Vous êtes sur la page 1sur 27

A Seminar Report on Silicon Photonics

CHAPTER 1 CONCEPT OF SILICON PHOTONICS

We all expect fast, free-flowing bandwidth whenever and wherever we connect with the world. However, there is a problem, as newer faster microprocessors roll out, the copper connections that feed those processors within computers and servers will prove inade uate to handle the crushing tides of data. Here is a better wa!" replace the copper with optical fiber and the electrons with photons. #hat is the promise of silicon photonics" affordable optical communications for ever!thing. $t will let manufacturers build optical components using the same semiconductor e uipment and methods the! use now for ordinar! integrated circuits, thereb! dramaticall! lowering the cost of photonics. $ts overarching goal is to develop high-volume, lowcost optical components using standard %&'S processing-the same manufacturing process used for microprocessors and semiconductor device. #he onl! wa! for photonics to move into the mass mar(et is to introduce integration, high-volume manufacturing, and low cost assembl!-that is, to )Siliconi*e) photonics. +! that, we mean integrating several different optical devices onto one silicon chip, rather than separatel! assembling each from exotic materials. #he researchers believe that with this development, silicon photonic chips containing do*ens or even hundreds of h!brid silicon lasers could someda! be built using standard highvolume, low-cost silicon manufacturing techni ues side b! side, communicating with each other to form a supercomputer far be!ond the scale of toda!,s fastest computer. Silicon photonics technolog! has the potential to use the power of optical networ(ing inside computers and to create new generation of miniaturi*ed and low-cost photonic components, among other applications. $n one potential use, man! boards containing these Silicon-photonic chips would coexist. #hus one can now see a path to integrating silicon h!brid lasers, photo detectors, modulators, and waveguides into a single highl! integrated photonic chip capable of transmitting - #bit.s of information down a single optical fiber all on a piece of silicon the si*e of !our fingernail. #he chips will also be well suited for use in general data communications and computing.

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

A Seminar Report on Silicon Photonics

CHAPTER 2 INTRODUCTION OF SILICON PHOTONICS

$ts overarching goal is to develop high-volume, low-cost optical components using standard %&'S processing the same manufacturing process used for microprocessors and semiconductor device. /espite silicon4s shortcomings, researchers have been stud!ing silicon photonics for more than 56 !ears, starting with Richard Soref4s pioneering wor( in the midioaos at the Air 7orce Research 3aborator!. Since then, there have been a host of silicon photonics brea(throughs at %ornell 2niversit!, the &assachusetts $nstitute of #echnolog!, the 2niversit! of %alifornia at 3os Angeles, the 2niversit! of %atania in Sicil!, the 2niversit! of Surre!, $+&, $ntel, and elsewhere. With optical interconnects in and around our des(top computers and servers, we4ll download movies in seconds rather than hours and conduct lightning-fast searches through gigab!tes of image, audio, or text data. &ultiple simultaneous streams of video arriving on our P%s will open up new applications in remote monitoring and surveillance, teleconferencing, and entertainment $n theor!, !ou could push fiber up to -86 trillion bits per second a rate that would deliver the text of all the boo(s in the 2.S. 3ibrar! of %ongress in about a second. 2nli(e electronic data, optical signals can travel tens of (ilometers without distortion or attenuation. 9ou can also pac( do*ens of channels of high-speed data onto a single fiber, separating the channels b! wavelength, a techni ue called wavelength-division multiplexing. #oda!, :6 separate signals, each running at -6 gigabits per second, can be s uee*ed onto a hair-thin fiber. #oda!4s devices are speciali*ed components made from indium phosphide, lithium niobate, and other exotic materials that can4t be integrated onto silicon chips. #hat ma(es their assembl! much more complex than the assembl! of ordinar! electronics, because the paths that the light travels must be painsta(ingl! aligned to micrometer precision. #he onl! wa! for photonics to move into the mass mar(et is to introduce integration, high-volume manufacturing, and low cost assembl!-that is, to )Siliconi*e) photonics. +! that, we mean integrating several different optical devices onto one silicon chip, rather than separatel! assembling each from exotic materials.
/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. 5

