news & views
Figure 1 | The delivery and clearance of nanoassemblies. The nanoassemblies comprising core (gold) and satellite (green) nanoparticles are constructed using complementary DNA sequences as linkers (red) and coated with ligands such as PEG (black strands, blue clouds) to control their interactions with cells. When injected intravenously, they are uptaken via the EPR effect to the target site. Once inside the cell (blue), the assemblies exert their therapeutic effect and breakdown into their constituents for elimination.
have a gold core nanoparticle, which is surrounded by one or more layers of gold satellite nanoparticles. Each nanoparticle was modified with single-stranded DNA and linked together using DNA strands with complementary sequences. This design strategy improved tumour accumulation of the nanoassembly via passive targeting and the EPR effect, while significantly reducing uptake in the reticuloendothelial system. Finally, the supramolecular structure decomposed into the original building blocks, which promoted efficient elimination of each component into urine (3–5-nm satellite nanoparticles). This strategy improves the biodistribution and renal clearance of injected inorganic nanoparticles in living animals, while maintaining tumour-targeting efficiency. Thus, the DNAmediated core–satellite supramolecular nanoassemblies have the potential to replace the toxic and hydrophobic inorganic nanoparticles, such as quantum dots and carbon nanotubes, that are typically used in drug-delivery systems (Fig. 1). During the past decade, it has been shown that a number of important factors
mediate the ultimate fate of nanoparticles in vivo including solubility, stability, shape and flexibility, size and size distribution, and formulation2,5,6. The design of precise nanostructures is typically focused on developing site-specific delivery and targeting capabilities but biodistribution and clearance of the nanostructures from the neighbouring tissue and local environment are also critical. Thus, it is important to increase specificity, selectivity and safety 1, which are the fundamental principles that govern the clinical translation of nanoparticles7. In particular, these factors are crucial for determining physiological behaviours and pharmacokinetic parameters of nanoparticles in the body, as well as to ascertain their potential cytotoxicity and in vivo toxicity. The use of biocompatible or biodegradable supramolecular structures is essential to enhance renal clearance, and eventually reduce potential in vivo toxicity (Fig. 1)3. While taking steps towards overcoming the fundamental limitations of nanoparticles, namely efficient targeting and renal clearance, the DNA
nanoassembly developed by Chan and colleagues3 still needs improvement as a targeted probe. First, the biodistribution and targeting process is mediated by macrophage uptake and phagosomal accumulation, which increases nonspecific uptake and retention in the liver and spleen. To avoid this immune response, longer PEG chains were used; however, this eventually will increase the overall hydrodynamic diameter of the nanoassembly, preventing renal clearance of decomposed building blocks. Second, passive targeting via the EPR effect requires the long circulation time of nanoparticles in the body and the EPR effect is only exploited when the size of the nanoparticle is larger than 10 nm in hydrodynamic diameter. Therefore, there is a balance to strike between pegylation, which increases the size of the nanoparticle, and keeping the diameter of the nanoparticles within limits to allow renal clearance and prevent undesired nonspecific background retentions. These important trade-offs should be explored in future studies. ❐ Hak Soo Choi is at the Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts 02215, USA. e-mail: [email protected] References 1. Lee, J. H., Park, G., Hong, G. H., Choi, J. & Choi, H. S. Quant. Imaging Med. Surg. 2, 266–273 (2012). 2. Choi, H. S. & Frangioni, J. V. Mol. Imaging 9, 291–310 (2010). 3. Chou, L. Y. et al. Nature Nanotech. 9, 148–155 (2014). 4. Choi, H. S. et al. Nature Biotechnol. 25, 1165–1170 (2007). 5. Liu, J., Yu, M., Zhou, C. & Zheng, J. Materials Today 16, 477–486 (December, 2013). 6. Albanese, A., Lam, A. K., Sykes, E. A., Rocheleau, J. V. & Chan, W. C. W. Nature Commun. 4, 2718 (2013). 7. Choi, H. S. et al. Nature Nanotech. 5, 42–47 (2010).
Published online: 26 January 2014
NANOWIRE TRANSISTORS
Room for manoeuvre
Kinked nanowire transistors that can be manipulated in three dimensions can be used to record the intracellular electrical signals of targeted cells.
