GB2449689A

GB2449689A - Zoned optical fibre

Info

Publication number: GB2449689A
Application number: GB0710437A
Authority: GB
Inventors: Michael Charles Parker; Makiko Hisatomi
Original assignee: Fujitsu Ltd
Current assignee: Fujitsu Ltd
Priority date: 2007-05-31
Filing date: 2007-05-31
Publication date: 2008-12-03
Anticipated expiration: 2027-05-31
Also published as: GB2449689B; GB0710437D0

Abstract

An optical fibre 400 (500; 600; 700; 900; 1000) comprising a core 410 (510; 610; 910; 1010) and at least one cylinder 450 455 (550, 555; 650, 655; 755; 950, 955; 1050, 1055) together forming a plurality of zones. In use, the zones have a primary optical function of confining light within the fibre 400 (500; 600; 700; 900; 1000). Successive zones have different refractive indices and different widths. At least one of the zones comprises a material that exhibits a second optical function such as optical gain or controllable refractive index.

Description

A Method and Apparatus Relatingto Optical Fibres

Field of the Invention

This invention relates to optical fibres. More particularly, the invention relates to an optical fibre comprising a core and a plurality of concentric cylinders surrounding the core, the core and the plurality of cylinders together forming a JO plurality of zones that confine light within the fibre, and wherein successive zones have different refractive indices and have a width that varies with radius. Even more particularly, although not exclusively, the invention relates to radially chirped Bragg fibres (RCBF)

Background Art

Optical fibres are used to guide light signals. They are used in many different applications, including sensing, short-range precision delivery of light and long-distance communications.

When an optical fibre is used to guide a light signal, the optical properties of the fibre can have a significant effect on the guided signal. For example, when a light signal travels over long distances in an optical fibre, various undesirable effects can occur, including attenuation of the signal due to losses in the fibre, and chromatic dispersion (different frequencies of the light travelling at different speeds) . When high-power light signals are transmitted, problems can result from non-linearities in the refractive index of the material from which the fibre is made; at very high powers, the fibre can suffer catastrophic failure due to self-focusing effects.

There is increasing interest in hollow-core optical fibre (i.e. fibre having an air-filled or vacuum core), due to its many applications. Such fibre presents reduced attenuation loss to light guided down the hollow core, since material absorption losses are greatly reduced. Similarly, problems associated with high-power signals are greatly reduced.

Hollow-core fibre can also be used to guide solid objects such as atoms (e.g. entangled molecules for quantum computing), ions (excited-state transport), carbon nanospheres or other particulate matter from one location to another. It can also be used for sensing applications (remote sensing), since the fibre's core can be filled with a microfluid, so that the fibre's optical properties (e.g. absorption, degree of evanescence, and modal field diameter) are influenced by the presence of specified molecules in the microfluid.

US 7110649 (Fujitsu Ltd) and Makiko Hisatomi, Michael C. Parker, Stuart D. Walker, "Binary Multi-Zoned Microstructured Fiber: A Comparative Dispersion Analysis of Radially Chirped Bragg Fibre, J Lighwave Tech 23, pp. 3551 -3557 (20005), the whole contents of which are hereby incorporated by reference, describe an unusual kind of optical fibre, comprising a number of concentric cylinders, or "zones" of optical material. At the junction of two zones the refractive index changes between lower and higher values discontinuously. The radius of the mth zone is: rm = In some ways that structure is reminiscent of a Fresnel zone plate, which of course is a plate having a number of concentric zones. The zoned structure has a focusing property, as does the Fresnel zone plate, and also the Fresnel zone plate has a square root variation for the radii of its zones.

Thus in Fig. 3 of the present drawings, an example RCBF has zones 250, 255 that again decrease in width according to the above square-root law, but the index is constant within each zone 250, 255, so the structure is binary in its refractive index. Thus core 210 and 2, 4th and subsequent even-numbered rings 255 are made from a material having a first, higher, refractive index; the 15t, 3rd, subsequent odd-numbered rings 250, and outer cladding 240 are made from a material having a second, lower refractive index.

Fibres substantially of the kind described in US 7110649 are described herein as "radially chirped Bragg fibres" (RCBFs) : their key feature is the radial chirping of the radial widths, i.e. the progressive decrease in the widths with increasing radius.

is Chromatic dispersion is a phenomenon that causes different wavelengths of light in a signal to travel at different speeds, causing chirping of the signal's spectrum and broadening of signal pulses. Broadening of optical pulses causes an increase in the time that must be left between signal pulses in order for them to remain distinguishable, and hence a decrease in the useable signal bandwidth provided by the fibre. If dispersion is left uncorrected, pulsed optical signals will, over sufficient distances, break up. Known methods of addressing chromatic dispersion include transmitting signal pulses at a signal wavelength at which the optical fibre exhibits zero dispersion (unfortunately, fibre losses tend to be relatively high at that wavelength), and interspersing lengths of "standard" fibre (the principal fibre used for long-haul telecommunications) with lengths of dispersion compensation fibre (DCF), which is fibre that exhibits dispersion having the opposite sign to the dispersion exhibited by the standard fibre, so that the net dispersion experienced by the signal pulses is zero.

For "next-generation" high speed (>40Gbps) optical networks, dispersion is potentially a serious problem, as at high data rates optical networks are sensitive to chromatic dispersion. The networks require tuneable devices, but DCF5 can only compensate a fixed amount of dispersion and are not tuneable; hence they are not suitable for dynamic, high-speed optical networking.