A Seminar Report on Silicon Photonics

2.1 Photon: A photon is a discrete bundle of light energ!. Photons are alwa!s in motion and, in a vacuum, have a constant speed of light to all observers, at the vacuum speed of light. #he photon is an elementar! particle, despite the fact that it has no mass. $t cannot deca! on its own, although the energ! of the photon can transfer ;or be created< upon interaction with other particles. Photons are electricall! neutral and are one of the rare particles. According to the photon theor! of light, photons &ove at the speed of light in free space Have *ero mass and rest energ!. %arr! energ! and momentum. %an be destro!ed .created when radiation is absorbed.emitted. %an have particle-li(e interactions ;i.e. collisions< with electrons and other particles, such as in the %ompton 0ffect.

2.2 Silicon: $t is semiconductor element which have s!mbol of silicon is Si and atomic number -:. Second onl! to ox!gen in abundance in 0arth4s crust= it never occurs free but is found in almost all roc(s and in sand, cla!, and soils, combined with ox!gen as silica. Pure silicon is a hard, dar( gra! solid with a metallic luster and the same cr!stal structure as diamond. $t is an extremel! important semiconductor doped with boron, phosphorus, or arsenic, it is used in various electronic circuits and switching devices, including computer chips, transistors, and diodes. Silicon presents a uni ue material for this research because the techni ues for processing it are well understood and it demonstrates certain desirable behaviors. 7or example, while silicon is opa ue in the visible spectrum, it is transparent at the infrared wavelengths used in optical transmission, hence it can guide light. &oreover, manufacturing silicon components in high volume to the specifications needed b! optical communication is comparativel! inexpensive.

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

>

A Seminar Report on Silicon Photonics

Silicon4s (e! drawbac( is that it cannot emit laser light, and so the lasers that drive optical communications have been made of more exotic materials, such as ?indium phosphide, and ?gallium arsenide,. However, silicon can be used to manipulate the light emitted b! inexpensive lasers so as to provide light that has characteristics similar to more-expensive devices. #his is @ust one wa! in which silicon can lower the cost of photonics.

2.3 Photonics: Photonics is the science of generating, controlling, and detecting photons, particularl! in the visible and near infra-red spectrum, but also extending to the ultraviolet ;6.5A6.>8 Bm wavelength<, long-wave infrared ;CA-5 Bm wavelength<, and far-infrared.#H* portion of the spectrum. #he science of photonics includes investigation of the emission, transmission, amplification, detection, and modulation of light. Photonic devices include optoelectronic devices such as lasers 1 Photo detector, as well as optical fiber and planar waveguides.

2.4 Silicon Photonics: Silicon photonics is an evolving technolog! in which data is transferred among computer chips b! optical ra!s. #he concept involves combining laser and silicon technolog! on the same chip. #he improved performance results from the greater available bandwidth and high propagation speed of infrared ;$R< beams compared with electric current. #he effective implementation of silicon photonics technolog! would dramaticall! increase the processing speed and power of computers.

2.4.1. Background: Photonics as a field reall! began in -DE6, with the invention of the 3AS0R ;3ight amplification b! stimulated emission of radiation<, and the laser diode followed in the -DF6s b! the development of optical fibers as a medium for transmitting information using light beams, and the 0rbium-doped fiber amplifier. #hese inventions formed the basis for the telecommunications revolution of the late 56th centur!, and provided the infrastructure for the internet.
/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. :

A Seminar Report on Silicon Photonics

Historicall!, the term photonics onl! came into common use among the scientific communit! in the -DC6s as fiber optic transmission of electronic data was adopted widel! b! telecommunications networ( operators. At that time, the term was adopted widel! within +ell 3aboratories. $ts use was confirmed when the $000 3asers and 0lectro-'ptics Societ! established an archival @ournal named Photonics #echnolog! 3etters at the end of the -DC6s. A huge further growth of photonics can be expected for the case that the current development of silicon photonics will be successful. 2.4.2. Need silicon photonics: $ntel cofounder Gordon &oore pro@ected that the number of transistors on a computer chip will double ever! 5 !ears. ;-DE8<. %omputer chips toda! have -.F billion transistors and growing, #he more transistors we can put on a chip, the more information it can process in a given time. #here alwa!s remains the limit on number of transistors. &ulticore processor solves the problem of transistors for now, but it never mean we can (eep @ust adding no. of cores to the processor forever and never limit the speed of our computer. %opper wires are used for the communication in computer. %opper has resistance. $ncreased resistance increases heat. $ncreased heat increases resistance that results in loss of information.