Ziliang Carter Lin and Bianxiao Cui
M
any biological cells use electricity to control their physical and chemical functions. Some cells do so by generating action potentials, which are rapid rises and falls in the electric potential difference across their 94
cell membrane. In particular, muscle cells, neurons and endocrine cells use action potentials to initiate mechanical contraction, perform signal processing and trigger chemical release, respectively. Furthermore, the propagation of action
potentials along axons and through synapses is the primary means for neurons to communicate over long distances. Action potentials are generated by the rapid opening and closing of ion channels on the cell membrane. Malfunctions of these ion
NATURE NANOTECHNOLOGY | VOL 9 | FEBRUARY 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
news & views channels are the cause of many neurological disorders, cardiac arrhythmias and druginduced toxicity in humans. Therefore, understanding the electrophysiology and pathology of these cells, from single-cell to circuit-network levels, requires accurate recording of their action potentials. Writing in Nature Nanotechnology, Charles Lieber and colleagues at Harvard University and Wuhan University of Technology now show that a platform that combines kinked nanowire transistors and three-dimensional positioning can be used to record intracellular action potentials of desired cells and cell regions1. Traditionally, action potentials are recorded using either extracellular or intracellular electrodes (Fig. 1a,b). Extracellular recording such as the chipbased multielectrode array method uses electrodes positioned close to the cell membrane to detect extracellular field potentials. Although this method is non-invasive and easily scalable, it lacks one-to-one correspondence between the cells and electrodes. The recorded field potentials also fail to represent the shape of the intracellular action potentials, thus making data interpretation difficult. On the other hand, intracellular recording such as patch-clamp recording provides accurate measurements of the action potentials, but this method is laborious and invasive, which limits the recording throughput and duration. Furthermore, the relatively large size of the patch pipette restricts its application primarily to cell bodies. In the past few years, significant progress has been made in the development of nanoelectrodes that aims to combine the advantages of intracellular and extracellular recording methods. These approaches are based on nanopillar 2, nanowire3 or goldspine4 electrodes, or field-effect nanowire transistors5 (Fig. 1c,d), and methods such as mechanical penetration, electroporation and chemical modification are employed to facilitate electrical access into the cell interior. Until now, most of these electrodes have been built on chip platforms, which allows parallel recording but provides little control over the position of the electrodes with respect to the targeted cells. As a result, only the cells that are in contact with the electrodes are recorded. Now Lieber and colleagues have created a freely movable kinked nanowire field-effect transistor for precise positioning and intracellular recording (Fig. 1e)1. The team has previously shown that, unlike conventional linear nanowire transistors, kinked nanowire transistors can have the source and drain on the same side of the nanowire5. In this latest work, each kinked
nanowire transistor is mounted at the end of a polymer shaft printed with source and drain lines, which are in turn connected to a printed circuit board. The whole assembly is mounted on a three-dimensional translational stage, which allows the nanoelectrodes to be accurately positioned with respect to the cells. The nanowire transistor is coated with phospholipids to facilitate mechanical insertion into the interior of the cell. Using the platform, the researchers achieve high-fidelity intracellular recording of cardiac muscle cells. In particular, action potentials and resting membrane potentials recorded with the approach are found to be almost identical to recordings by patchclamp probes. This suggests that the kinked nanowire transistors are capable of achieving tight seals with the cell membrane — a key element in realizing high-quality recording. Lieber and colleagues also demonstrate that the technique can clearly identify the change in action potential on treatment with various ion channel blockers.
Compared with chip-based nanoelectrodes, this new platform affords great freedom of recording from any desired cell location. It also records almost full action potential amplitude, which is significantly larger than the attenuated action potentials recorded by chip-based nanoelectrodes. Compared with the patch-clamp technique, which uses a similar positioning method and measures the full action potentials, the new platform uses a much smaller probe. The diameter of the nanowire (~80 nm) is an order of magnitude smaller than that of a pipette tip (~1 μm), which makes it much less invasive. Furthermore, unlike the single-use pipette probes, the nanowire probe can be repeatedly inserted and withdrawn from cells without the need of cleaning or re-coating but still possesses similar sensitivity. Looking ahead, the platform could provide a variety of new scientific opportunities. One of the most exciting potential applications is subcellular recording at axons, dendrites and even single synapses. c
a
b
z
d
y
x
e
z
y
x
Figure 1 | Intra- and extracellular recording methods. a, Conventional extracellular recording consists of planar electrodes (orange) measuring field potentials of a nearby cell (blue). b, Conventional intracellular recording involves three-dimensional positioning of a glass pipette (purple) to a target cell. c, On-chip metal nanoelectrodes have recently been developed, which use a variety of different nanostructures as electrodes2–4. d, On-chip kinked nanowire transistor electrodes have also been recently developed5. The nanowires consist of straight silicon segments connected at sharp angles, with the junction regions doped to create field-effect transistors. e, By creating free-standing kinked-nanowire transistors, three-dimensional positioning becomes possible with nanowire sensors.