More generally, next-generation, high-speed networks would benefit from flexible, dynamic components that provide rapid network reconfigurations that are actively controlled and do not require network operators to keep extensive stock of fixed-property alternative components. One particular drawback of DCFs (or any fixed dispersion compensator) is that a large number of spare components is required, in order to cover the whole range of desired dispersion with different lengths of DCF. Provision of a variable dispersion compensator (VDC) would enable dynamic re-configuration of a network, and hence reduce operational costs.

Various VDC devices have been suggested, including the virtually-imaged phased array (VIPA), fibre Bragg gratings (FBGs), arrayed waveguide gratings (AWGs), lattice filters, and electronic compensation. Although several such ideas have been known for some time, they are still immature technologies, and they tend to have strongly constrained bandwidth-dispersion products (i.e. higher ranges of variable dispersion tend to be available only for narrower bandwidths) VIPA is the most developed solution for tuneable dispersion compensation; however, it is rather large in size, power-hungry, and slow, as its switching is mechanical. FBGs are a fibre-based technology, and hence coupling to standard single-mode fibre (SMF) is straightforward; however, ripple in both the amplitude response and group-delay response is generally high across the device's passband. AWGs and other planar light circuit (PLC) technologies are promising, as they have the potential to miniaturise photonic components; however, those technologies have a smaller tuning range, and are still immature.

In most optical fibres in use today, including SMF (see SMF 1 in Fig. 1), propagating light is mostly confined to a core 10, which is surrounded by a layer of cladding 20. Both the core 10 and the cladding 20 are typically glass, but the core 10 is made of a higher-refractive-index material, whilst the cladding 20 is of a lower-refractive-index material, so that light in th& core 10 undergoes total internal reflection at the core-cladding interface. (Most optical fibres are encased in a protective jacket 30, which is not shown in the accompanying drawings other than Fig. 1, for ease of is illustration.) Another class of fibres is the Bragg fibre (see Bragg fibre 100 in Fig. 2), a fibre in which a core 110 is surrounded by concentric rings 120, 125 of material, with every second ring 125 being of the same material as the core 110 and having the same width as each other second ring 125, and every other ring 120 being of a different material from the second rings 125 and having the same width as each other ring 120. Bragg fibres show superficial similarity to RCBFs but, despite those similarities (and their similar names), they guide light by a very different mechanism. As discussed above, RCBFs guide light by periodic refocusing, a phenomenon resulting from the Fresnel-lens-like nature of their zonal structure. Bragg fibre, on the other hand, consists of concentric rings of different refractive indices, where the incremental radius of each ring represents a constant quarter-wavelength path length change such that light at the Bragg wavelength is reflected back into the core by the photonic bandgap effect; they do not exhibit the characteristic periodic refocusing of RCBF. RCBF is not translationally periodic, and does not exhibit a photonic band gap (at least not in a conventional manner); one cannot readily understand the properties of RCBF using known theories of the operation of Bragg fibres.

W02005/053116 (Corning) teaches a double-clad, hollow-core, Bragg fibre (i.e. a Bragg fibre in which there are two groups of the concentric rings having different thicknesses, forming an outer cladding region and an inner cladding/outer core region) . The inner cladding/outer core region includes an inner, interface, portion in which there are a plurality of regions of optically active dopants. In an alternative embodiment, the optically active dopants are disposed in the hollow core of the fibre. The fibre is said to work by encouraging coupling of energy from surface modes at the core/cladding interface to modes in the core. However, use of surface modes is inefficient; normally, one aims to suppress them. The fibre has to be designed to encourage their excitation and the cross-coupling. Moreover, the energy coupling can occur in both directions, with energy being lost from the core modes to the surface modes.

Fig. 4 shows dispersion characteristics of examples of SMF 330, Bragg fibre 320 and RCBF 310. In Fig. 4, the abscissa (the normalised frequency V) is given by rai2 2 V=-j--in1 -n2 and the normalised propagation constant b is given n2 -n2 by b= ô with neff being the effective refractive index of nI -n2 the mode and a being the appropriate diameter of the structure. The ordinate is a measure of the normalised dispersion exhibited by each type of fibre.

In recent years, a great deal of attention has been focused on another class of optical fibre, called "photonic crystal fibres" (PCFs). Various PCF structures have been proposed. Most of those structures include a core, which may be solid, or an elongate hole, running along the length of the fibre, and a cladding comprising a plurality of elongate holes embedded in a matrix material, which, in the case of a solid-core fibre, is typically the same material as the core of the fibre. The cladding holes are typically arranged in a periodic pattern, usually a lattice having six-fold symmetry.

The fibres typically guide light by one of two mechanisms. In the first, the cladding exhibits a photonic band-gap, so that light of at least one range of frequencies is unable to propagate in the cladding, essentially due to interference is effects. In the second, light is confined to the core by total internal reflection, as in standard fibres, with the required index difference resulting from the effective refractive index of the cladding being lower than that of the core as a result of the holes in the cladding.

Sigang Yang, Yejin Zhang, Shizhong Xie in "Transformation of a single mode photonic crystal fibre into a tunable dispersion compensator", We3-Pl, ECOC 06, Cannes, France, September 2006, discloses a fibre having translational symmetry, a periodic hexagonal structure, which is not chirped. The fibre is a VDC photonic crystal fibre (VDC-PCF) which has a large negative dispersion when a specific set of holes is filled with index-tuneable polymer. In their scheme, the dispersion of the fibre is made tuneable by changing the refractive index of the polymer. The VDCPCF is a "holey fibre" of the kind discussed above: the fibre is fabricated from silica glass, with an array of micro-scale holes that run along the entire length of the fibre. The arrangement of the

S

holes is periodic, with the holes arranged on a triangular lattice, giving six-fold rotational symmetry.