.
Fig: 2.1! "oo#$%s L&'

Right now, fiber optics is the best wa! to move large streams of data among computing devices. #o find fiber optics, !ou need loo( no further than the large-trun(, fiber optic lines that carr!
/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. 8

A Seminar Report on Silicon Photonics

telephone and $nternet traffic among service providers around the world. what about moving large amounts of data inside !our computerH %omputer manufacturers don,t use fiber optics to move data from component to component inside !our P% because it,s expensive to use and implement fiber optics inside computers, which is wh! the! rel! on electrical copper lin(s. However, as computer components and chips wor( faster, those slower copper lin(s will begin to hold bac( P%s. #he copper lin(s won,t be able to move data uic(l! enough to let the chips and other components wor( at top speed. #he chips ma! be waiting for data to arrive over the copper lin(s, leaving them idle. 7iber optics can carr! thousands of times more data than copper cable lin(s. Silicon photonics ma! be the answer to these problems. Silicon photonics brings laser technolog! to silicon, allowing for the use of fiber optic communications from a silicon chip. +ecause silicon is inexpensive, implementation of fiber optics in man! new areas including among servers, across networ(s, and inside computers ma! become possible. #o understand how optical data might one da! travel through silicon in !our computer, it helps to (now how it travels over optical fiber toda!. 7irst, a computer sends regular electrical data to an optical transmitter, where the signal is converted into pulses of light. #he transmitter contains a laser and an electrical driver, which uses the source data to modulate the laser beam, turning it on and off to generate -s and 6s. $mprinted with the data, the beam travels through the glass fiber, encountering switches at various @unctures that route the data to different destinations. $f the data must travel more than about -66 (ilometers, an optical amplifier boosts the signal. At the destination, a photo detector reads and converts the data encoded in the photons bac( into electrical data.

CHAPTER 3
/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. E

A Seminar Report on Silicon Photonics

(UILDIN) (LOC*S OF SILICON PHOTONICS

3.1. Int#o+,ction Silicon Photonics incl,+$s -ollo'ing -lo' o- P#oc$ss -lo' 1. Light source (low cost external laser) 2. Guide light (silicon on insulator) 3. odulation (!i "! capacitor de#ice)

4. $hoto detection (!i %ased photo detector) A laser generates light this light ma! be filtered and tuned to a specific wavelength. $t is then modulated, which is the process of placing data on the light. $f multiple optical channels are desired, the light then passes through a multiplexer that combines it with wavelengths from other lasers and places the resulting light onto a glass fiber. A bloc( diagram of this setup appears in 7ig ;>.-<. 'n the receiving end, the process operates in reverse. A demultiplexer separates the wavelengths on the fiber to create individual data streams. #hese are then routed to photo detectors that convert the light into electrical signals

Fig 3.1!: Si./l$ 0loc1 +i&g#&. o- Silicon Photonics.

3.2. Light so,#c$ LASER!:


/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. F

A Seminar Report on Silicon Photonics

3asers generate a beam of a single wavelength. 0lectrical or optical energ! is pumped into a gain medium, which is surrounded b! mirrors to form a Icavit!.J $nitial photons are either electricall! generated within the cavit! or in@ected into the cavit! b! an optical pump. the! trigger the release of other photons with the same optical properties ;wavelength, phase and polari*ation<. As the photons move bac( and forth between the mirrors, the! gather additional photons. And forms strong beam of photon light which is, sufficientl! strong to form a coherent laser beam. #his stimulated emission is created b! changing the state of electrons. As their state changes, the! release a photon, which is the particle that composes light which is a 3AS0R.

