WO1990008970A1

WO1990008970A1 - Method for the manufacture of an electro-optic device

Info

Publication number: WO1990008970A1
Application number: PCT/GB1990/000137
Authority: WO
Inventors: Mark Farries; Martin Fermann; Luksun Li; David Payne; Philip Russell
Original assignee: Plessey Overseas Limited
Priority date: 1989-02-04
Filing date: 1990-01-31
Publication date: 1990-08-09
Also published as: EP0408715A1; GB8902533D0; JPH03504772A

Abstract

An optical fibre is electrically poled by applying an intense external electric field across the fibre core and sustaining this field for a prolonged period of time. In this manner, a permanent linear electro-optic property is imparted to the fibre. Poling may be accelerated by performing this method at elevated temperature. Highest field intensities can be achieved using a fibre having one or two internal electrodes i.e. electrodes incorporated in the fibre cladding. Preferably the internal electrodes are of a metal or metal alloy that is liquid at poling temperature.

Description

METHOD FOR THE MANUFACTURE OF AN ELECTRO-OPTIC DEVICE

TECHNICAL FIELD

The present invention concerns a method for the manufacture of an electro-optic device, in particular a method for the manufacture of an optical fibre having a permanently induced linear electro-optic property and having utility thus for electro-optic device application.

Optical fibre-based electro-optic devices, for example optical switches, modulators and electric field sensors, are desirable because they should exhibit low loss and offer compatibility with existing fibre transmission systems and the like. BACKGROUND

Bulk-crystal based electro-optic devices are well known. In these devices the phase velocity of light propagating therethrough may be changed by the application of a transverse electric field, this resulting in a change in the phase of the emerging light signal. For a given electrode geometry this change in phase is a function of the applied voltage. In the case of the linear electro-optic effect, the phase change is proportional to the applied voltage. In the case of the quadratic electro-optic effect, the phase change varies with the square of the voltage. Electroded electro-optic birefringent crystals have been used between polarisers to serve as optical switches and optical modulators. Both the linear effect (Pockels cell) and the quadratic effect (Kerr cell) have been exploited.

For the linear effect to be present it is well established on theoretical grounds that it is necessary for the material to lack inversion symmetry. Glass materials, glass fibres in general and fused silica optical fibres in particular, normally possess inversion symmetry. Understandably, and so far as known, no linear electro- optic effect has been reported for such materials and structures. The lowest-order electro-optic effect thus far observed in optical fibres is the quadratic (or Kerr) electro-optic effect. This effect usually has been small with the result that thus far fibre-based electro-optic devices have been generally inefficient and of limited utility outside of the laboratory. DISCLOSURE OF THE INVENTION

The present invention is intended as a means of providing an optical fibre having a permanently-induced linear electro-optic property.

Contrary to expectation it is now found that, with a sustained application of an intense electric field, it is possible to impart such a property.

Accordingly there is here provided a method for the manufacture of an electro-optic device, this method including the following procedural steps, namely: providing an optical fibre comprised of a core and cladding; and, applying in a direction transverse to the optical fibre core an intense external electric d.c. poling field for a prolonged period of time sufficient to impart to the optical fibre a permanent linear electro- optic property.

The above phenomenon has now been established as demonstrable for fused silica fibres, which fibres have incorporated cores doped with either germanium or phosphorus or a combination of both. The case of alternative dopants however, is not precluded. It is also thought that a number of other glass compositions, including compound glasses, would exhibit this property to varying degrees.

The linear electro-optic property aforesaid corresponds to an induced second-order non-linearity in the non-linear polarisation of the optical fibre. It is found that the sign of this induced second- order non-linearity (as also that of the linear electro-optic coefficient) is dependent on the polarity of the applied poling field. Moreover, it is found that different dopant materials within the fibre core produce different signs for the second-order non-linearity.