In common with other holey fibres, the Yang fibre has a complex dispersion map, with its dispersion depending significantly on the size and distribution of the cladding holes; the underlying principles of its geometry, including its guidance mechanism -the generation of a photonic band gap -are complex.

The present invention seeks to ameliorate at least some of the above-mentioned problems.

Disclosure of the Invention

In a first aspect, the invention provides an optical fibre Is comprising a core and at least one cylinder concentric with and surrounding the core, the core and the at least one cylinder together forming a plurality of zones, said zones having a primary optical function, in use, of confining light within the fibre, and successive zones having different refractive indices and different widths; CHARACTERISED IN THAT at least one of the zones comprises a material that exhibits a secondary optical function.

Preferably, at least one of the cylinders comprises a material that exhibits the secondary optical function. The core may or may not exhibit the secondary optical function.

The fibre does not guide light by virtue of a conventional (periodic) photonic band gap. Preferably, the fibre does not include cladding holes.

The zones are of course formed in the transverse cross-section of the fibre, and extend along at least a major part of its length. The cylinders may have a circular cross-section, or they may have a non-circular cross-section.

The widths of the zones may decrease with increasing radius. The decrease may be monotonic. The outer diameter of the mth zone may be the radius of the core times the square root of m. (Where herein the value of a physical parameter is specified mathematically, that value is of course accurate only within the reasonable limits of manufacture and/or measurement; thus, for example, fibres in which the outer diameter of the mth zone is only substantially equal to the radius of the core times the square root of m are encompassed by the statement in the preceding sentence. Also, the square-root relationship presupposes closely similar refractive indices for the different zones, such that overall optical path lengths are in the same proportion as geometric path lengths; where the optical and geometric path lengths differ is significantly, the optical path length is preferred.) The outer diameter of the rnth zone may be a constant plus a given radius times the square root of m.

The widths may increase with increasing radius. The increase may be monotonic.

There may be two cylinders in addition to the core.

There may be more than two cylinders in addition to the core.

There may be five cylinders in addition to the core. There may be nine or ten cylinders in addition to the core. There may be more than ten cylinders in addition to the core.

It may be that every other zone has same refractive index; i.e. in an example fibre with five cylinders, the core, 2 cylinder, and 4th cylinder may have the same refractive index and the ring, 3rd ring, and 5th ring may have the same refractive index.

The material exhibiting the secondary optical function may be, for example, a polymer (e.g. PMMA -poiy methyl methacrylate) or a glass (e.g. Erbium-doped glass) The core may be a hollow core. The core may be filled with a fluid, that is a liquid or a gas. There may be a vacuum in the core.

In a group of example embodiments of the invention, the secondary optical function is optical gain. Thus, at least one of the cylinders may be formed from an active material.

That active cylinder may be the first cylinder. The core may include an active material, which may be solid or liquid.

The active cylinders may be separated from each other by passive (non-active) cylinders. Although this group of embodiments is concerned with amplification of modes, in contrast to the Corning paper discussed above, it does not require excitation of surface modes, and so the active cylinders need not be adjacent to the core, or continuous with each other.

The secondary optical function may be a controllable optical property. In another group of example embodiments of the invention, the secondary optical function is controllable refractive index. Controllable refractive index can be used to enable overall dispersion control.

Returning to Fig. 4, one can see that if the range of V d2(Vb is chosen where V " / is positive then RCBF 310 has negative dV2 dispersion, which can be used for dispersion compensation.

The inventors have noted that in the vicinity of the cut-off frequency 350 (i.e. the normalised frequency above which the fibre ceases to be single-mode), which in the example of Fig. 4 is just under V=5, the dispersion of the RCBF 310 is significant; in contrast, Bragg fibre 320 and SMF 330 have a dispersion close to zero at that frequency. The inventors have realised that that relatively high dispersion near the cut-off wavelength 350 makes RCBF 310 a good choice of fibre for dispersion compensation. In use, an optical fibre will 11 tend to guide lowest-order-mode light most effectively when the light is as close as possible to the cut-off wavelength for higher modes. The high dispersion shown at the cut-off wavelength 350 in Fig. 4 is therefore particularly advantageous.

The material that exhibits the secondary optical function may be a material with similar thermal expansion to a material from which the rest of the fibre is comprised, but a different rate of change of refractive index with temperature (e.g. to appropriately engineered PMMA) . The material that exhibits the secondary optical function may be a material the refractive index of which changes when a voltage is applied across the material. The material that exhibits the secondary optical function may be a polymer incorporating a non-linear IS chromophore. Examples of suitable arrangements are given for example in M. Oh, et al., "Recent Advances in Electrooptic Polymer Modulators Incorporating Highly Nonlinear Chromophore", IEEE Journal of Selected Topics in Quantum Electronics 7, p826 (2001) and G. Xu, et al., "Organic electro-optic modulator using transparent conducting oxides as electrodes", Optics Express 13, p7380 (2005) One or more of the zones may exhibit a secondary optical function in the form of a controllable refractive index.

