news & views a surface code11. Notably, this model can benefit hugely if the nodes consist of small clusters of coupled qubits, such as those now demonstrated. ❐
Engineering, University College London, London WC1H 0AH, UK. e-mail: [email protected]; [email protected]
John J. L. Morton and Jeroen Elzerman are at the London Centre for Nanotechnology and Department of Electronic & Electrical
1. 2. 3. 4.
5. Awschalom, D. D., Epstein, R. & Hanson, R. Sci. Am. 297, 84–91 (2007). 6. Filidou, V. et al. Nature Phys. 8, 596–600 (2012). 7. Viola, L. & Lloyd, S. Phys. Rev. A 58, 2733–2744 (1998). 8. Dennis, E., Kitaev, A., Landahl, A. & Preskill, J. J. Math. Phys. 43, 4452–4505 (2002). 9. Yao, X.‑C. et al. Nature 482, 489–494 (2012). 10. Bernien, H. et al. Nature 497, 86–90 (2013). 11. Nickerson, N. H., Li, Y. & Benjamin, S. C. Nature Commun. 4, 1756 (2012).
References Waldherr, G. et al. Nature 506, 204–207 (2014). Taminiau, T. H. et al. Nature Nanotech. 9, 171–176 (2014). Shor, P. W. Phys. Rev. A 52, 2493–2496 (1995). Steane, A. M. Phys. Rev. Lett. 77, 793–797 (1996).
QUANTUM DOTS
Electrifying cavities
A design that allows electrical contacts to be created on semiconductor microphotonic structures brings quantum networks based on semiconductor single photon sources one step closer.
Ruth Oulton
T
he ideal photonic quantum network would have access to a large supply of single photons. We could imagine a chip that contains a large array of devices, with each of them producing exactly one photon when triggered electrically or optically. Single photons from each source must be indistinguishable in wavelength, polarization and bandwidth1. Reporting in Nature Communications, Pascale Senellart and co-workers from the Laboratoire de Photonique et de Nanostructures, CNRS, France, Université Paris Diderot, and attocube systems AG, Munich demonstrate a significant step towards the practical realization of an array of single photon sources2. They have realized a semiconductor quantum dot (QD) single photon source that allows an electric field to be applied to the QD without impeding the path of single photons. Self-assembled QDs have long been considered viable practical single photon sources. Each dot can be populated by an electron and a hole, and when these recombine to emit a photon, the QD empties and emits no further photons at the same wavelength until it gets populated again. Key to the success of QD technology is that a QD may be embedded in a distributed Bragg reflector (DBR) cavity structure that allows efficient out-coupling of the photons (Fig. 1). Etching a pillar structure in a DBR confines the light, via total internal reflection, to a small volume. In the past decade, improvements in pillar design and fabrication have allowed physicists to replicate in the solid state many of the cavity quantum electrodynamic effects observed first with atoms. Careful design and fabrication of QD pillars has led to a
bright source of identical (indistinguishable) photons3 and entangled photons4 ready for integration into a quantum chip. Further progress in micropillar design has been hampered by the significant difficulties in applying an electric field to the micropillar. Application of an electric field is an important functionality if QDs are to be a viable quantum technology. The self-assembly process produces QDs with almost-perfect quantum efficiency. But it is not possible to control the position a
at which the QDs form in the pillar, and even knowing the precise location of the QDs is challenging. This makes it difficult to fabricate a pillar with a QD placed in the middle of the cavity. Moreover, the QDs are not homogeneous in size, which means that the light emitted may not match the cavity wavelength exactly. An electric field can be used to fine-tune the photon wavelength, overcoming the problem of size inhomogeneity to produce an array of indistinguishable photon sources. b
Photon emission
DBR layers
λ/2
QD 5 μm
Figure 1 | Dot in a pillar. a, Schematic of a QD emitter embedded in the centre of a micropillar cavity. Alternating quarter-wavelength layers of GaAs/AlAs, known as DBRs, cause destructive interference, which allows them to effectively act as mirrors. A GaAs cavity of depth λ/2 allows confinement of light with wavelength λ, with further confinement achieved by etching the DBR into a pillar a few micrometres in diameter. Individual QDs incorporated in the centre of the cavity are identified spatially and spectrally using spectroscopy, allowing a pillar of a suitable mode wavelength and position to be fabricated. b, A scanning electron microscopy image of a set of fabricated micropillars. Panel b reproduced with permission from ref. 8, © 2009 AIP.