NATURE NANOTECHNOLOGY | VOL 9 | FEBRUARY 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
95
news & views The good seal offered by the technique should also allow the measurement of subthreshold events such as excitatory and inhibitory postsynaptic potentials. However, similar to patch-clamp recording, the positioning method severely limits the scalability of the approach. Furthermore, nanowire transistors and nanoelectrodes in general will become more useful when they acquire the ability to inject stable current into
the cell. In comparison, current injection is a useful feature of the patch-clamp technique, which allows clamping of the cell membrane potential at predetermined levels to reveal the current–voltage relationship of specific types of ion channel. ❐ Ziliang Carter Lin is at the Department of Applied Physics Stanford University, Stanford, California 94305, USA. Bianxiao Cui is at the
Department of Chemistry, Stanford University, Stanford, California 94305, USA. e-mail: [email protected] References 1. Qing, Q. et al. Nature Nanotech. 9, 142–147 (2014). 2. Xie, C., Lin, Z. L., Hanson, L., Cui, Y. & Cui, B. X. Nature Nanotech. 7, 185–190 (2012). 3. Robinson, J. T. et al. Nature Nanotech. 7, 180–184 (2012). 4. Hai, A., Shappir, J. & Spira, M. E. Nature Methods 7, 200–250 (2010). 5. Tian, B. Z. et al. Science 329, 830–834 (2010).
MAGNETIC NANOSTRUCTURES
Vortices on the move
Magnetic vortices can be controllably transferred in an extended system by electrical means.
Teruo Ono
M
agnetic vortices are in-plane curling spin structures that have a nanoscale core with an out-of plane magnetization1. They are formed in ferromagnetic disks and the binary nature of the cores (whose magnetization can point up or down) combined with their thermal stability make vortices appealing for future non-volatile memory devices. Importantly for such applications, it has been shown that the cores can be switched by applying suitable electrical or magnetic field pulses2–5. A magnetic vortex is formed in the presence of a confining potential such as that provided by the geometrical confinement of a disk. However,
applications other than memory devices could be developed if magnetic vortices could be freed from such geometrical constraints. Towards this end, it has been shown that the Oersted field generated by passing an electric current into a ferromagnetic film through a metallic nanocontact can create a suitable confining potential for the formation of magnetic vortices6,7. This technique is advantageous because it allows magnetic vortices to be arbitrarily positioned in an extended film, and for the strength of the confining potential to be tuned by means of the amplitude of the electric current. Writing in Nature Nanotechnology, Thibaut Devolder
West
East SiO2 Free layer Metal Fixed layer Bottom electrode
Current
West
Current
East
Ground
Figure 1 | Schematic illustration of a device composed of two nanocontacts fabricated on top of a spin valve. The spin valve consists of a magnetically soft free-layer stacked onto a metallic spacer and a magnetically hard fixed-layer8. 96
and colleagues at IMEC, Katholieke Universiteit Leuven, CNRS and University of Paris-Sud have now shown that magnetic vortices can also be controllably transported in an extended magnetic film8. Devolder and colleagues have exploited the advantages of the nanocontact method and fabricated a device composed of two nanocontacts (east and west) on top of a spin valve structure. The spin valve is formed from a magnetically soft freelayer — that is, with a magnetization that is easy to rotate away from the equilibrium direction — stacked onto a metallic spacer and a magnetically hard fixed-layer (Fig. 1). The resistance of the spin valve depends on the relative orientation of the magnetization in the two magnetic layers. The Oersted field from the electrical current flowing through one of the nanocontacts can nucleate a magnetic vortex in the free layer. The nucleated vortex is excited by spin torque from the spin-polarized current flowing radially outward from the nanocontact in the film plane6,7,9, and undergoes circular motion in the vicinity of the nanocontact, due to an attractive force from the Zeeman energy associated with the Oersted field. This attractive force works like the gravity of a planet, with the vortex behaving like a satellite. The circular motion of the vortex around the nanocontact can be detected as a time evolution of the electrical resistance of the spin valve, which varies in dependence of the relative orientation of the magnetization in the free and pinned layers. The researchers first nucleated a vortex on the east nanocontact by injecting a current IE = 20 mA, and detected its circular motion around this contact by
NATURE NANOTECHNOLOGY | VOL 9 | FEBRUARY 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
Figure 1 | The delivery and clearance of nanoassemblies. The nanoassemblies comprising core (gold) and satellite (green) nanoparticles are constructed using complementary DNA sequences as linkers (red) and coated with ligands such as PEG (black strands, blue clouds) to control their interactions with cells. When injected intravenously, they are uptaken via the EPR effect to the target site. Once inside the cell (blue), the assemblies exert their therapeutic effect and breakdown into their constituents for elimination.