Every other zone may have a controllable refractive index. It may be that only the central zone (i.e. the core) has a controllable refractive index. It may be that the central zone and the first zone have controllable refractive indices.

The material may exhibit an optical non-linearity, for example a non-linear refractive index.

The size of the non-linear effect may be controllable.

The fibre may, in use, guide light by periodically refocusing it. The period of the refocusing may be chosen to achieve quasi-phase-matching (QPM) with the optical non-linearity. Thus, the geometry of example embodiments of the invention, for example the RCBF geometry, may be exploited to enhance by QPM the efficiency of optical nonlinear effects.

That may be achieved by phase-matching of modes in a multimoded waveguide structure; thus, the modes may become spatially and periodically "mode-locked". In mode-locking, the propagation constant for the modes differ by a constant increment, so that the modes come back into phase after a length determined by the constant.

The fibre may be a fibre for use in a communication system.

In a second aspect, the invention provides use of a fibre according to the first aspect of the invention to provide varying dispersion compensation in an optical network.

As discussed above, since an appropriate compensation amount has to be calculated for each link in network, fixed dispersion DCF does not enable dynamic reconfiguration. Also, the compensation required can vary from link to link, so a large number of spare DCFs is required; hence, a large inventory is required, and there is a lack of flexibility.

Use of a fibre according to an example embodiment of the invention can enable a drastic reduction in inventory, and hence in maintenance costs.

Fibres according to embodiments of the invention may provide a large tuning range. As discussed above, next-generation high-speed (>40Gbps) networks with dynamic and flexible network configurations are more sensitive to dispersion than legacy networks; hence tuneable dispersion is wanted. Example embodiments of the invention may provide more efficient use of network capacity by enabling dynamic reconfiguration of the dispersion compensation provided in the network, and hence reduced operational costs.

Thus, in a third aspect, the invention provides an optical network comprising a fibre according to the first aspect of the invention.

Embodiments of the invention may be especially useful in coherent transmission systems. Thus, the fibre may be used to provide a coarse dispersion control, in conjunction with another dispersion compensator used to provide fine dispersion control. The other dispersion compensator may be a digital signal processor.

In a fourth aspect, the invention provides use of an optical fibre according to the first aspect of the invention, the fibre having a cut-off wavelength, the use comprising propagating along the fibre light having a wavelength within nm of the cut-off wavelength to achieve a dispersion of more than 1000 ps/nm.

The dispersion may be more than 2000 ps/nm. The wavelength may be within 100 nm of the cut-off wavelength.

In a fifth aspect, the invention provides a method of dispersion compensation, the method including propagating light in a fibre according to the first aspect of the invention, at a wavelength near to a cut-off wavelength of the fibre.

The dispersion may be more than 1000 ps/nm. The wavelength may be within 200 nm of the cut-off wavelength.

In a sixth aspect, the invention provides a method of amplifying a light signal, comprising propagating the light signal along a an amplifying fibre as described above and being a fibre according to the first aspect of the invention, and using the optical gain to supply energy to an evanescent part of a mode of the fibre core.

The mode may be the fundamental mode of the fibre core.

In a seventh aspect, the invention provides an amplifier, a laser or a sensor including a fibre according to the first aspect of the invention.

It will be appreciated that features of the present invention described in relation to any aspect of the present invention are equally applicable to the other aspects of the present invention.

Brief Desct-iption of the Drawings l0 Certain illustrative embodiments of the invention will now be described in detail, by way of example only, with reference to the accompanying schematic drawings, in which: Fig. 1 is a transverse cross-section through an example IS of a prior-art standard single-mode fibre (SMF); Fig. 2 is a transverse cross-section through an example of a prior-art Bragg fibre; Fig. 3 is (a) a transverse cross-section through an example of a prior art, solid-core, Fresnel-zoned (RCBF) fibre, and (b) a plot of the radial variation of the refractive index of the fibre; Fig. 4 is a plot showing the normalised dispersion characteristics of the prior-art fibre of Figs. 1 to 3; Fig. 5 is a transverse cross-section through a first example of a fibre according to an embodiment of the invention Fig. 6 is (a) a transverse cross- section through a second example of a fibre according to an embodiment of the invention, being a solid-core, nonlinear, RCBF, and (b) a corresponding refractive-index profile with ri=1.67 microns; Fig. 7 is (a) a transverse cross-section through a third example of a fibre according to an embodiment of the invention, being another solid-core, nonlinear RCBF, and (b) a corresponding refractive-index profile with Pi3.92 microns; Fig. 8 is a 2D plot showing modal evolution along 55 micron lengths of the fibres of Figs. 6 and 7, with: (a) r1-l.67 microns, input light source D5=9 microns, 1\=17 microns, (b) p1=3.92 microns, input light source D=9 microns, 1 =30 microns, (C) Pi=392 microns, input light source D5=6 microns, I\ =43 microns; Fig. 9 is a schematic longitudinal cross-section through a fourth example fibre according to an embodiment of the invention, incorporating electro-optically active material; Fig. 10 is a block diagram showing how variable dispersion compensation may be used in conjunction with digital signal processing; Fig. 11 is (a) a transverse cross-section through a fifth example of a fibre according to an embodiment of the invention, being a hollow-core, amplifying RCBF, and (b) a corresponding refractive-index profile; and Fig. 12 is a transverse cross-section through a sixth example of a fibre according to an embodiment of the invention, being a solid-core, amplifying RCBF, including amplifying material in its core.