The required prolonged period of time is dependent on the intensity of the poling electric field and on ambient temperature. It is found that the higher the poling field intensity, the shorter the time required to produce the same induced second-order non- linearity. At elevated temperature, a similar acceleration of this poling process is observed. Furthermore, for a given ambient temperature and fixed period of time, the resultant linear electro- optic coefficient is found to increase with increased intensity of the poling electric field. No value of the field intensity necessary for the onset of this poling phenomenon has yet been determined. However, for practical purposes, a lower limit of approximately 20v/μm is postulated. This corresponds to a poling period of time on the order of hours for practical values of linear electro-optic coefficient. The term "intense" as used herein shall be construed accordingly.

The external electric poling field is applied using electrodes located on opposite sides of the fibre core. It is advantageous to provide an optical fibre in which at least one of these electrodes is internal i.e. incorporated within the cladding. This permits closer spacing of the electrodes and therefore higher field intensity for a given applied voltage. Preferably, both electrodes are internal for not only does this permit close spacing but it is also less susceptible to breakdown effects. Commensurate with this aim it is also advantageous to provide electrical contacts at respective ends of the optical fibre. It is also advantageous to use as internal electrode material a metal or alloy that is liquid for the ambient poling temperature. These materials provide better electrodes and permit higher voltages to be applied because they are smooth and provide better surface contact. For example, for poling at room temperature the metal gallium may be employed. Although this metal has a melting temperature of 29.6° C, it can be liquid at room temperature due to its supercooling property.

The poling may be applied uniformally. Alternatively it may be structured. The latter may be achieved using a patterned external electrode, for example one with a comb-like or other periodic configuration. BRIEF INTRODUCTION OF THE DRAWINGS

In the drawings accompanying this specification:-

Figure 1 is a perspective view of a sectioned optical fibre, with internal electrodes, suitable for use in the method disclosed herein;

Figure 2 is a schematic view of the optical fibre shown in the preceding figure, which view shows a preferred arrangement of the internal electrodes;

Figure 3 is a graph illustrating measured optical phase retardation as a function of an applied a.c. test field for a fibre "as drawn", and "as poled" by the method disclosed herein, and illustrates quadratic and linear electro-optic effects, respectively;

Figure 4 is a graph showing measured linear electro-optic coefficients as a function of poling electric field intensities each applied for a fixed period of time; and,

Figures 5 to 8 respectively are perspective views of sectioned optical fibres each with a different external electrode configuration, also suitable for use in the method disclosed herein. DESCRIPTION OF THE PREFERRED EMBODIMENTS

So that this invention might be better understood embodiments thereof will now be described and reference will be made to the drawings aforesaid. The description that follows is given by way of example only.

The method disclosed herein is based on the application of an intense d.c. electric field (the poling field) across the fibre core for a prolonged period of time (the poling time). The intense poling field is best applied by means of internal, conductive electrodes integrated in the cladding region of the fibre (Figures 1, 2) or by using a combination of internal and external electrodes (Figures 5 to 8). By way of reference the reader is directed to United Kingdom Patent Application No. 2,192,289 wherein details of electroded fibre construction are disclosed. By such means it is possible to apply d.c. electric field as intense as 600 volts per micron across the fibre core. Owing to breakdown effects it is extremely difficult to obtain such high intense field using only external electrodes.

The following example is given as an illustration of the production of a fibre having the form of that shown in Figure 1 and 2. In these figures the optical fibre 1 is comprised of a core 3 and cladding 5. A pair of electrodes 7, 9 are incorporated in the cladding 5 and located each side of the core 3.

A fibre preform containing a germanosilicate core (composition 10 mol. % Geθ₂) is made by the standard MCVD technique. The diameters of the preform and the core are approximately 11.5mm and 0.38mm respectively. The refractive index difference is typically 0.77%. Two holes having a diameter of 3mm, are drilled on opposite sides of the core by means of an ultra-sonic diamond- drilling machine. The separation of these holes is approximately 1.7mm. A fibre is drawn from this preform to a diameter of 140μm , which gives a higher-order mode cutoff wavelength of about 0.59μm . The dielectric separation of the electrode holes is about 20μm. A length of fibre of about 1 meter is taken.