3.2.1. &he 'a(an )**ect: A phenomenon observed in the scattering of light is, as it passes through a transparent medium the light undergoes a change in fre uenc!, and random alteration in phase due to a change in rotational or vibrational energ! of the scattering. +! using this effect in silicon, enabling silicon to be used for the first time to amplif! signals and create continuous beams of laser light. #he Raman 0ffect is widel! used toda! to ma(e modulators and detectors. #hese devices are built b! directing a laser beam (nown as the pump beam into a fiber. As the light enters, the photons collide with vibrating atoms in the material and, through the Raman 0ffect= energ! is transferred to photons of longer wavelengths.

Fig,#$ 3.2! LASER th#o,gh silicon &n+ o/tic&l -i0$#

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

A Seminar Report on Silicon Photonics

&! #he Raman 0ffect allows energ! from a pump beam to amplif! data at longer wavelengths in glass fiber. ;b< #his can now done in silicon as well with small distance. $f a data beam is applied at the appropriate wavelength, it will pic( up additional photons. After traveling several (ilometers in the fiber, the beam ac uires enough energ! to cause a significant amplification of the data signal ;7igure >.> A<. +! reflecting light bac( and forth through the fiber, the repeated action of the Raman 0ffect can produce a pure laser beam. However, fiber-based devices using the Raman 0ffect are limited because the! re uire (ilometers of fiber to provide sufficient amplification. #he Raman 0ffect is more than -6,666 times stronger in silicon than in glass optical fiber, ma(ing silicon an advantageous material. $nstead of (ilometers of fiber, onl! centimeters of silicon are re uired ;7igure >.>b<. +! using the Raman 0ffect and an optical pump beam, silicon can now be used to ma(e useful amplifiers and lasers.

3.2.1.1 &he +hallenge: #he process of building a Raman amplifier or laser in silicon begins with the creation of a waveguide - a conduit for light - in silicon. #his can be done using standard %&'S techni ues to etch a ridge or channel into a silicon wafer ;7igure - b<. 3ight directed into this waveguide will be contained and channeled across the chip. $n an! waveguide, some light is lost through absorption b! the material, imperfections in the ph!sical structure, roughness of the surfaces and other optical effects. #he challenge that $ntel researchers surmounted is ma(ing a waveguide in which the amplification provided b! the Raman effect exceeds the loss in the silicon waveguide.

3.2.1.2. ,ow Lasers -ork: 3asers generate a beam of a single wavelength b! amplif!ing light. As shown in 7igure 5, electrical or optical energ! is pumped into a gain medium which is surrounded b! mirrors to form a Icavit!J. $nitial photons are either electricall! generated within the cavit! or in@ected into the cavit! b! an optical pump. As the photons stream through the gain medium, the! trigger the
/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. D

A Seminar Report on Silicon Photonics

release of duplicate photons with the same optical properties ;wavelength, phase and polari*ation<. As the photons move bac( and forth between the mirrors, the! gather additional photons. #his gain has the effect of amplif!ing the light. 2ltimatel!, the light is sufficientl! strong to form a IcoherentJ laser beam in which all the photons stream in parallel at the same wavelength. #his laser beam is shown exiting the cavit! b! the red beam at the right of the figure below. #he world,s first laser was built b! #ed &aiman in -DE6. #his device used a white flash lamp to opticall! pump a rub! cr!stal and generate red laser light $n mid-566:, $ntel researchers discovered that increasing the pump power be!ond a certain point no longer increased the amplification and eventuall! even decreased it. #he reason turned out to be a ph!sical process called two-photon absorption ;see next section<, which absorbs a fraction of the pump beam and creates free electrons. #hese electrons build up over time and collect in the waveguide. #he problem is that the free electrons absorb some of the pump and signal beams, reducing the net amplification. #he higher the power densit! in the waveguide, the higher the loss incurred. $ntel4s brea(through is a solution that minimi*es the extra electrons caused b! two-photon absorption - so that an amplified, continuous laser beam can be generated. $n fact, $ntel recentl! demonstrated the first silicon device with a continuous net amplification with a gain that more than doubled the input signal power.