Detailed Description

As discussed above, a radially chirped Bragg fibre (RCBF) features concentric zones of alternating materials along the fibre length, with the radius of the mth zone given by a square-root relationship. The radii of RCBF zones are not periodic, and RCBF guides light by periodically refocusing the light within the fibre structure.

The zoned structure can be regarded as a longitudinal equivalent of a binary Fresnel zone plate, such that light is guided by periodic refocusing back within the fibre structure.

Alternatively, the periodic refocusing of the light within the fibre core can be viewed as being analogous to the Talbot self-imaging effect, exploited within multi-mode interference (MMI) couplers. However, rather than employing an essentially periodic grating as is required for the 1D case, the cylindrical geometry of a fibre requires aperiodic gratings to achieve the same self-imaging effect.

An example RCBF utilises a simple Fresnel zoning geometry to achieve the refocusing effect. Figs. 6 and 7 show schematic cross- sections of two example solid-core RCBF waveguide geometries. Each fibre 500, 600 includes a material with a nonlinear refractive index in its core 510, 610.

In the fibre 500 of Fig. 6, the centre (geometric) radius (i.e. the radius of core 510) is r1 = 1.67 microns, with a core refractive index of n2=2.25. Zones 555 have the same index, i5 whilst the index of zones 550 and outer cladding 540 is n1=2.35. There are 9 zones 550, 555. The fibre 600 of Fig. 7 has a larger central zone 610, having a radius p=3.92 microns.

However, rather than having the simple zone radii of the Fresnel formula, this fibre 600 has the same zone radii increments as the fibre 500 of Fig. 6, such that the radii of subsequent zones 650, 655 are given by pm=pi+(rm-ri), where p=3.92 microns, r1=1.67 microns, and rm follows from the square-root Fresnel formula. (Outer cladding 640 is again not part of the Fresnel pattern.) We have used a fully vectorial 3D finite-difference time-domain (FDTD) package, E'oynting for Optics, to simulate electric field evolution in our RCBF designs. Fig.8(a) shows the evolution of a linearly-polarised D5= 9 micron width Gaussian input mode source, having a wavelength A = 1.55 micron along a 55 micron-length of the fibre 500 shown in Fig. 6. The periodic refocusing of the light is a marked effect, with a spatial period of A=17 microns within the low refractive index core 510 of the fibre 500. The vertical lines indicate the boundaries of the zones 550, 555. In the FDTD simulation, time steps were dt=133as, whilst a minimum spatial resolution of dx=dy=lOOnm for transverse dimensions, and dz=5Onm (along the fibre axis) were employed. Fig. 8(b) and 8(c) show how the spatial period A can be tuned by varyingboth fibre geometry and also the Gaussian diameter D5 of the illuminating light. In this case, the period A can be increased by either increasing the radius of the central zone 510, e.g. A can be increased to A=30 microns for the central ID zone radius of p1=3.92 microns of fibre 600. Furthermore, reducing the diameter of the source light to D5=6 microns also increases the spatial period further to A=43 microns. In those two latter cases, the FDTD simulation parameters were increased slightly to dt=144as and dx=dy=l3lnm, with dz=5Onm, in order to accommodate the larger fibre structure 600 within the computing memory resources.

The periodic refocusing may thus be used to achieve phase-matching of nonlinear effects resulting from the nonlinear refractive index of core 510, 610.

In the example embodiment of the invention shown in Fig. 5, a RCBF 400 has a similar structure to the RCBF 200 of Fig. 3. RCBF 400 comprises a zeroth zone, inner core 410 made of a material having a first refractive index, and a plurality of concentric, annular zones or rings, the even-numbered ones 455 of which have another refractive index, and the odd-numbered ones 450 of which have a third refractive index (as does outer cladding 440) . The radius of the mth zone is again given by the square-root relationship.

The outer radius of the outermost of the annular zones has in this example a diameter equal to that of standard SMF; i.e., 9-10 microns.

Of course, in alternative embodiments, the fibre of the invention could have a different structure, for example the structure of fibre 500 or fibre 600, or a hollow-core structure.

A fibre having the structure of fibre 400, and zones 455 that exhibit a secondary optical function, is an example of an embodiment of the invention.

With reference to Fig. 5, the inner core 410 is passive but each even-numbered ring 455 is made of an active material.

When that active material is an amplifying material this offers various advantages such as: 1) differential gain of the evanescent tail, leading to overall higher amplification; 2) highly efficient fibre laser; In addition if the inner core is hollow this offers the additional advantages: 3) ultra-high light intensities at centre of hollow core, without fears of material damage due to the high concentration of optical power; 4) spatially different optical pumping (i.e. in the cladding), compared with the main lobe of signal light (at centre of hollow core) Example applications of such a device include high-sensitivity sensing, ultra-high power optical fibre laser for use in welding, and concentrated high-intensity optical power delivery.

The advantage for high-intensity optical power delivery is that the majority of the optical power of the mode resides at the centre of the hollow-core, and only the evanescent tail finds itself in the material surrounding the hollow core. If the surrounding material is active and offers optical gain, then it will amplify only the relatively low power of the evanescent tail. Since that power is relatively low, the active, amplifying material is less likely to suffer from gain saturation effects, and so can provide a high gain. The guiding mechanism of the RCBF acts to maintain the overall shape of the guided mode, in both hollow-core and surrounding material. Hence, as the evanescent tail increases in power, optical power is automatically shifted from the surrounding material towards the centre of the hollow core, so as to maintain the same overall modal shape of the guided mode.

