GB2270173A

GB2270173A - Optical modulator

Info

Publication number: GB2270173A
Application number: GB9218336A
Authority: GB
Inventors: Nicholas John Parsons
Original assignee: GEC Marconi Ltd; Marconi Co Ltd
Current assignee: BAE Systems Electronics Ltd
Priority date: 1992-08-28
Filing date: 1992-08-28
Publication date: 1994-03-02
Anticipated expiration: 2012-08-28
Also published as: GB9218336D0; GB2270173B

Abstract

An optical modulator comprises a Mach-Zehnder interferometer in which an electric field is applied across each arm 4, 5 to modulate an optical signal. One of the arms 5 includes a length portion 14 offset relative to electrodes 8, 9 to a region of substantially zero electric field strength, so as to provide the device with a predetermined level of chirp. <IMAGE>

Description

Optical Modulator This invention relates to an optical amplitude or intensity modulator for use in an optical communication system.

It is well known that an optical signal can be modulated using the electro-optic effect. When broad band digital or analogue signals are to be transmitted over long distances of optical fibre and at a high data rate the modulation characteristics of the optical source must be closely controlled. In particular the phase or frequency deviation associated with the intensity modulation must be very small when transmitting at wavelengths where significant dispersion exists. It has been proposed to use a Mach-Zehnder interferometer type modulator in which light is input into a waveguide which splits into two arms and then recombines. The phases of the light propagating through the two arms are modulated by an electric field of equal magnitude applied in opposing directions.This reduces the frequency spread of the signal passing along the waveguide and reduces sensitivity penalties caused by optical dispersion. Such a balanced interferometric modulator has an dE parameter, or so-called chirp of zero. The a; parameter represents the ratio of phase modulation to amplitude modulation and has been expressed as a = (d/dt)l(Dildr)lS where + and 5 are the instantaneous phase and intensity respectively of the optical output and t denotes the time.

However, recent work on high data rate systems, eg. in excess of 5 Gbit\s, has shown that optimum performance in standard fibre can be obtained for a non-zero value of owt of between, say, -0.4 and -0.5. The value varies according to the length of the optical link, the wavelength of operation and the dispersion characteristics of the fibre amongst other factors. US-A-5,074,631 discloses a modulator of the Mach-Zehnder type having an adjustable rt parameter. However, the device requires separate driving circuitry for each of the two arms of the interferometer which, while ensuring that the device is versatile, increases the expense of installation.In practice for most installations such versatility is not necessary and it is an object of the invention to provide an optical modulator of the Mach-Zehnder type having a fixed ot parameter, high performance, which can be made easily available at low cost and which does not require expensive driving circuitry.

This invention provides an optical modulator comprising a Mach-Zehnder interferometer arrangement in which an optical waveguide splits into two arms and then recombines, electrodes being disposed about each arm to define electrode gaps of substantially equal length and width such that an electric field of substantially equal strength can be applied across each arm, one of the arms including a portion of predetermined length located relative to its respective electrode gap such that the electric field applied is substantially zero.

By selecting the length which is offset from the electric field, the C parameter can be fixed at a required value.

It is much preferred that the length portion is disposed symmetrically about the middle of the length of the respective arm. This ensures that the base band frequency response of each arm remains substantially the same - i.e.

the Cs parameter at relatively higher frequencies is the same as that at relatively lower frequencies.

In a preferred arrangement the two arms extend in parallel to each other on either side of a central common electrode, a further electrode being disposed on the other side of each arm and arranged such that a single driving signal can be applied across both electrode gaps.

In a preferred arrangement each arm, apart from the offset portion, is located centrally with respect to its respective electrode gap. Such an arrangement ensures that the modulator is easy to make using conventional masking techniques since any problems that can be caused by tolerances in aligning each respective arm with its electrode gap are minimised. This is to be compared to a case, for example, where the applied electric field were to be adjusted by offsetting one or both arms by a small amount only so as to gradually adjust the refractive index. It is much preferred that the length portion extends substantially parallel to one side of its respective electrode gap.

