SPIE Membership Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:
SPIE Photonics West 2018 | Call for Papers




Print PageEmail PageView PDF

Optoelectronics & Communications

Optical switching in the event horizon

A novel concept for an optical transistor allows a weak pulse to switch a much stronger one, overcoming a serious limitation of previous approaches for all-optical logic functions.
26 September 2011, SPIE Newsroom. DOI: 10.1117/2.1201109.003844

There is a growing trend to move data in optical networks directly using optical methods, avoiding electronic processing wherever possible. But despite the numerous functionalities that can now be implemented all-optically, it appears virtually impossible to accomplish the most basic active electronic component—the transistor—in a satisfactory way. Many concepts for optical transistors have been suggested and demonstrated. Nonlinear resonators have been extensively explored,1, 2 the optical Kerr effect (intensity-dependent nonlinear increase of the refractive index) has been exploited in many different ways, and optical switching has been shown using single molecules.3 While all these advances certainly do have their virtues, they often fail to fulfil rather straightforward criteria on a practical transistor, such as fanout and cascadability. One of the intriguing features of an electronic transistor is its ability to control a large current with a much smaller one: the amplified fanned-out output of a single switch can thus drive hundreds of cascaded switches. Such elementary staging of optical transistors, however, is not an option for any of the concepts shown so far.

On the physics side, this dilemma is readily understandable, as photons do not interact directly but only via nonlinear optical effects in matter. The latter are much weaker than direct electrostatic interaction between electrons. Consequently, it is much more difficult to build a photonic transistor. Nevertheless, to be useful, an optical transistor needs to provide a certain set of minimum functionalities. In a commentary in Nature Photonics in 2010, David Miller of Stanford University defined a list of seven criteria necessary for a practical optical transistor.4 At the time, not a single published concept complied with the full set of specifications. In particular, nearly every optical transistor proposed so far would require a much stronger pulse to switch a weaker one.

While this basic dilemma had long been known, no practical means had been found to effectively enhance the optical nonlinearity for the switching process, at least not without sacrificing other major criteria. One apparently straightforward idea is simply to choose materials that exhibit a stronger nonlinearity, e.g., Kerr media close to the bandgap. In this region, however, two-photon absorption occurs,5 effectively thwarting the approach.

Recently, however, a concept was introduced that may enable circumventing the physical limitations in a much smarter way: the so-called optical event horizon. Ulf Leonhardt's group in St. Andrews showed that one weak pulse may be captured in the nonlinear wake of a stronger pulse if both pulses propagate at nearly the same linear group velocity.6 According to the Kerr effect, a strong optical pulse decreases the group velocity in its immediate surroundings, not only for itself but also for light propagating at other wavelengths. If such a second pulse approaches the strong pulse from behind at slightly higher linear velocity, it will decelerate once it enters the realm of the optical horizon induced by the stronger pulse. Depending on initial conditions, pulses will find it impossible to penetrate this event horizon, and they will be captured in it for extended periods of time.

This ‘solitary confinement’ actually enables a strong nonlinear interaction between the two pulses that would otherwise not be possible. On basis of this interaction, we reported7 a scheme for efficient all-optical control of light pulses, fulfilling all criteria for a practical optical transistor. The underlying mechanism is most easily visualized as an elastic scattering process between the two pulses, transferring energy from one pulse to the other while conserving the total energy. Because the underlying four-wave mixing process also conserves the photon number, this interaction causes an energy change of the individual photons on both sides, i.e., a wavelength change among other secondary effects (see Figure 1).

Figure 1. Visualization of the optical switching process. Delays are shown relative to the unperturbed propagation of the strong signal pulse (soliton) along the coordinate z. When a weaker pulse is launched at positive delays with slightly larger group velocity, it will collide with the soliton and remain captured in the event horizon for extended periods of time. In this phase, both pulses will strongly interact until they eventually break free. As the result of this interaction, all soliton parameters, including its center wavelength and peak power, are modified.

In combination with specific material properties, the additional impact on the strong pulse now enables versatile manipulation of the strong pulse. Analyzing this situation in numerical simulations and also investigating it experimentally, we find that a pulse that is up to five times weaker may be sufficient to frequency-shift a stronger pulse by its full width at half-maximum,7, 8 enabling all-optical switching with near-perfect contrast. Moreover, we find that all other criteria defined by Miller4 are also fulfilled by our scheme.

In conclusion, the concept of an optical event horizon quite surprisingly implies a very effective means of locking two pulses together, strongly enhancing their mutual interaction. For the latter, the only prerequisite is the near-coincidence of group velocities for spectrally well separable signals, a condition that is always met provided that a zero-dispersion wavelength exists. Consequently, this concept can be readily scaled to integrated optical waveguides, enabling logic functions on a chip and bringing us one significant step closer to integrated all-optical logic functions and, finally, to the dream of an all-optical computer. Our future research will be directed toward miniaturizing all-optical, cascaded transistors and integrating them on-chip to implement practical logic-gate functions.

Günter Steinmeyer
Max Born Institute
Berlin, Germany
Günter Steinmeyer
Optoelectronics Research Centre Tampere University of Technology
and, Finland
Shalva Amiranashvili
Weierstrass Institute for Applied Analysis and Stochastics
Berlin, Germany

1. K. Jain, G. W. Pratt Jr., Optical transistor, Appl. Phys. Lett. 28, pp. 719, 1976.
2. P. W. Smith, E. H. Turner, A bistable Fabry-Perot resonator, Appl. Phys. Lett. 30, pp. 280, 1977.
3. J. Hwang, M. Pototschnig, R. Lettow, G. Zumofen, A. Renn, S. Götzinger, V. Sandoghdar, A single-molecule optical transistor, Nature 460, pp. 76, 2009.
4. D. A. B. Miller, Are optical transistors the next logical step?, Nat. Photon. 4, pp. 3, 2010.
5. M. Sheik-Bahae, D. C. Hutchings, D. J. Hagan, E. W. Van Stryland, Dispersion of bound electron nonlinear refraction in solids, IEEE J. Quantum Electron. 27, pp. 1296, 1991.
6. T. G. Philbin, C. Kuklewicz, S. Robertson, S. Hill, F. König, U. Leonhardt, Fiber-optical analog of the event horizon, Science 319, pp. 1367, 2008.
7. A. Demircan, Sh. Amiranashvili, G. Steinmeyer, Controlling light by light with an optical event horizon, Phys. Rev. Lett. 106, pp. 163901, 2011.
8. J. Bethge, C. Brée, Sh. Amiranashvili, F. Noack, G. Steinmeyer, A. Demircan, Solitonic transistor in the optical event horizon, OSA Tech. Digest, 2011. Paper CMJ6