NATURE NANOTECHNOLOGY | VOL 9 | MARCH 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
169
news & views
1 μm wire
d
λ/2 cavity Bragg mirrors
z
y x
Figure 2 | Micropillar with electric contacts2. After identification of the QD, a pillar is etched (with diameter, d), which is attached to four ‘nanowires’ connected to a frame. Doping in the DBRs allows an electrical contact (yellow) on top of the DBR layer, enabling an electric field to be applied.
Furthermore, contacts allow electrical carrier injection: one could apply the same compact and convenient method to inject carriers as used in semiconductor lasers for telecommunication technologies. Although large devices may be successfully contacted to produce single photon sources5, contacting small pillars has proven difficult. The top of the freestanding pillars, just a few micrometres in diameter should ideally be kept free from metal contacts and electrical bond wires that absorb and scatter the photon emission. This technical difficulty was only previously solved by one group, who were able to back-fill the structure with polymer,
170
allowing a support for a ring-shaped metallic contact around the pillar 6,7. The fabrication steps, however, are complex and difficult to replicate reliably. The work by Senellart and co-workers offers an alternative solution. Their pillar design allows the realization of a top electrical contact without bonding directly to the micropillar. In their design (Fig. 2), the pillar is connected to a larger frame via four 1-μm-wide one-dimensional ‘wires’. A metal contact is located away from the pillar at the frame, with p- and n-doping in the DBR stacks allowing a vertical electrical field to be applied. The quality of the pillars is comparable to that of isolated pillars of similar diameter (3 μm). Senellart and co-workers used a method they previously pioneered to locate the position of a single QD with high accuracy 8, and then fabricated around the structure a micropillar whose size is tuned to the same frequency as the QD emission. Combining this fabrication technique with the ability to fine-tune the QD states by an electric field allows the production of single photons with very high extraction efficiency (>50%) and represents a key breakthrough. Over the past decade, cutting-edge QD physics has progressed in tandem with QD cavity fabrication techniques. High-quality fabrication of individual devices is now becoming standard. Arrays of identical devices are a vital next step for a compact and scalable quantum technology. The designs and techniques utilized by Senellart and co-workers offer a potential solution to
these key challenges. As this work illustrates, the ability to combine several established techniques is a significant achievement. The >50% efficiency, high-quality single photon sources demonstrated here would already outperform standard spontaneous parametric downconversion sources, the workhorse photon-pair source used in quantum photonics, which is intrinsically limited to ~5% efficiency 1. Theoretical work9 shows that for scalable quantum computing, a combined source–detector efficiency should be >2/3. With conventional single-photon-detector technology giving an efficiency of ~70%, the source should be ~95% efficient to represent the next breakthrough in technology. Such advances will, however, likely result only from incremental progress in combining different fabrication technologies, as shown here. ❐ Ruth Oulton is at the H. H. Wills Physics Laboratory and the Department of Electrical and Electronic Engineering, Nanoscience and Quantum Information Building, University of Bristol, Tyndall Avenue, Bristol BS8 1FD, UK. e-mail: [email protected] References 1. O’Brien, J. L., Furusawa, A. & Vučković, J. Nature Photon. 3, 687–695 (2009). 2. Nowak, A. K. Nature Commun. 5, 3240 (2014). 3. Gazzano, O. et al. Nature Commun. 4, 1425 (2012). 4. Dousse, A. et al. Nature 466, 217–220 (2010). 5. Yuan, Z. et al. Science 295, 102–105 (2001). 6. Böckler, C. et al. Appl. Phys Lett. 92, 091107 (2008). 7. Heindl, T. et al. Appl. Phys. Lett. 96, 011107 (2010). 8. Dousse, A. et al. Appl. Phys. Lett. 94, 121102 (2009). 9. Varnava, M., Browne, D. E. & Rudolph, T. Phys. Rev. Lett. 100, 060502 (2008).
