news & views NANOLASERS
Lasing from 2D atomic crystals
The coupling of monolayer tungsten diselenide and a photonic-crystal cavity leads to ultralow-threshold lasing.
N
Figure 1 | Schematic of a 2D semiconductor nanolaser, consisting of a monolayer of WSe2 on top of a photonic crystal. Image courtesy of Tal Galfsky.
lead to enhanced spontaneous emission2–4, strong coupling and the formation of exciton polaritons5. By using a polymer-transfer process, Xu and co-authors placed a mechanically exfoliated monolayer of WSe2 — which plays the role of the gain medium — on top of a GaP photonic-crystal cavity, where light is localized (Fig. 1). The photonic crystal was made from GaP because of its high refractive index and transparency to the emission wavelength of WSe2. Although contrary to conventional laser designs, placing the gain medium outside the cavity is feasible because the atomically thin WSe2 crystal is confined within 1 nm of the surface of the photonic-crystal cavity. Hence, the crystal still has considerable overlap with the optical field. Such an architecture opens up the possibility of placing gain medium at desired locations after the fabrication of the photonic crystal or even larger photonic circuits. Additionally, this approach also alleviates restrictions imposed by traditional semiconductor-fabrication technologies, such as surface recombination arising from the etching of the embedded gain medium. In contrast to previous attempts with monolayer semiconductors, Xu and colleagues used cavities with a high quality factor, which ensures that photons will stay within the cavity long enough and interact with the 2D gain medium. In their devices, lasing is manifested by a distinct nonlinear threshold (a ‘kink’; Fig. 2) in the light output at an incident pump power of 27 nW, accompanied by a collapse of the spectral linewidth (Fig. 2). These signatures are indicative of a phase transition from spontaneous to stimulated emission occurring in the device. The ratio of input to output emission (typically defined as the fraction of the spontaneous emission that goes into the cavity mode, or β factor 6–8) is β = 0.19, meaning that only 19% of the spontaneous emission goes into the cavity mode. Although this might not be a big number, it is already comparable to some of the best photonic-crystal-based nanolasers embedded with quantum dots9. The measured peak output power was found to be 10 fW at an incident pump power of 100 nW.
NATURE MATERIALS | ADVANCE ONLINE PUBLICATION | www.nature.com/naturematerials
© 2015 Macmillan Publishers Limited. All rights reserved
β= β= β=
Linewidth
anolasers offer the possibility of reaching the quantum limit of a single emitter, and are essential for reducing power consumption in optical interconnects, high-resolution imaging and sensing. As the physics governing the operation of lasers — in essence, stimulated emission by a two-level system — became established, research shifted to their technological characteristics: emission at different frequencies, exploitation of different materials, and reduction of their size. Now, Xiaodong Xu, Arka Majumdar and co-workers report in Nature that lasing can arise from a monolayer of tungsten diselenide (WSe2) — a two-dimensional (2D) semiconductor — when coupled to an underlying photonic-crystal cavity 1. The researchers achieve continuous-wave lasing under optical pumping with a very low threshold of 27 nW at 130 K. Among 2D atomic crystals, monolayers of transition metal dichalcogenides have emerged as the most attractive candidates for photonic and optoelectronic applications. Transition metal dichalcogenides exhibit bandgaps ranging from the visible to near-infrared frequencies, and the crystals’ interaction with light can be controlled when they are incorporated into photonic structures such as microcavities. This can
Power out
Vinod Menon
1 1 0. 1 00 0.
Power in
Figure 2 | Characteristics of a 2D semiconductor nanolaser. The dependence of power-out versus power-in for different β values shows the appearance of a nonlinear ‘kink’ along with the collapse in linewidth (as seen by Xu and colleagues).
