Example: biology

08 modulators 2015 - UCY

1 Electrical-to-optical conversion: modulators HMY 645 Lecture 08 Spring Semester 2015 Stavros IezekielDepartment of Electrical and Computer EngineeringUniversity of Cyprus2PL(mW)IL(mA)[])(cos1tiItmIIBmBL+= += BI())(cos100tpPtmPPmL+=+= 0P)()(tIstPLLL=Consider the static optical power-versus-current characteristic of a laser diode; if we bias at point IBand then superimpose modulation, then the optical power will track changes in this. We show it here for sinusoidal modulation:Direct modulationLILP3 Although the method of direct modulation is a useful one, it suffers a number of problems:1. As well as the intensity of the light, the wavelength is modulated. (This phenomenon is called chirp.) Along with fibre dispersion, this leads to a chirp-induced dispersion limit on transmission The maximum bandwidth we can modulate up to is only a few tens of GHz at the very The maximum quantum efficiency ( ) in theory is 100%, and this places an upper limit on the slope efficiency (and therefore the gain ).

MACH-ZEHNDER MODULATORS 13 Lucent 14 Light from a laser can be described by its electric field. To keep things simple we consider a purely monochromatic laser (i.e. a …

Tags:

  2015, Modulator, 08 modulators 2015

Information

Domain:

Source:

Link to this page:

Please notify us if you found a problem with this document:

Other abuse

Advertisement

Transcription of 08 modulators 2015 - UCY

1 1 Electrical-to-optical conversion: modulators HMY 645 Lecture 08 Spring Semester 2015 Stavros IezekielDepartment of Electrical and Computer EngineeringUniversity of Cyprus2PL(mW)IL(mA)[])(cos1tiItmIIBmBL+= += BI())(cos100tpPtmPPmL+=+= 0P)()(tIstPLLL=Consider the static optical power-versus-current characteristic of a laser diode; if we bias at point IBand then superimpose modulation, then the optical power will track changes in this. We show it here for sinusoidal modulation:Direct modulationLILP3 Although the method of direct modulation is a useful one, it suffers a number of problems:1. As well as the intensity of the light, the wavelength is modulated. (This phenomenon is called chirp.) Along with fibre dispersion, this leads to a chirp-induced dispersion limit on transmission The maximum bandwidth we can modulate up to is only a few tens of GHz at the very The maximum quantum efficiency ( ) in theory is 100%, and this places an upper limit on the slope efficiency (and therefore the gain ).

2 QhcsL=EXTERNAL MODULATION4CW lightModulated lightTi-diffused optical waveguideLithiumniobatesubstrateElectrod esIn addition to direct modulation, we can also modulate the light from a laser with an external component known as a modulator . Hence the terms external modulation and external Bias point and modulation depth chosen to give incrementally linear slopeOpticalpowerThis will depend on the CW laser output power as well as drive conditions)(tvVB+)(0tpP+Externalmodulato rCWLaserRF input + BiasExternal modulationModulatedlightoutputLaser emits constant optical power. This then passes through an optical modulator (external modulator ) this is a voltage driven device. As we adjust the voltage, the amount of optical power absorbed will vary.

3 In this way, we achieve modulation of the optical power coming out of the advantage of external modulation is that it can be used to implement optical phase modulation, which opens up the possibility of coherent optical communications and therefore increased receiver advantages of external modulation compared to direct modulation: Laser diodes suffer from chirp which then introduces dispersion penalty. Not an issue with external modulation. It is possible to produce formats such as single sideband (SSB) or double-sideband suppressed carrier (DSB-SC) Slope efficiency of laser diodes is limited by fundamental quantum efficiency (100% max), whereas for external modulation the slope efficiency scales with CW laser power.

4 Laser diodes limited to 30 GHz max (unless optical injection locking is used), whereas up to at least 100 GHz has been reported with modulators . Many Mach-Zehnder modulators are based on lithium niobate and are difficult to integrate with other components, but electro-absorption modulators lend themselves to monolithic integration with driver electronics. Recent work by Intel, for example, on silicon modulators paves the way for integration with CMOS Considerations8 Obviously the key requirement is that some optical property of the material must change in response to a changing electrical parameter. Electro-optic effect An applied electric field changes the refractive index This leads to phase changes Can also produce intensity modulation when combined with an interferometer Acousto-optic effect A sound wave (resulting from electric field applied to a piezoelectric) changes the refractive index Electro-absorption effect Applied electric field changes the absorption9To date, the dominant type of modulator is the lithium niobate Mach-Zehnder, which is based on the electro-optic effect:A small change in refractive index nresults from an electric field E.