Metal or metal alloy is introduced into the holes either directly during the fibre drawing process or by filling after the draw with a liquid metal. In the latter case liquid gallium, for example, is injected from one end into one of the holes while the other hole is blocked. The length of the metal is about 75cm. Similarly, the other hole is filled with gallium, up to a length of 75cm, from the other end. The overlap length of the two electrodes 7,9 is thus 50cm. Copper or gold wires 11, 13 of 20μm diameter are inserted into the two holes from either end to make contact with the electrodes 7,9 (Figure 2). In this way a d.c. voltage of 6kV can be applied to the electrodes 7,9 without dielectric breakdown, since this technique avoids the problem of having two closely-spaced electrode wires emerging from one end of the fibre, as would be the case for filling both holes from one end. The above fibre is then poled by applying an intense electric field between the electrodes for a prolonged period of time. A linear electro-optic coefficient of 2xlO"¹⁵m/V can be obtained by poling the above fibre 1 (Figures 1 and 2) at room temperature using a poling field of 300Vμm intensity for a period of 20 minutes. A similar result can be obtained using a less intense poling field (lOOV/μm ) for a longer period of time (1 hour) at an elevated temperature (350°C). A higher linear electro-optic coefficient can be obtained by applying a more intense poling field (the dielectric breakdown field is about 600 V/μm) over the core 3 for a longer time. The induced effect is permanent. No significant decay has been observed under laboratory conditions for at least 100 days.

The results of further experiments are given in Figures 3 and 4. In these experiments fused silica fibres each with a germanium/phosphorous doped core 3 (7 mol % Geθ2 and 0.5 mol % P2 O₅) were used. The electrodes 7, 9 had a separation of 15μm and each fibre was of length 65cm. These fibres were birefringent and were produced by allowing the core 3 to deform from circular to elliptical cross-section by drawing at a higher temperature. These exhibited a linear birefringence of 1.06 x lO"⁴ and in each case the principal birefringent axis of the fibre was aligned with the electrodes 7, 9. This intrinsic birefringence is useful for preserving optical polarisation when long interaction lengths are employed. Measurements were made using a linearly polarised He-Ne laser beam (λ = 633nm) which was launched into the fibre at an angle of 45° with respect to the x-polarisation axis. This was phase modulated using an a.c. electric field of 10kHz of up to 14V/μm magnitude applied between the electrodes 7,9.

Figure 3 shows the variation of a.c. phase retardation in rads/m with applied a.c. field. "As drawn" fibres and fibres poled at 400 V/μ for 10 mins are compared. For the "as drawn" fibres, the graph shows that the effect is quadratic. The plots lie on a square-law curve calculated for a quadratic electro-optic coefficient of 2.2 x 10" ¹⁶ m/V². For the "as poled" fibres the graph shows that the effect is linear. The plots lie on a line corresponding to a linear electro-optic coefficient of 2.1 x 10^-15 m/V. A Pockels modulator constructed form a 6 metre length of this "as poled" fibre would require a mere 60V voltage to provide π/2 phase retardation. From Figure 3 it is clear that the a.c. optical phase retardation is at all practical values of applied a.c. field of significantly greater magnitude for the "as poled" fibres.

The relationship between the linear electro-optic coefficient and the poling field intensity is shown in Figure 4. Prior to measurement, each fibre sample was poled for a fixed period of 10 mins. The increase in coefficient value with poling field intensity is seen to be monotonic. No saturation was observed over the range of measurement.

Mention has been made above of fibres in which the cores are doped with germanium or with germanium and phosphorous. Linear effects are also found for poled fibres with cores including only phosporous dopant.

Other electrode configurations that can be employed are shown in Figures 5 to 8. in each of these examples a D-shaped fibre 15 has been drawn from a shaped preform, one which has been ground down to expose a plane face 17 extending parallel with the core 3. Surface irregularities are smoothed during drawing. Employing one internal electrode and one external electrode provides a greater versatility with little detriment to the maximum magnitude of field which can be applied. Thus the external electrode can be either continuous 19 (Figure 5) or periodic 21, for example comb-like (Figure 6). In addition, the poling field can be applied directly or from the discharge of electrons from a needle-shaped external electrode 23 (Figure 8) which can be moved along the length of the fibre 15 and repositioned for repeated application of the poling field (here localised).