have a gold core nanoparticle, which is surrounded by one or more layers of gold satellite nanoparticles. Each nanoparticle was modified with single-stranded DNA and linked together using DNA strands with complementary sequences. This design strategy improved tumour accumulation of the nanoassembly via passive targeting and the EPR effect, while significantly reducing uptake in the reticuloendothelial system. Finally, the supramolecular structure decomposed into the original building blocks, which promoted efficient elimination of each component into urine (3–5-nm satellite nanoparticles). This strategy improves the biodistribution and renal clearance of injected inorganic nanoparticles in living animals, while maintaining tumour-targeting efficiency. Thus, the DNAmediated core–satellite supramolecular nanoassemblies have the potential to replace the toxic and hydrophobic inorganic nanoparticles, such as quantum dots and carbon nanotubes, that are typically used in drug-delivery systems (Fig. 1). During the past decade, it has been shown that a number of important factors
mediate the ultimate fate of nanoparticles in vivo including solubility, stability, shape and flexibility, size and size distribution, and formulation2,5,6. The design of precise nanostructures is typically focused on developing site-specific delivery and targeting capabilities but biodistribution and clearance of the nanostructures from the neighbouring tissue and local environment are also critical. Thus, it is important to increase specificity, selectivity and safety 1, which are the fundamental principles that govern the clinical translation of nanoparticles7. In particular, these factors are crucial for determining physiological behaviours and pharmacokinetic parameters of nanoparticles in the body, as well as to ascertain their potential cytotoxicity and in vivo toxicity. The use of biocompatible or biodegradable supramolecular structures is essential to enhance renal clearance, and eventually reduce potential in vivo toxicity (Fig. 1)3. While taking steps towards overcoming the fundamental limitations of nanoparticles, namely efficient targeting and renal clearance, the DNA
nanoassembly developed by Chan and colleagues3 still needs improvement as a targeted probe. First, the biodistribution and targeting process is mediated by macrophage uptake and phagosomal accumulation, which increases nonspecific uptake and retention in the liver and spleen. To avoid this immune response, longer PEG chains were used; however, this eventually will increase the overall hydrodynamic diameter of the nanoassembly, preventing renal clearance of decomposed building blocks. Second, passive targeting via the EPR effect requires the long circulation time of nanoparticles in the body and the EPR effect is only exploited when the size of the nanoparticle is larger than 10 nm in hydrodynamic diameter. Therefore, there is a balance to strike between pegylation, which increases the size of the nanoparticle, and keeping the diameter of the nanoparticles within limits to allow renal clearance and prevent undesired nonspecific background retentions. These important trade-offs should be explored in future studies. ❐ Hak Soo Choi is at the Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts 02215, USA. e-mail: [email protected] References 1. Lee, J. H., Park, G., Hong, G. H., Choi, J. & Choi, H. S. Quant. Imaging Med. Surg. 2, 266–273 (2012). 2. Choi, H. S. & Frangioni, J. V. Mol. Imaging 9, 291–310 (2010). 3. Chou, L. Y. et al. Nature Nanotech. 9, 148–155 (2014). 4. Choi, H. S. et al. Nature Biotechnol. 25, 1165–1170 (2007). 5. Liu, J., Yu, M., Zhou, C. & Zheng, J. Materials Today 16, 477–486 (December, 2013). 6. Albanese, A., Lam, A. K., Sykes, E. A., Rocheleau, J. V. & Chan, W. C. W. Nature Commun. 4, 2718 (2013). 7. Choi, H. S. et al. Nature Nanotech. 5, 42–47 (2010).
Published online: 26 January 2014
NANOWIRE TRANSISTORS
Room for manoeuvre
Kinked nanowire transistors that can be manipulated in three dimensions can be used to record the intracellular electrical signals of targeted cells.