Fig,#$ 3.3!: (2 ins$#ting & +io+$3li1$ PIN +$4ic$ in th$ '&4$g,i+$5 Int$l #$.o4$+ th$ $l$ct#ons g$n$#&t$+ 02 t'o3/hoton &0so#/tion &n+ /#o+,c$ contin,o,s &./li-ic&tion

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

-6

A Seminar Report on Silicon Photonics

3.2.2. &wo.photon a%sorption: 2suall!, silicon is transparent to infrared light, meaning atoms do not absorb photons as the! pass through the silicon because the infrared light does not have enough energ! to excite an electron. 'ccasionall!, however, two photons arrive at the atom at the same time in such a wa! that the combined energ! is enough to free an electron from an atom.

Fig 3.4!: PIN /3t2/$3int#insic3n3t2/$! +io+$ /l&c$+ on $ith$# si+$ o- th$ light 0$&.

When the pump pulse propagates through the waveguide, free carriers are generated due to the #PA effect. #he free-carriers effect not onl! causes excess absorptions, as mentioned before, it also induces a change in the refractive index of the Silicon ;Kn<. #he phase of the output probe light is therefore modified ;KL< due to this effect as"

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

--

A Seminar Report on Silicon Photonics

Where, 3M 3ength of waveguide NM Wavelength of the light, OM #PA coefficient PpM $ntensit! pulse of the probe, A MArea of waveguide.

Fig. 3.6!: En$#g2 (&n+ Di&g#&.

2suall!, this is a ver! rare occurrence. However, the higher the pump power, the &ore li(el! it is to happen. 0ventuall!, these free electrons recombine with the cr!stal 3attice and pose no further problem. However, at high power densities, the rate at which the free electrons are created exceeds the rate of recombination and the! build up in the waveguide. 2nfortunatel!, these free electrons begin absorbing the light passing through the silicon waveguide and diminish the power of these signals. #he end result is a loss significant enough to cancel out the benefit of Raman amplification.

3.2.3. Breakthrough Laser: #he solution is to change the design of the waveguide so that it contains a semiconductor structure, technicall! called a P$P ;P-t!pe -$ntrinsic - P-t!pe< device. When a voltage is applied to this device, it acts li(e a vacuum and removes the electrons from the path of the light. Prior to this brea(through, the two-photon absorption problem would draw awa! so man! photons as to

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

-5

A Seminar Report on Silicon Photonics

not allow net amplification. Hence, maintaining a continuous laser beam would be impossible. $ntel4s brea(through is the use of the P$P to ma(e the amplification continuous. 7igure ;>.E< is a schematic of the P$P device. #he P$P is represented b! the p- and ndoped regions as well as the intrinsic silicon in between. #his silicon device can direct the flow of current in much the same wa! as diodes and other semiconductor devices do toda! in common electronics. Hence, the manufacture of this device relies on established manufacturing technologies and it reinforces the basic goal of silicon photonics" inexpensive, high-performance optical components.

Fig,#$ 3.7!: Th$ 0#$&1th#o,gh silicon l&s$# ,s$+ & PIN +$4ic$ &n+ th$ R&.&n E--$ct to &./li-2 light &s it 0o,nc$+ 0$t'$$n t'o .i##o#s co&t$+ on th$ '&4$g,i+$ $n+s5 /#o+,cing & contin,o,s l&s$# 0$&. &t & n$' '&4$l$ngth.

#o create the brea(through laser, $ntel coated the ends of the P$P waveguide with mirrors to form a laser cavit! ;7igure above<. After appl!ing a voltage and a pump beam to the silicon, researchers observed a stead! beam of laser light of a different wavelength exiting the cavit! A the first continuous silicon laser

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

->

A Seminar Report on Silicon Photonics

Fig,#$ 3.8! An $9&./l$ o- c#$&ting .,lti/l$ silicon l&s$# so,#c$s -#o. on$ /,./ 0$&.