In order that the invention may be well understood, an embodiment thereof will now be described with reference to the accompanying diagrammatic drawings in which: Figure 1 shows a plan view from above of an optical modulator according to the invention; and Figure 2 is a cross sectional view along line A-A of Figure 1; and Figure 3 shows the impulse response within each arm of the modulator shown in Figures 1 and 2.

An optical modulator comprises a substrate 1 formed from electro-optic crystal material such as x-cut, y propagating Lithium Niobate (LiNbO3) and a single mode optical waveguide 2 formed from, eg. Titanium (Ti) which has been diffused, implanted or otherwise formed by proton exchange procedures in the substrate. The waveguide 2 comprises an input 3 which splits into two arms 4, 5 of substantially equal length and then recombines before passing to an output 6. Electrodes 7, 8, 9 and 10 arranged in a travelling wave configuration and formed from a suitably conductive material such as copper are provided on the top of the substrate 1. The outermost electrodes 7 and 9 are connected to ground while, in use, an R.F. signal containing the data to be transmitted is applied to the input end 11 of the middle electrode 8 from where it travels to the output 15.Electrode 10 is used to apply a D.C. bias to the interferometer. The electrodes 7,8,9 define electrode gaps 12, 13, of substantially equal length and width, one disposed above each of the waveguides 4, 5 so that electric fields E of equal magnitude, but of opposite direction, can be applied across each waveguide arm 4, 5. Each waveguide arm 4, 5, is located centrally with respect to its respective electrode gap substantially equidistantly from each of its respective electrodes, except for a portion 14 of arm 5.

Arm 5 includes a portion 14 which is offset to one side of its electrode gap 13, as best seen in Figure 2, to a region where the electric field E between electrodes 8 and 9, is substantially zero. The portion 14 is disposed symmetrically about the middle of the length of the electrode gap 13.

At high data rates and with the travelling wave electrode configuration shown, electrical and optical signals travel in the same direction along the device. The speeds of propogation are different, depending on the relative refractive indices, and this gives rise to a rectangular impulse response of duration equal to the transit-time difference between the signals. As shown in Figure 3, the impulse response within arm 4 is shown in solid line while the response within arm 5 is shown in dotted line. The length in time of each pulse is equal to L/(nm-nO) C, where L is the length of each electrode gap, nm and nO are the refractive indices of the electrodes and the waveguides respectively, and C is the speed of light. The value of the $ parameter is proportional to the length of portion 14 divided by L.

It can be appreciated that if one pulse were much shorter than the other, caused for example by one electrode gap being shorter than the other, and those pulses were Fourier transformed into the frequency domain the shorter pulse would tend to have a broader spread of frequency components. The fact that the electrode gaps are of the same length, and the central disposition of the offset portion 14 ensures that the base band frequency response of each arm is substantially the same.

Claims

1. An optical modulator comprising a Mach-Zehnder interferometer arrangement in which an optical waveguide splits into two arms and recombines, electrodes being disposed about each arm to define electrode gaps of substantially equal length and width such that an electric field of substantially equal strength can be applied across each arm, one of the arms including a portion of predetermined length located relative to its' respective electrode gap such that the electric field applied is substantially zero.

2. An optical modulator according to claim 1 in which the length portion is disposed symmetrically about the middle of the length of the respective electrode gap.

3. An optical modulator according to claim 1 or 2 in which the two arms extend in parallel to each other on either side of a central common electrode, a further electrode being disposed on the other side of each arm and arranged such that a single driving signal can be applied across both electrode gaps.

4. An optical modulator according to any preceding claim in which each arm, apart from the portion of predetermined length, is located centrally with respect to its respective electrode gap.

5. An optical modulator according to any preceding claim in which the length portion extends substantially parallel to one side of its respective electrode gap.

6. An optical modulator according to any preceding claim comprising a substrate of Lithium Niobate having a Titanium waveguide.

7. An optical modulator substantially as described with reference to the accompanying drawings.