Optical Manufacturing: Femtosecond-laser direct-written waveguides produce quantum circuits in glass

THOMAS MEANY

Quantum information science (QIS) is attracting significant funding and excitement. In the United Kingdom, funding of 270 million pounds has recently been allotted specifically to quantum-enabled technologies. Reasons for this interest and investment include the prospect of disruptive capabilities and the significant and accelerating progress over the last century. Initial progress in understanding and applying quantum mechanics has led to the development of semiconductor technologies and all their practical benefits. However, this century will lead to technological innovations that are inherently based on the quantum-mechanical behavior of particles.

For example, the possibility exists of encoding a quantum bit, a qubit, in the polarization of a single photon. This is natural because, as photonics researchers, we regularly manipulate both the vertical (V, corresponding to a bit value of 1) and horizontal (H, corresponding to a bit value of 0) polarizations of the electric field using waveplates. We are also familiar with the most arbitrary superposition of these states: an elliptically polarized beam described by a linear combination of horizontally and vertically polarized states with a phase between 0 and 2π.

In this article, we will discuss the prospect of using photons as information carriers, specifically using photons in lightwave circuits fabricated using a femtosecond-laser direct-write technique. This approach has the advantage of being useful not only for the so-called "holy grail" of QIS, which is quantum computation, but also as a natural choice for communication.

Confining photons within waveguides on a chip provides a stable environment for nested interferometric components; the photons don't undergo quantum decoherence or unintentionally interact with their surrounding environment. In fact, it is likely that photons will play some role in any quantum technology due to their extremely long coherence time-even photons arriving from space have been shown to display a polarization orientation.

Writing waveguides in glass

Integrated optics has been a critical development in the field of quantum photonics. It offers the prospect of stabilizing nested interferometers, which are essential to managing qubit manipulation and ultimately qubit interactions (a nested interferometer contains a second complete interferometer in one of its arms). The small size offered by integrated optics is of course also crucial to taking experiments out of laboratories and producing practical technologies. These on-chip optical circuits operate primarily by the process of total internal reflection (alternative methods using bandgap confinement in photonic crystals are also regularly used, but will not be discussed here).

A range of integrated devices is in use, reflecting what is occurring in the field of advanced optical interconnects and telecommunications components. The diverse range of on-chip platforms is accelerating progress in QIS; ridge, buried, and photonic-crystal waveguide structures have been demonstrated in semiconductors such as silicon, gallium arsenide, silicon carbide, and even diamond. Furthermore, crystalline materials have been used to fabricate waveguide circuits; lithium niobate offers a second-order optical nonlinearity enabling both high-speed switching and photon-generation capabilities.

However, glasses also offer quite a convenient platform for QIS experiments since they offer the prospect of low-loss transmission and efficient coupling to an optical fiber. The value of this property cannot be overestimated because, regardless of the complexity of the optical function enabled by an integrated-optical platform, it is useless unless the photons can be efficiently extracted.

A fundamental theorem, called the "no-cloning" theorem, precludes the amplification of quantum information; as a result, every photon in a quantum-information system is precious. Indeed, the first demonstration of integrated quantum photonics exploited a glass waveguide precisely for its efficient properties.¹

Glass waveguides have been fabricated using photolithographic techniques that, although effective, mean that for current and emerging prototype device designs, a new mask must be fabricated for each new structure. This is a slow and costly prototyping procedure.

An alternative technique to developing waveguides in glass is called the femtosecond laser direct-write (FLDW) technique. This exploits a nonlinear-absorption process that occurs at high laser intensities in a transparent material. As a result, when a laser is focused below the surface of a transparent material, causing this nonlinear effect, the absorption is highly localized at the laser focus.

This absorption process is quite complex and the resulting material properties are dependent on the laser intensity. At high intensities, damage and even void formation can occur, while at medium intensities, a periodic subwavelength structure can be observed. At low intensities, a modification of the material's refractive index can be observed. By smoothly translating the substrate with respect to the laser focus, it is possible to form a waveguide (see Fig. 1). This can occur in three dimensions and means that this technique can easily produce low-loss waveguiding devices at multiple depths.^2,3

The laser-writing technique is both cheap and flexible in comparison to the cleanroom facilities required for photolithographic waveguide processing. These qualities have given rise to a diverse range of research groups that have sprung up across the globe since this technique was first demonstrated in 1996.⁴

This technique is already becoming a commercially viable technology. For example, a company called Femtoprint (Muzzano, Switzerland; www.femtoprint.ch) has developed a glass-processing system using FLDW.⁵ Another company, Optoscribe (Livingston, Scotland; optoscribe.com), specializes in the fabrication of custom multimode fan-out devices using FLDW. In addition, a laser processing technology developed by Nanoscribe (Eggenstein-Leopoldshafen, Germany; www.nanoscribe.de) can be used to fabricate a diverse range of structures in many materials on scales of centimeters with feature resolution of nanometers.