Hence the evanescent tail is continually losing power towards the hollow-core, so that it will continue to benefit from the high differential gain afforded by its relatively lower power.

In such a way, very high gains and optical powers may be achieved, before the evanescent tail itself accumulates sufficient optical power to start causing gain saturation effects within the amplifying medium.

By bringing the two ends of the hollow-core MSF together into a closed ring configuration, or simply placing reflectors at either end of the hollow core device, a laser may be constructed. For the ring laser case, power must be coupled out of the device either by evanescent coupling (which typically involves passage of the light through the material side of the hollow-core fibre); or by introducing a "Y"-shaped hollow-core splitter within the ring configuration, so that a proportion of the light is automatically fed out, yet remains within an air-core geometry (to avoid any dangers of material damage due to the high optical intensities) A significant aspect of the potential of the device for sensing applications lies in its enhanced sensitivity to very low concentrations of an arialyte under consideration. The amplifying medium can increase the power of the sensing signal (e.g. the Raman signature, if Raman spectroscopy is used to analyze the solution in question) so as to make it easier to measure.

The geometry of the RCBF also allows more efficient optical pumping to invert the active medium. Often, in fibre amplifier geometries, the optical pumping power is located at the centre of the waveguide, alongside the optical signal to be amplified. However, by making the first annular zone around the hollow-core the active region, a number of benefits arise: the total cross-sectional area of the active region is potentially higher than that of the hollow-core, which allows greater efficiency of inversion of the amplifying medium, since the pumping power is at a lower intensity and is less likely to saturate the inverting material. It also therefore allows a higher optical pumping power, as there are less likely to be nonlinear effects occurring in the cladding regions, which might reduce efficiency and cause unwanted side-effects (e.g. spurious harmonic generation, and cross-phase/gain modulations etc.) Additional filtering to separate the signal from the pump is also therefore unnecessary, so reducing device complexity. Of course, some of the advantages are also present if a passive solid-core design (rather than the hollow-core geometry) surrounded by active zones/cladding is adopted. In that case, the avoidance of unwanted non-linear effects in the core is still achieved, as well as the greater efficiency and high optical power advantages already noted.

Ultra high-power fibre lasers are becoming increasingly important, but the high intensities can cause problems for solid-core devices (e.g. end-facet damage, and any impurities or defects cause reflections and higher than expected local concentration of optical power leading to local heating and possible fracture or damage) . However, by employing a hollow-core fibre where the majority of the optical power is concentrated at the centre, where there is no material, many of these problems are avoided, so allowing ever higher optical powers to be employed, e.g. for spot-welding applications, or other high energy concentration applications.

Complex combinations of active and passive zones are also possible. For example, with the hollow-or solid-core geometries, all odd zones (i.e. first cylinder or annulus surrounding the core, followed by the 3 annulus etc may be made active, with even zones (i.e. zeroth zone or core and 2 zone etc) made passive. Conversely, all even zones (i.e. the central solid-core and 2nd, 4th zones etc.) may be made active and the odd-order zones made passive. Other combinations are possible, e.g. the first zone and all consequent higher-order zones may be made active, with only the core made passive.

IS The waveguide dispersion of the fibre of Fig. 3 is given by: D/L=_12)V'"'1') [ps/nm/kin) c dV2 It is interesting to note why, by employing an essentially "radially chirped" NSF geometry, our RCBF design exhibits a strong dispersive characteristic, as noted above.

It can be considered by analogy to chirped FBGs for dispersion compensation, where the associated longitudinal Bragg grating is chirped along the fibre axis and the light is chromatically dispersed by the grating as it propagates along the fibre. In the case of the RCBF geometry, the chirping is transverse to the fibre axis (i.e. transverse to the propagation of the light.) Intuitively, that might be thought therefore not to impart any dispersion onto the light. However, transversal and longitudinal grating structures are essentially equivalent, imparting similar filtering and dispersion characteristics onto the light. Given that, one can understand why a radially chirped MSF structure also exhibits a strong dispersive nature.

Our RCBF fibre exhibits no periodicity (e.g. translational, or radial) and relies on the Fresnel (re)focusing effect in order to confine light to the centre of the fibre and effect waveguiding. Variation of the Fresnel lens causes the lensing properties (e.g. focal length, degree of chromatic aberration) of the fibre to change. Variation of the Fresnel lens is caused by introducing a variable refractive index in the various zones of the Fresnel lens.

The RCBF fibre has a circular symmetry due to the overall cylindrical geometry of the fibre; hence the index variations should also maintain that circular, rotational symmetry.

Varying the refractive index of the zones changes the lens properties of the RCBF and hence changes the dispersive characteristics of the fibre, and so allows variable dispersion compensation.

In contrast, the Yang approach to VDC (described above) employs a photonic crystal fibre (PCF) geometry. Here, the PCF causes a photonic band (PBG) to arise due to the translational periodicity of its holey hexagonal symmetry.

The PBG confines the light to the centre of the fibre and causes waveguiding. A photonic bandgap is analysed using a dispersion diagram/map, which plots the variation of propagation constant with wavenumber. By introducing variations in the holey structure of the PCF geometry, the dispersion diagram can be modified, so as to cause an overall change in the dispersion properties of the fibre. In order to preserve the cylindrical, rotational symmetry of the fibre, any variations in the holey structure must also have rotational symmetry: hence the variation in holey refractive index exhibits a hexagonal, 6-fold rotational symmetry.