Engineering efforts should lead to improvements in the output power and the operational temperature of Xu and collaborators’ nanolaser. These could involve suitable transition metal dichalcogenides that emit at wavelengths compatible with silicon photonics and that achieve lasing under electrical injection. A bigger challenge would be the understanding of the fundamental aspects of the laser, such as its coherence properties and photon statistics. This is important to substantiate the true lasing characteristics of nanolasers and to help improve future laser designs and performance. Xu and colleagues’ work along with other recent demonstrations of singlephoton emission from defect states in 2D materials10,11, of the formation of roomtemperature polaritons5 and of enhanced spontaneous emission2–4, will surely stimulate further research activity on devices that incorporate 2D materials in or in contact with photonic structures such as micro- and nanocavities. Because of their inherently strong interaction with light, and other attractive properties such as valley polarization and strong spin–orbit coupling, 2D materials could open up avenues for 1
news & views the development of heretofore inaccessible device features with potential applications in classical and quantum photonics. ❐ Vinod Menon is in the Department of Physics, City College & Graduate Centre, City University of New York, New York, New York 10031, USA. e-mail: [email protected]
2
References
1. Wu, S. et al. Nature http://dx.doi.org/10.1038/nature14290 (2015). 2. Gan, X. et al. Appl. Phys. Lett. 103, 181119 (2013). 3. Wu, S. et al. 2D Mater. 1, 011001 (2014). 4. Schwarz, S. et al. Nano Lett. 14, 7003–7008 (2014). 5. Liu, X. et al. Nature Photon. 9, 30–34 (2015). 6. Chow, W. W., Jhanke, F. & Gies, C. Light Sci. Appl. 3, e201 (2014). 7. Khajavikhan, M. et al. Nature 482, 204–207 (2012).
8. Strauf, S. & Jahnke, F. Laser Photon. Rev. 5, 607–633 (2011). 9. Ellis, B. et al. Nature Photon. 5, 297–300 (2011). 10. Srivastava, A. et al. Preprint at http://arXiv.org/abs/1411.0025 (2014). 11. He, Y.-M. et al. Preprint at http://arXiv.org/abs/1411.2449 (2014).
Published online: 16 March 2015
NATURE MATERIALS | ADVANCE ONLINE PUBLICATION | www.nature.com/naturematerials
© 2015 Macmillan Publishers Limited. All rights reserved
Lasing from 2D atomic crystals
The coupling of monolayer tungsten diselenide and a photonic-crystal cavity leads to ultralow-threshold lasing.
N
Figure 1 | Schematic of a 2D semiconductor nanolaser, consisting of a monolayer of WSe2 on top of a photonic crystal. Image courtesy of Tal Galfsky.
lead to enhanced spontaneous emission2–4, strong coupling and the formation of exciton polaritons5. By using a polymer-transfer process, Xu and co-authors placed a mechanically exfoliated monolayer of WSe2 — which plays the role of the gain medium — on top of a GaP photonic-crystal cavity, where light is localized (Fig. 1). The photonic crystal was made from GaP because of its high refractive index and transparency to the emission wavelength of WSe2. Although contrary to conventional laser designs, placing the gain medium outside the cavity is feasible because the atomically thin WSe2 crystal is confined within 1 nm of the surface of the photonic-crystal cavity. Hence, the crystal still has considerable overlap with the optical field. Such an architecture opens up the possibility of placing gain medium at desired locations after the fabrication of the photonic crystal or even larger photonic circuits. Additionally, this approach also alleviates restrictions imposed by traditional semiconductor-fabrication technologies, such as surface recombination arising from the etching of the embedded gain medium. In contrast to previous attempts with monolayer semiconductors, Xu and colleagues used cavities with a high quality factor, which ensures that photons will stay within the cavity long enough and interact with the 2D gain medium. In their devices, lasing is manifested by a distinct nonlinear threshold (a ‘kink’; Fig. 2) in the light output at an incident pump power of 27 nW, accompanied by a collapse of the spectral linewidth (Fig. 2). These signatures are indicative of a phase transition from spontaneous to stimulated emission occurring in the device. The ratio of input to output emission (typically defined as the fraction of the spontaneous emission that goes into the cavity mode, or β factor 6–8) is β = 0.19, meaning that only 19% of the spontaneous emission goes into the cavity mode. Although this might not be a big number, it is already comparable to some of the best photonic-crystal-based nanolasers embedded with quantum dots9. The measured peak output power was found to be 10 fW at an incident pump power of 100 nW.