Ziliang Carter Lin and Bianxiao Cui
M
any biological cells use electricity to control their physical and chemical functions. Some cells do so by generating action potentials, which are rapid rises and falls in the electric potential difference across their 94
cell membrane. In particular, muscle cells, neurons and endocrine cells use action potentials to initiate mechanical contraction, perform signal processing and trigger chemical release, respectively. Furthermore, the propagation of action
potentials along axons and through synapses is the primary means for neurons to communicate over long distances. Action potentials are generated by the rapid opening and closing of ion channels on the cell membrane. Malfunctions of these ion
NATURE NANOTECHNOLOGY | VOL 9 | FEBRUARY 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
news & views channels are the cause of many neurological disorders, cardiac arrhythmias and druginduced toxicity in humans. Therefore, understanding the electrophysiology and pathology of these cells, from single-cell to circuit-network levels, requires accurate recording of their action potentials. Writing in Nature Nanotechnology, Charles Lieber and colleagues at Harvard University and Wuhan University of Technology now show that a platform that combines kinked nanowire transistors and three-dimensional positioning can be used to record intracellular action potentials of desired cells and cell regions1. Traditionally, action potentials are recorded using either extracellular or intracellular electrodes (Fig. 1a,b). Extracellular recording such as the chipbased multielectrode array method uses electrodes positioned close to the cell membrane to detect extracellular field potentials. Although this method is non-invasive and easily scalable, it lacks one-to-one correspondence between the cells and electrodes. The recorded field potentials also fail to represent the shape of the intracellular action potentials, thus making data interpretation difficult. On the other hand, intracellular recording such as patch-clamp recording provides accurate measurements of the action potentials, but this method is laborious and invasive, which limits the recording throughput and duration. Furthermore, the relatively large size of the patch pipette restricts its application primarily to cell bodies. In the past few years, significant progress has been made in the development of nanoelectrodes that aims to combine the advantages of intracellular and extracellular recording methods. These approaches are based on nanopillar 2, nanowire3 or goldspine4 electrodes, or field-effect nanowire transistors5 (Fig. 1c,d), and methods such as mechanical penetration, electroporation and chemical modification are employed to facilitate electrical access into the cell interior. Until now, most of these electrodes have been built on chip platforms, which allows parallel recording but provides little control over the position of the electrodes with respect to the targeted cells. As a result, only the cells that are in contact with the electrodes are recorded. Now Lieber and colleagues have created a freely movable kinked nanowire field-effect transistor for precise positioning and intracellular recording (Fig. 1e)1. The team has previously shown that, unlike conventional linear nanowire transistors, kinked nanowire transistors can have the source and drain on the same side of the nanowire5. In this latest work, each kinked
nanowire transistor is mounted at the end of a polymer shaft printed with source and drain lines, which are in turn connected to a printed circuit board. The whole assembly is mounted on a three-dimensional translational stage, which allows the nanoelectrodes to be accurately positioned with respect to the cells. The nanowire transistor is coated with phospholipids to facilitate mechanical insertion into the interior of the cell. Using the platform, the researchers achieve high-fidelity intracellular recording of cardiac muscle cells. In particular, action potentials and resting membrane potentials recorded with the approach are found to be almost identical to recordings by patchclamp probes. This suggests that the kinked nanowire transistors are capable of achieving tight seals with the cell membrane — a key element in realizing high-quality recording. Lieber and colleagues also demonstrate that the technique can clearly identify the change in action potential on treatment with various ion channel blockers.
Compared with chip-based nanoelectrodes, this new platform affords great freedom of recording from any desired cell location. It also records almost full action potential amplitude, which is significantly larger than the attenuated action potentials recorded by chip-based nanoelectrodes. Compared with the patch-clamp technique, which uses a similar positioning method and measures the full action potentials, the new platform uses a much smaller probe. The diameter of the nanowire (~80 nm) is an order of magnitude smaller than that of a pipette tip (~1 μm), which makes it much less invasive. Furthermore, unlike the single-use pipette probes, the nanowire probe can be repeatedly inserted and withdrawn from cells without the need of cleaning or re-coating but still possesses similar sensitivity. Looking ahead, the platform could provide a variety of new scientific opportunities. One of the most exciting potential applications is subcellular recording at axons, dendrites and even single synapses. c
a
b
z
d
y
x
e
z
y
x
Figure 1 | Intra- and extracellular recording methods. a, Conventional extracellular recording consists of planar electrodes (orange) measuring field potentials of a nearby cell (blue). b, Conventional intracellular recording involves three-dimensional positioning of a glass pipette (purple) to a target cell. c, On-chip metal nanoelectrodes have recently been developed, which use a variety of different nanostructures as electrodes2–4. d, On-chip kinked nanowire transistor electrodes have also been recently developed5. The nanowires consist of straight silicon segments connected at sharp angles, with the junction regions doped to create field-effect transistors. e, By creating free-standing kinked-nanowire transistors, three-dimensional positioning becomes possible with nanowire sensors.