Varying the refractive index of the holey hexagonal ring changes the dispersion map of the fibre, and hence allows variable dispersion compensation.

Thus, compared with the Yang fibre discussed above, RCBF fibre according to an example embodiment of the invention has a fully "concentric-circular" design, with full circular (rotational) symmetry, without holes. Rather than relying on holes in a translationally periodic (often hexagonal) geometry to create a photonic bandgap and hence confine and guide the light, the fibre employs concentric circles, with aperiodic zoning (i.e. non-periodic geometry) to guide the light by repeated refocusing, analogous to a longitudinally extended Fresnel zone plate. As such, the design fully (and more efficiently) exploits the circular symmetry required in a microstructured fibre (MSF) to make it dispersive. In having a fully circular-symmetric design, the RCBF achieves a higher dispersion for a given length of fibre, so that shorter lengths, and/or lower control voltages are required, with the additional advantage that a more compact VDC unit is made possible.

As discussed above, RCBFs of the type known in the prior art provide a fixed amount of dispersion when operated near their cut-off wavelength. However, in a dynamically reconfigurable network, signal pulses may arrive at a point having travelled along different network paths, experiencing different dispersions, and hence requiring different dispersion compensation. The fibre of the example embodiment of the invention meets that need by providing an actively variable amount of dispersion compensation.

Thus, in another example embodiment of the invention the inner core 410 and each even-numbered ring 455 is made, not of an amplifying material as discussed above but of a material the refractive index of which is variable, in this example because the material is an electro-optic material.

(Alternatively, only the odd-numbered zones are made of an electro-optic material.) Tuning of the dispersion in RCBF can be achieved by fabricating the fibre with a material of controllable refractive index n1 for every other zone or for particular zones. By having a material whose refractive index can be changed by external control, the dispersion property of the fibre is altered, so that different amounts of dispersion can be compensated, depending on the external control. The to geometry of the fibre 400 of Fig. 5 is one such variable dispersion compensation (VDC) MSF scheme. There are a number of ways in which the refractive index of material can be changed: mechanically, thermo-optically, electro-optically.

Amongst such schemes, electro-optic methods tend to provide the fastest switching response.

An example of an electro-optic approach is shown in Fig. 9. Fibre 700 is connected in line with two standard fibres 705. Electro- optic tuning of the dispersion of fibre 700 is achieved by creating a potential difference along its length, within each even-numbered ring 755, using indium tin oxide (ITO) electrodes 790. Electrodes 790 are arranged at each end of the fibre 700, surrounding the fibre in the vicinity of its join to the standard fibres 705.

This embodiment of the invention aims to achieve tuneable dispersion compensation in the well-established DCF regime, without the conventional bandwidth-dispersion constraint; however, this is probably at the expense of size and speed.

That said, it is to be expected that future optical transmission systems will still require relatively coarse and (possibly slow) VDC, in conjunction with a higher-speed, fine dispersion compensation system. For example, future coherent detection transmission systems can theoretically compensate for all dispersion using digital signal processing (DSP) techniques. But it would still make sense to employ optical VDC using the invention for "coarse" and relatively slow dispersion compensation and to use the DSP for residual, higher-speed dispersion compensation. Such an approach takes away a lot of the load from the DSP (allowing it to perform other important additional functions more efficiently), and uses relatively straightforward optics to perform a useful, underlying compensation function. A discussion of the use of DSP techniques for variable dispersion compensation is given in S.J. Savory et al., "Transmission of 42.8Gbit/s Polarization Multiplexed NRZ-QPSK over 6400km of Standard Fiber with no Optical Dispersion Compensation", Proceedings of Optical Fiber Communications conference (OFC'07), Paper OTuA1, Anaheim, March 2007.

Fig. 10 is a simple 3 block diagram indicating how VDC is used in conjunction with DSP for coarse/fine dispersion compensation. In this example system, a light signal including dispersion that is to be compensated passes first into coarse VDC 800, which includes an active RCBF according to an example embodiment of the invention. The coarsely compensated signal then passes into coherent detector 810, where a local oscillator is used to retrieve the optical phase of the signal. A DSP 820 is then used for fine dispersion compensation of residual optical phase impairments in the retrieved phase of the signal.

It will be understood that, as propagating light is typically most concentrated at the centre of core 410, variations in the refractive index of core 410 have most effect on the dispersion provided by fibre 400. In an alternative example embodiment of the invention, only core 410 has an actively controlled refractive index. In another example embodiment, only the core 410 and ring 455a have actively controlled refractive indices, whereas rings 455b-d do not, and need not be made of a material having a variable refractive index. Similarly, in another example embodiment, only the core 410, and the inner two rings 455 a, b have the variable refractive index.

The core of a fibre according to an embodiment of the invention may be a hollow core, or it may be a solid core that exhibits the secondary function, or it may be a simple, passive solid core.

For example, the fibre 900 of Fig. 11 has a hollow core 910, surrounded by a first set of cylinders 950, which are passive glass, and which alternate with a second set of cylinders 955, which are doped with amplifying, active ions.

Outer region 940 is again passive.

In another example (Fig. 12), a fibre 1000 includes a solid core 1010, which is made of a material having a variable refractive index, with cylinders 1055 being made of the same material. Between the core 1010 and the active cylinders 1050, and between each active cylinder 1055 are provided passive cylinders 1050. Thus, the dispersion provided by the fibre 1000 may be varied by simultaneously altering the refractive indices of the active core 1010 and the active cylinders 1055.