NATURE MATERIALS | ADVANCE ONLINE PUBLICATION | www.nature.com/naturematerials
© 2015 Macmillan Publishers Limited. All rights reserved
β= β= β=
Linewidth
anolasers offer the possibility of reaching the quantum limit of a single emitter, and are essential for reducing power consumption in optical interconnects, high-resolution imaging and sensing. As the physics governing the operation of lasers — in essence, stimulated emission by a two-level system — became established, research shifted to their technological characteristics: emission at different frequencies, exploitation of different materials, and reduction of their size. Now, Xiaodong Xu, Arka Majumdar and co-workers report in Nature that lasing can arise from a monolayer of tungsten diselenide (WSe2) — a two-dimensional (2D) semiconductor — when coupled to an underlying photonic-crystal cavity 1. The researchers achieve continuous-wave lasing under optical pumping with a very low threshold of 27 nW at 130 K. Among 2D atomic crystals, monolayers of transition metal dichalcogenides have emerged as the most attractive candidates for photonic and optoelectronic applications. Transition metal dichalcogenides exhibit bandgaps ranging from the visible to near-infrared frequencies, and the crystals’ interaction with light can be controlled when they are incorporated into photonic structures such as microcavities. This can
Power out
Vinod Menon
1 1 0. 1 00 0.
Power in
Figure 2 | Characteristics of a 2D semiconductor nanolaser. The dependence of power-out versus power-in for different β values shows the appearance of a nonlinear ‘kink’ along with the collapse in linewidth (as seen by Xu and colleagues).
Engineering efforts should lead to improvements in the output power and the operational temperature of Xu and collaborators’ nanolaser. These could involve suitable transition metal dichalcogenides that emit at wavelengths compatible with silicon photonics and that achieve lasing under electrical injection. A bigger challenge would be the understanding of the fundamental aspects of the laser, such as its coherence properties and photon statistics. This is important to substantiate the true lasing characteristics of nanolasers and to help improve future laser designs and performance. Xu and colleagues’ work along with other recent demonstrations of singlephoton emission from defect states in 2D materials10,11, of the formation of roomtemperature polaritons5 and of enhanced spontaneous emission2–4, will surely stimulate further research activity on devices that incorporate 2D materials in or in contact with photonic structures such as micro- and nanocavities. Because of their inherently strong interaction with light, and other attractive properties such as valley polarization and strong spin–orbit coupling, 2D materials could open up avenues for 1
news & views the development of heretofore inaccessible device features with potential applications in classical and quantum photonics. ❐ Vinod Menon is in the Department of Physics, City College & Graduate Centre, City University of New York, New York, New York 10031, USA. e-mail: [email protected]
2
References
1. Wu, S. et al. Nature http://dx.doi.org/10.1038/nature14290 (2015). 2. Gan, X. et al. Appl. Phys. Lett. 103, 181119 (2013). 3. Wu, S. et al. 2D Mater. 1, 011001 (2014). 4. Schwarz, S. et al. Nano Lett. 14, 7003–7008 (2014). 5. Liu, X. et al. Nature Photon. 9, 30–34 (2015). 6. Chow, W. W., Jhanke, F. & Gies, C. Light Sci. Appl. 3, e201 (2014). 7. Khajavikhan, M. et al. Nature 482, 204–207 (2012).
8. Strauf, S. & Jahnke, F. Laser Photon. Rev. 5, 607–633 (2011). 9. Ellis, B. et al. Nature Photon. 5, 297–300 (2011). 10. Srivastava, A. et al. Preprint at http://arXiv.org/abs/1411.0025 (2014). 11. He, Y.-M. et al. Preprint at http://arXiv.org/abs/1411.2449 (2014).
Published online: 16 March 2015
NATURE MATERIALS | ADVANCE ONLINE PUBLICATION | www.nature.com/naturematerials
© 2015 Macmillan Publishers Limited. All rights reserved