5 112202+++=RErEnnPockels effectOnly certain crystalline solids show the Pockels effect, as it requires lack of inversion symmetryIt is linear with respect to electric field and hence voltageKerr effectObserved in all optical materials with varying magnitudes, but generally weaker than general, the Pockels effect is used since it is stronger (Kerr effect is primarily exploited for optical fibre solitons).The Pockels coefficients rijare elements of a 6 x 3 tensor10 Moodie, CIPC omparison of electro-optic materialsApart from presence of electro-optic effect, other important material properties include optical loss, maximum optical power handling capability and stability (thermal and optical). modulators can be made from inorganic materials, semiconductors or polymers11 Polymer modulator fabricated using SU-8 based rubber stamp as a potential route to low cost , for the moment lithium niobate (an inorganic material) dominates, not because it excels with respect to loss, stability, maximum optical power or electro-optic sensitivity, but because it offers the best compromise between all four key niobate is also relatively cheap since it is also widely used in surface acoustic wave filters.

6 It can be grown using the Czochralskiprocess in wafer sizes large enough to accommodate the relatively long and narrow structures required for from a laser can be described by its electric field. To keep things simple we consider a purely monochromatic laser ( a perfect laser), for which the emitted field at some fixed distance from the laser is given by:))()(()()(tttjoooetEtE +=Amplitude (complex quantity)Optical frequency ( 100 s of THz)Optical phaseIn analogy with electronic communications, we are able to modulate amplitude, frequency or modulation in optical communications is known as intensity modulation, and this is the most common approach. It can be achieved either through direct or external and phase modulation can only be achieved with an external modulator , and can only be detected with a coherent photoreceiver.

7 We will not consider these techniques any further optical intensity is directly proportional to the square of the electric field magnitude. The optical power emitted by the laser is, in turn, directly proportional to the intensity. So we can write:22)()(tEtEo= power opticalSo the optical power varies only with variations in the amplitude of the electric field, and this is achieved either through direct modulation or an external will now consider the operation of an external modulator based on the principle of an interferometer:ModulatorElectrical input (modulation)Unmodulated light from laser16 External modulators that are based on the interferometer principle are known as Mach-Zehnder modulators (MZM). To understand the basic principle, we need to remember something about superposition (and constructive and destructive interference).

8 Consider some examples:+= +=timeConstructive interference:+=time"Quadrature phase" 90 consider the optical waveguide structure of a MZM:Input lightY-junction. The incoming light is split equally into two paths at this point. So the light on each of these paths for an ideal device will be 3 dB less in optical power compared to the input two waveguide arms have equal length, so the delay and hence phase shift is equal for both Y-junction. Here light from the two arms is combined in phase. However, the optical power of the output will be lower than that of the input due to losses in the waveguides and at the Y-junctions. We refer to this as the insertion lossof the MZMO utput light18CW lightModulated lightTi-diffused optical waveguideLithiumniobatesubstrateElectrod esThe waveguides are formed from titanium which is diffused onto a layer of lithium niobate, which forms the substrate.

9 Lithium niobate is a material that has a strong electro-optic effect if we apply a voltage to it, then its refractive index changes. We can show that this is equivalent to introducing a phase the MZM shown above, a voltage applied to the electrodes will introduce a phase shift into the upper arm. For zero volts there is no phase shift and we have constructive interference, but if we increase the voltage to some value (called V ) then there is a radians relative phase shift leading to total extinction. Values in between will lead to varying levels of VVmioPPffTReduction due tooptical insertion lossof modulatorJust as we have a light-current characteristic for a laser diode, we have a voltage-light characteristic for a MZM: VVm=1<=ffiffoTPTP20 The transfer characteristic is given by: += VVTPP mffiocos12If we apply a bias voltage of nV /2(where nis odd) and a small-signal modulation component given by vm(t), then linearization of the above equation around the bias point will yield:() ++= VtvTVtvVTPP mffmBffio)(12)(cos12from which the slope efficiency (in W/V) is obtained as.

10 IffmoPVTdvdP 2=So by increasing the optical power from the CW laser, we can increase the efficiency of the of the bias point is an important consideration, because the sinusoidal shape of the MZM transfer characteristic means there are no linear parts to the curve, so if we want to have almost linear operation we must choose points on the curve that are good approximations to a straight line. Also, we can show that the best bias points will be those for which the slope of the characteristic is maximised (in order to prove efficiency).If we assume constant CW input power, then: += VVPTP miffocos12 = VVVPTdVdPmiffmosin2 Finding the maxima/minima for this derivative yields the following as suitable bias points.


Related search queries