in terms of emissivity. within an acceptable error. In addition, due to Measurements were done under the same the spiral layout of the electrode, a stronger conditions before and after Pt coating, as thermal coupling effect must happen in the shown in figure 3(a). The temperature profile middle of the membrane, which decreases 46 becomes much smoother after Pt coating. the temperature gradient in that area. The
- 47. nanoICT relatively uniform temperature distribution in coils with different resistance were measured, the centre of the membrane provides a stable which provides important information for thermal environment for the stencil apertures. optimizing the designs in future generations. The calibrated temperature coefficient of resistance (TCR) is extremely useful in monitoring and controlling the whole process. Temperature profiles were also studied to correlate with the clogging rate in different areas. Extra Pt thin layers were coated on the membrane for providing a more uniform surface in terms of emissivity, leading to a more accurate temperature measurement. The temperature distribution provides us great feasibility of studying the dependence of temperature on clogging rate. The achieved results from this exchange program would be combined with other results to be considered to submit to MEMS2012, following with other journals. References [1] M. A. F. van den Boogaart, et al., “Corrugated membranes for improved pattern definition with micro/nanostencil lithography”, Sensors and Actuators A, vol. 130-131, p. 568-574, 2006. [2] V. Savu, et al., “Dynamic stencil lithography on full wafer scale”, Journal of Vacuum Science and Technology B, 26(6), 2008. Fig. 3 > (a) The temperature profile of the heated [3] M. Lishchynska, et al., “Predicting Mask stencil with and without Pt coating on the frontside Distortion, Clogging and Pattern Transfer of the membrane. (b) The average temperature of for Stencil Lithography”, Microelectronic the membrane versus the applied power in ambient Engineering, 84 (2007) 42-53. conditions. [4] S. Xie, V. Savu, J. Brugger, “Heated membranes prevent clogging of apertures in nanostencil lithography”, The 16th International Another information can be derived from Conference on Solid-State Sensors, Actuators the measurement is the temperature versus and Microsystems (Transducers’11), Beijing, power consumption, as shown in figure 3(b). China. The power consumption is not critical in our preliminary stage of experiments, but it has to be considered in future development if more heated stencils are powered simultaneously. 3. Conclusions We have successfully carried out the thermal measurement on heated stencils by using the IR microscope in IEMN. Various designs of the 47