Various patterns of second-order non-linearity can be written electrically into the fibre core. For example, where the external electrode 21 is spatially periodic (Figure 6), the poling electric-field will alternate between zero and its full value along the length of the fibre 15 and this results in a second-order non-linearity having the same spatial periodicity as the external grating electrode structure 21. A positive and negative spatially alternating d.c. electric-fields can be applied using the interleaved electrode structure 21, 25 shown in Figure 7 and this results in an induced second-order non- linearity which similarly alternates positive and negative. The same effect can be achieved by a voltage to the point electrode 23 shown in Figure 8 for the required poling time, then moving the electrode 23 along the fibre 15 to a new position and applying a voltage of opposite sign to the point electrode 23. This method has the advantage of allowing a higher applied field without risk of breakdown between external electrodes 21, 25, as can occur using the configuration of Figure 7. Various known fibre configurations can be treated in the abo^-- way to produce a linear electro-optic effect, for example highly-birefringent fibres, spun fibres, rare-earth-doped fibres and twin-core fibres. In each case the induced electro-optic effect can be uniform or varying along the fibre length. TECHNICAL APPLICATION

Poled fibres produced by the method described herein will provide the basis for optical switches and light modulators, such as required for Q-s witching and mode-locking of fibre lasers. They will also find application in tuneable spectral filters, optical signal processing, fibre sensing systems and in other areas where active control of light is necessary. In addition such fibres, which possess linear electro-optic properties, could be used for electric-field sensing devices where a linear response to applied field is more convenient than the conventional quadratic response.

Claims

CLAIMS:-What I/we claims is:-

1. A method for the manufacture of an electro-optic device, this method including the following procedural steps, namely: providing an optical fibre comprised of a core and cladding; and, applying in a direction transverse to the optical fibre core an intense external electric d.c. poling field for a prolonged period of time sufficient to impart to the optical fibre a permanent linear electro- optic property.

2. The method as claimed in Claim 1 wherein the optical fibre is maintained at an elevated temperature whilst the d.c poling field is applied.

3. The method as claimed in either Claims 1 or 2 wherein the optical fibre provided includes at least one internal electrode.

4. The method as claimed in Claim 3 wherein the optical fibre provided has one internal electrode and one external electrode, which external electrode is located on a plane face of the optical fibre, this plane face being parallel to the fibre core.

5. The method as claimed in Claim 4 wherein the external electrode is configured to a comb-like pattern.

6. The method is claimed in Claim 5 wherein a further external electrode is provided on the plane face, this electrode also being configured to a comb-like pattern, the external electrodes being interleaved.

7. The method as claimed in Claim 3 wherein the optical fibre provided has one internal electrode and one external electrode, which external electrode is movable in a direction along the length of the optical fibre and, the d.c. poling field is applied repeatedly for different positions of the external electrode.

8. The method as claimed in Claim 3 wherein the optical fibre provided includes a pair of internal electrodes.

9. The method as claimed in Claim 8 wherein the internal electrodes external inwardly from respective ends of the optical fibre and overlap for a part only of the length of the optical fibre.

10. The method as claimed in any one of Claim 3 to 9 wherein each said internal electrode is of metal or metal alloy, and the d.c. poling field is applied at a temperature such that the metal or metal alloy is liquid.

1 1. The method as claimed in Claim 10 wherein each said internal electrode is of germanium or germanium alloy and the d.c. poling field is applied at room temperature.

12. The method as claimed in any one of the preceding claims wherein the optical fibre provided is of fused silica.

13. The method as claimed in any one of the preceding claims wherein the optical fibre core includes germanium or phosphorous dopant or a combination of these dopants.

14. The method as claimed in any one of the preceding claims wherein the optical fibre provided is birefringent.

15. The method as claimed in Claim 14 wherein the d.c. poling field is applied in a direction lying in a principal plane of the birefringent fibre.

16. A method for the manufacture of an electro-optic device when performed substantially as described hereinbefore with reference to the drawings.