NATURE NANOTECHNOLOGY | VOL 9 | MARCH 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
Engineering, University College London, London WC1H 0AH, UK. e-mail: [email protected]; [email protected]
John J. L. Morton and Jeroen Elzerman are at the London Centre for Nanotechnology and Department of Electronic & Electrical
1. 2. 3. 4.
5. Awschalom, D. D., Epstein, R. & Hanson, R. Sci. Am. 297, 84–91 (2007). 6. Filidou, V. et al. Nature Phys. 8, 596–600 (2012). 7. Viola, L. & Lloyd, S. Phys. Rev. A 58, 2733–2744 (1998). 8. Dennis, E., Kitaev, A., Landahl, A. & Preskill, J. J. Math. Phys. 43, 4452–4505 (2002). 9. Yao, X.‑C. et al. Nature 482, 489–494 (2012). 10. Bernien, H. et al. Nature 497, 86–90 (2013). 11. Nickerson, N. H., Li, Y. & Benjamin, S. C. Nature Commun. 4, 1756 (2012).
References Waldherr, G. et al. Nature 506, 204–207 (2014). Taminiau, T. H. et al. Nature Nanotech. 9, 171–176 (2014). Shor, P. W. Phys. Rev. A 52, 2493–2496 (1995). Steane, A. M. Phys. Rev. Lett. 77, 793–797 (1996).
QUANTUM DOTS
Electrifying cavities
A design that allows electrical contacts to be created on semiconductor microphotonic structures brings quantum networks based on semiconductor single photon sources one step closer.
Ruth Oulton
T
he ideal photonic quantum network would have access to a large supply of single photons. We could imagine a chip that contains a large array of devices, with each of them producing exactly one photon when triggered electrically or optically. Single photons from each source must be indistinguishable in wavelength, polarization and bandwidth1. Reporting in Nature Communications, Pascale Senellart and co-workers from the Laboratoire de Photonique et de Nanostructures, CNRS, France, Université Paris Diderot, and attocube systems AG, Munich demonstrate a significant step towards the practical realization of an array of single photon sources2. They have realized a semiconductor quantum dot (QD) single photon source that allows an electric field to be applied to the QD without impeding the path of single photons. Self-assembled QDs have long been considered viable practical single photon sources. Each dot can be populated by an electron and a hole, and when these recombine to emit a photon, the QD empties and emits no further photons at the same wavelength until it gets populated again. Key to the success of QD technology is that a QD may be embedded in a distributed Bragg reflector (DBR) cavity structure that allows efficient out-coupling of the photons (Fig. 1). Etching a pillar structure in a DBR confines the light, via total internal reflection, to a small volume. In the past decade, improvements in pillar design and fabrication have allowed physicists to replicate in the solid state many of the cavity quantum electrodynamic effects observed first with atoms. Careful design and fabrication of QD pillars has led to a
bright source of identical (indistinguishable) photons3 and entangled photons4 ready for integration into a quantum chip. Further progress in micropillar design has been hampered by the significant difficulties in applying an electric field to the micropillar. Application of an electric field is an important functionality if QDs are to be a viable quantum technology. The self-assembly process produces QDs with almost-perfect quantum efficiency. But it is not possible to control the position a
at which the QDs form in the pillar, and even knowing the precise location of the QDs is challenging. This makes it difficult to fabricate a pillar with a QD placed in the middle of the cavity. Moreover, the QDs are not homogeneous in size, which means that the light emitted may not match the cavity wavelength exactly. An electric field can be used to fine-tune the photon wavelength, overcoming the problem of size inhomogeneity to produce an array of indistinguishable photon sources. b
Photon emission
DBR layers
λ/2
QD 5 μm
Figure 1 | Dot in a pillar. a, Schematic of a QD emitter embedded in the centre of a micropillar cavity. Alternating quarter-wavelength layers of GaAs/AlAs, known as DBRs, cause destructive interference, which allows them to effectively act as mirrors. A GaAs cavity of depth λ/2 allows confinement of light with wavelength λ, with further confinement achieved by etching the DBR into a pillar a few micrometres in diameter. Individual QDs incorporated in the centre of the cavity are identified spatially and spectrally using spectroscopy, allowing a pillar of a suitable mode wavelength and position to be fabricated. b, A scanning electron microscopy image of a set of fabricated micropillars. Panel b reproduced with permission from ref. 8, © 2009 AIP.