Appropriate methods for manufacturing fibres according to the invention will readily present themselves to the skilled artisan. For example, to form a hollow-core RCBF, hollow cylindrical preforms, each positioned and dimensioned to correspond to a zone in the final fibre, may be assembled concentrically and drawn to appropriate dimensions. The zone or zones having the secondary optical function will in this example be formed from cylinders made from an appropriate material. A solid core RCBF may be produced by a similar method, with the addition of a central solid cane, around which the concentric hollow performs are positioned. * 27

Whilst the present invention has been described and illustrated with reference to particular embodiments, it will be appreciated by those of ordinary skill in the art that the invention lends itself to many different variations not specifically illustrated herein. Some examples of such variations and alternatives have been described above.

Where in the foregoing description, integers or elements are mentioned which have known, obvious or foreseeable equivalents, then such equivalents are herein incorporated as if individually set forth. Reference should be made to the claims for determining the true scope of the present invention, which should be construed so as to encompass any such equivalents. It will also be appreciated by the reader that integers or features of the invention that are described is as preferable, advantageous, convenient or the like are optional and do not limit the scope of the independent claims.

Claims

Claims 1. An optical fibre comprising a core and at least one cylinder

concentric with and surrounding the core, the core and the at least one cylinder together forming a plurality of zones, said zones having a primary optical function, in use, of confining light within the fibre, and successive zones having different refractive indices and different widths; CHARACTERISED IN THAT at least one of the zones comprises a material that exhibits a secondary optical function.
2. A fibre as claimed in claim 1, in which the widths of the zones decrease with increasing radius.
3. A fibre as claimed in claim 2, in which the decrease is monotonic.
4. A fibre as claimed in claim 3, in which the outer diameter of the mth zone is the radius of the core times the square root of m.
5. A fibre as claimed in claim 3, in which the outer diameter of the mth zone is a constant plus a given radius times the square root of in.
6. A fibre as claimed in claim 1, in which the widths of the zones increase with increasing radius.
7. A fibre as claimed in any preceding claim, in which there are two cylinders in addition to the core.
8. A fibre as claimed in any of claims 1 to 6, in which there are more than two cylinders in addition to the core.
9. A fibre as claimed in any preceding claim, in which every other zone has the same refractive index.
10. A fibre as claimed in any preceding claim, in which the material exhibiting the secondary optical function is a polymer or a glass.
11. A fibre as claimed in any preceding claim, in which the core is a hollow core.
12. A fibre as claimed in any of claims 1 to 11, in which the secondary optical function is optical gain.
13. A fibre as claimed in claim 12, in which the at least one of the cylinders is formed from an active material.
14. A fibre as claimed in claim 13, in which that active cylinder is the first cylinder.
15. A fibre as claimed in any of claims 12 to 14, in which the core includes an active material.
16. A fibre as claimed in any of claims.1 to 11, in which the secondary optical function is a controllable optical property.
17. A fibre as claimed in claim 16, in which the secondary optical function is controllable refractive index.
18. A fibre as claimed in claim 17, in which the material that exhibits the secondary optical function is a material with similar thermal expansion to a material from which the rest of the fibre is comprised, but a different rate of change of refractive index with temperature.
19. A fibre as claimed in claim 17, in which the material that exhibits the secondary optical function is a material the refractive index of which changes when a voltage is applied across the material.
20. A fibre as claimed in any of claims 17 to 19, in which one or more of the zones exhibits a secondary optical function in the form of a controllable refractive index.
21. A fibre as claimed in claim 20, in which every other zone has a controllable refractive index.
22. A fibre as claimed in claim 20, in which only the central zone (i.e. the core) has a controllable refractive index.
23. A fibre as claimed in claim 20, in which the central zone and the first zone have a controllable refractive index.
24. A fibre as claimed in any of claims 1 to 11, in which the secondary optical function is an optical non-linearity.
25. A fibre as claimed in claim 24, in which the size of the non-linear effect is controllable.
26. A fibre as claimed in any preceding claim, which, in use, guides light by periodically refocusing it.
27. A fibre as claimed in claim 26 when dependent on claim 25, in which the period of the refocusing is chosen to achieve quasi-phase-matching with the optical non-linearity.
28. Use of a fibre according to any of claims 21 to 23 to provide varying dispersion compensation in an optical network.

io
29. Use as claimed in claim 28, in which the fibre is used to provide a coarse dispersion control, in conjunction with another dispersion compensator used to provide fine dispersion control.
30. An optical network comprising a fibre according to any of IS claims 1 to 27.
31. Use of an optical fibre according to any of claims 21 to 23, the fibre having a cut-off wavelength, the use comprising propagating along the fibre light having a wavelength within nm of the cut-off wavelength to achieve a dispersion of more than 1000 ps/nm/km.
32. A method of dispersion compensation, the method including propagating light in a fibre according to any of claims 21 to 23, at a wavelength near to a cut-off wavelength of the fibre.
33. A method of amplifying a light signal, comprising propagating the light signal along a fibre according to any of claims 12 to 15, and using the optical gain to supply energy to an evanescent part of a mode of the fibre core.
34. A method as claimed in claim 33, in which the mode is the fundamental mode of the fibre core.
35. An amplifier, a laser or a sensor including a fibre according to any of claims 1 to 27.
36. A device substantially as herein described, with reference to Figs. 5, 6, 7, 9, 11 and 12 of the accompanying drawings.