Laser-written quantum photonics

In part due to its simplicity, laser inscription of circuits has been used to perform a number of critical tasks. Photon generation, manipulation, and detection have all individually been achieved using laser-inscribed circuits. Efficient photon generation can be achieved using the small third-order nonlinearity of fused silica combined with the high extraction efficiency of the material.⁶

Manipulation of the polarization state on-chip has been achieved using birefringent waveguides that may be rotated and incorporated into polarization-dependent power splitters.⁷ Both tunable devices and waveguide-based single-photon detection have been achieved by using waveguides fabricated in photosensitive glass using a UV laser.⁸ Furthermore, the ease of integration with other photonic components has meant that laser-written circuits can be used as a means to bridge multiple technologies (see Fig. 2).⁹

FIGURE 2. A conceptual image of a photonic source shows a circuit-detection network enabled by laser-inscribed waveguides [9].

Quantum computing is an extremely exciting prospect. However, it is a challenging goal, and in the short term there are a number of stepping stones that promise to significantly disrupt current computing and communication technologies. One of these is quantum simulation, where photonic analogues offer enormous potential to provide simulations of everything from photosynthesis to chemistry.¹⁰

Laser-written circuits have been transformative in the study of quantum simulation; groups based in Sydney, Australia; Milan, Italy; and Jena, Germany have demonstrated waveguide arrays intended for the study of photon correlations. These circuits use waveguides that, when placed in close proximity, allow energy to be exchanged and can be used to simulate lattices resembling, for instance, graphene.¹¹

Because of the potential to interface multiple different device platforms, it is likely that future small-scale quantum circuits will exploit the convenience of laser-written devices. However, there is still scope to push into hitherto-unexplored regimes of both high-refractive-index contrast (greater than 2%) and large nonlinearity, both critical for increasing circuit density.

For example, McMillen et al. have shown significant progress toward developing low-loss photonic circuitry in a chalcogenide glass.¹² This material, composed of gallium lanthanum sulphide, is already available at low cost; a laser-inscription group in Edinburgh, Scotland, has begun to explore the possibilities that this material holds, such as high nonlinearity and mid-infrared transparency.¹³

These examples show the advantages and prospects of using laser-written waveguides for not only photon manipulation, but also production and detection. Challenges such as large nonlinearities and high-index-contrast waveguides are being researched and a range of small-scale-quantum computations have already been completed. This puts the FLDW technique among the few contenders that in the short term could produce an entirely monolithic demonstration of a basic quantum processor.

REFERENCES

1. A. Politi, et al., Sci. 320 (5876), 646-649 (2008); doi:10.1126/science.1155441.

2. T. Meany, et al., Appl. Phys.A 114 (1), 113-118 (2013); doi:10.1007/s00339-013-8090-8.

3. T. Meany, et al., Opt. Express 20 (24), 26895-26905 (2012); doi:10.1364/OE.20.026895.

4. K.M. Davis, et al., Opt. Lett. 21 (21), 1729-1731 (1996).

5. Y.J. Bellouard, Laser Micro/Nanoengineering, 7 (1), 1-10 (2012); doi:10.2961/jlmn.2012.01.0001.

6. J. Spring, et al., Opt. Express 21 (11), 5932-5935 (2013); doi:10.1364/OE.21.013522.

7. A. Crespi, et al., Nat. Commun. 2 (566) (2011).

8. T. Gerrits, et al., Phys. Rev. A. 84 (6), 60301 (2011).

9. T. Meany, et al., Laser Photon. Revi. 8 (3), L42-L46 (2014).

10. B.P. Lanyon, et al., Nat. Chem. 2 (2), 106-111 (2010); doi:10.1038/nchem.483.

11. A. Crespi, et al., New J. Phys. 15 (1), 013012 (2013); doi:10.1088/1367-2630/15/1/013012.

12. McMillen et al., Opt. Lett. 37 (9), 1418 (2012).

13. J.E. McCarthy, et al., Opt. Express 20 (2), 1545-1551 (2012).

Thomas Meany completed his Ph.D. at the Centre for Ultrahigh-Bandwidth Devices for Optical Systems (CUDOS), Macquarie University, North Ryde, Australia; email: [email protected]; http://physics.mq.edu.au.