NATURE NANOTECHNOLOGY | VOL 9 | MARCH 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
169
news & views
1 μm wire
d
λ/2 cavity Bragg mirrors
z
y x
Figure 2 | Micropillar with electric contacts2. After identification of the QD, a pillar is etched (with diameter, d), which is attached to four ‘nanowires’ connected to a frame. Doping in the DBRs allows an electrical contact (yellow) on top of the DBR layer, enabling an electric field to be applied.
Furthermore, contacts allow electrical carrier injection: one could apply the same compact and convenient method to inject carriers as used in semiconductor lasers for telecommunication technologies. Although large devices may be successfully contacted to produce single photon sources5, contacting small pillars has proven difficult. The top of the freestanding pillars, just a few micrometres in diameter should ideally be kept free from metal contacts and electrical bond wires that absorb and scatter the photon emission. This technical difficulty was only previously solved by one group, who were able to back-fill the structure with polymer,
170
allowing a support for a ring-shaped metallic contact around the pillar 6,7. The fabrication steps, however, are complex and difficult to replicate reliably. The work by Senellart and co-workers offers an alternative solution. Their pillar design allows the realization of a top electrical contact without bonding directly to the micropillar. In their design (Fig. 2), the pillar is connected to a larger frame via four 1-μm-wide one-dimensional ‘wires’. A metal contact is located away from the pillar at the frame, with p- and n-doping in the DBR stacks allowing a vertical electrical field to be applied. The quality of the pillars is comparable to that of isolated pillars of similar diameter (3 μm). Senellart and co-workers used a method they previously pioneered to locate the position of a single QD with high accuracy 8, and then fabricated around the structure a micropillar whose size is tuned to the same frequency as the QD emission. Combining this fabrication technique with the ability to fine-tune the QD states by an electric field allows the production of single photons with very high extraction efficiency (>50%) and represents a key breakthrough. Over the past decade, cutting-edge QD physics has progressed in tandem with QD cavity fabrication techniques. High-quality fabrication of individual devices is now becoming standard. Arrays of identical devices are a vital next step for a compact and scalable quantum technology. The designs and techniques utilized by Senellart and co-workers offer a potential solution to
these key challenges. As this work illustrates, the ability to combine several established techniques is a significant achievement. The >50% efficiency, high-quality single photon sources demonstrated here would already outperform standard spontaneous parametric downconversion sources, the workhorse photon-pair source used in quantum photonics, which is intrinsically limited to ~5% efficiency 1. Theoretical work9 shows that for scalable quantum computing, a combined source–detector efficiency should be >2/3. With conventional single-photon-detector technology giving an efficiency of ~70%, the source should be ~95% efficient to represent the next breakthrough in technology. Such advances will, however, likely result only from incremental progress in combining different fabrication technologies, as shown here. ❐ Ruth Oulton is at the H. H. Wills Physics Laboratory and the Department of Electrical and Electronic Engineering, Nanoscience and Quantum Information Building, University of Bristol, Tyndall Avenue, Bristol BS8 1FD, UK. e-mail: [email protected] References 1. O’Brien, J. L., Furusawa, A. & Vučković, J. Nature Photon. 3, 687–695 (2009). 2. Nowak, A. K. Nature Commun. 5, 3240 (2014). 3. Gazzano, O. et al. Nature Commun. 4, 1425 (2012). 4. Dousse, A. et al. Nature 466, 217–220 (2010). 5. Yuan, Z. et al. Science 295, 102–105 (2001). 6. Böckler, C. et al. Appl. Phys Lett. 92, 091107 (2008). 7. Heindl, T. et al. Appl. Phys. Lett. 96, 011107 (2010). 8. Dousse, A. et al. Appl. Phys. Lett. 94, 121102 (2009). 9. Varnava, M., Browne, D. E. & Rudolph, T. Phys. Rev. Lett. 100, 060502 (2008).
NATURE NANOTECHNOLOGY | VOL 9 | MARCH 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