NATURE NANOTECHNOLOGY | VOL 9 | FEBRUARY 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
95
news & views The good seal offered by the technique should also allow the measurement of subthreshold events such as excitatory and inhibitory postsynaptic potentials. However, similar to patch-clamp recording, the positioning method severely limits the scalability of the approach. Furthermore, nanowire transistors and nanoelectrodes in general will become more useful when they acquire the ability to inject stable current into
the cell. In comparison, current injection is a useful feature of the patch-clamp technique, which allows clamping of the cell membrane potential at predetermined levels to reveal the current–voltage relationship of specific types of ion channel. ❐ Ziliang Carter Lin is at the Department of Applied Physics Stanford University, Stanford, California 94305, USA. Bianxiao Cui is at the
Department of Chemistry, Stanford University, Stanford, California 94305, USA. e-mail: [email protected] References 1. Qing, Q. et al. Nature Nanotech. 9, 142–147 (2014). 2. Xie, C., Lin, Z. L., Hanson, L., Cui, Y. & Cui, B. X. Nature Nanotech. 7, 185–190 (2012). 3. Robinson, J. T. et al. Nature Nanotech. 7, 180–184 (2012). 4. Hai, A., Shappir, J. & Spira, M. E. Nature Methods 7, 200–250 (2010). 5. Tian, B. Z. et al. Science 329, 830–834 (2010).
MAGNETIC NANOSTRUCTURES
Vortices on the move
Magnetic vortices can be controllably transferred in an extended system by electrical means.
Teruo Ono
M
agnetic vortices are in-plane curling spin structures that have a nanoscale core with an out-of plane magnetization1. They are formed in ferromagnetic disks and the binary nature of the cores (whose magnetization can point up or down) combined with their thermal stability make vortices appealing for future non-volatile memory devices. Importantly for such applications, it has been shown that the cores can be switched by applying suitable electrical or magnetic field pulses2–5. A magnetic vortex is formed in the presence of a confining potential such as that provided by the geometrical confinement of a disk. However,
applications other than memory devices could be developed if magnetic vortices could be freed from such geometrical constraints. Towards this end, it has been shown that the Oersted field generated by passing an electric current into a ferromagnetic film through a metallic nanocontact can create a suitable confining potential for the formation of magnetic vortices6,7. This technique is advantageous because it allows magnetic vortices to be arbitrarily positioned in an extended film, and for the strength of the confining potential to be tuned by means of the amplitude of the electric current. Writing in Nature Nanotechnology, Thibaut Devolder
West
East SiO2 Free layer Metal Fixed layer Bottom electrode
Current
West
Current
East
Ground
Figure 1 | Schematic illustration of a device composed of two nanocontacts fabricated on top of a spin valve. The spin valve consists of a magnetically soft free-layer stacked onto a metallic spacer and a magnetically hard fixed-layer8. 96
and colleagues at IMEC, Katholieke Universiteit Leuven, CNRS and University of Paris-Sud have now shown that magnetic vortices can also be controllably transported in an extended magnetic film8. Devolder and colleagues have exploited the advantages of the nanocontact method and fabricated a device composed of two nanocontacts (east and west) on top of a spin valve structure. The spin valve is formed from a magnetically soft freelayer — that is, with a magnetization that is easy to rotate away from the equilibrium direction — stacked onto a metallic spacer and a magnetically hard fixed-layer (Fig. 1). The resistance of the spin valve depends on the relative orientation of the magnetization in the two magnetic layers. The Oersted field from the electrical current flowing through one of the nanocontacts can nucleate a magnetic vortex in the free layer. The nucleated vortex is excited by spin torque from the spin-polarized current flowing radially outward from the nanocontact in the film plane6,7,9, and undergoes circular motion in the vicinity of the nanocontact, due to an attractive force from the Zeeman energy associated with the Oersted field. This attractive force works like the gravity of a planet, with the vortex behaving like a satellite. The circular motion of the vortex around the nanocontact can be detected as a time evolution of the electrical resistance of the spin valve, which varies in dependence of the relative orientation of the magnetization in the free and pinned layers. The researchers first nucleated a vortex on the east nanocontact by injecting a current IE = 20 mA, and detected its circular motion around this contact by
NATURE NANOTECHNOLOGY | VOL 9 | FEBRUARY 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