Fig 3.:!: T'o3Photon A0so#/tion in Silicon

3.2.4. 'a(an.Based !ilicon $hotonics Raman scattering was proposed and demonstrated in 5665 as a mean to b!pass these limitations, and to create optical amplifiers and lasers in silicon. #he approach was motivated b! the fact that the stimulated Raman gain coefficient in silicon is -6>--6: times larger than that in fiber. #he modal area in a silicon waveguide is roughl! -66 times smaller than in fiber, resulting in a proportional increase in optical intensit!. #he combination ma(es it possible to
/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. -:

A Seminar Report on Silicon Photonics

reali*e chip-scale Raman devices that normall! re uire (ilometers of fiber to operate. #he initial demonstration of spontaneous Raman emission from silicon waveguides in 5665 was followed b! the demonstration of stimulated Raman scattering and parametric Raman wavelength conversion , both in 566>. 'ther merits of the Raman 0ffect include the fact that it occurs in pure silicon and hence does not re uire rare earth dopants ;such as 0rbium<, and that the spectrum is widel! tunable through the pump laser wavelength.

3.3. Silicon "o+,l&to#:

3.3.1. /ntroduction: &odulation is one of the hottest topics where device scaling and power consumption targets are ta(ing over the race for high modulation speed that occurred in the last few !ears. #he bandwidth has increased from a few tens of &H* to multi GH* in little more than half a decade. &oreover, a %&'S compatible fabrication process optimised to maximi*e device !ield is also highl! desirable. Since the first ma@orit! carrier device based on accumulation demonstrated b! 3iu et al. and the first depletion t!pe device proposed b! Gardes et al, later demonstrated b! 3iao et al with an optical bandwidth of 56 GH* and >6 GH*, a plethora of devices based on accumulation and depletion have been demonstrated. #hese devices are mostl! based on &ach Qehnder interferometers and ring resonators of different si*es. $n the case of depletion modulators, the @unction is positioned in different wa!s with a particular trend towards vertical @unction implanted in the middle of the waveguide to maximi*e the optical mode overlap and the depletion region, nevertheless the fabrication of such a @unction is sub@ect to alignment tolerance which ma! impair the device characteristics from wafer to wafer. 0fficienc! is also a matter of importance as from this, will result the device operating voltage.footprint and ultimatel! the power consumption. #he optical modulator, which contains two p-n @unctions, each in a silicon-on-insulator waveguide that is an arm of a &ach-Qehnder interferometer, is based on the free-carrier plasmadispersion effect, in which the refractive index changes as a function of electric-field-induced
/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. -8

A Seminar Report on Silicon Photonics

carrier depletion. 'ptimum electronic performance is crucial= the device is designed so that the electrical and optical signals propagate together down the waveguide.

3.3.2. ,ow !ilicon

odulator -orks:

#o understand how the modulator functions, we need to touch briefl! on the nature of light. 3ight is a form of radiation that occurs at specific fre uencies, some of which are visible, and some, li(e ultraviolet and infrared, that are invisible. When light is emitted it travels in a pattern that loo(s ver! much li(e a sine wave. ;See the top row of 7igure -.< #he total distance reached b! the pea(s and troughs of this sine wave is (nown as amplitude. When the sine wave is nearl! flat, the light is at its dimmest and has low amplitude. When the pea(s and troughs are ver! high and deep, the light shines brightl! and has greater amplitude.

Fig: 3.;! &./li-ic&tion /h$no.$non

When two wavelengths are combined, the resulting sine wave is the sum of the two constituent sine waves. 7or example, if two sine waves are perfectl! in s!nc and added together ;left column of 7igure >.D<, the resulting sine wave has twice the amplitude of the individual waves. $n contrast, when two waves are completel! out of s!nc ;right column of 7igure >.-6< the resulting wave has no amplitude

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

-E

A Seminar Report on Silicon Photonics

3.3.3.

ach.0ehnder inter*ero(eter:

Fig 3.1<!: Enco+ing Photons 'ith D&t&

;a< An optical modulator encodes -s and 6s b! first splitting a laser beam in two and then appl!ing an electric field to the beams, so that one beam is dela!ed b! half a wavelength relative to the other. When the beams recombine, both beams will be out of phase, and the! will cancel out. ;b< When no voltage is applied, on the other hand, the beams remain in phase when recombined. 0ncoding the beam with -s and 6s then means ma(ing the beams interfere ;6< or (eeping them in phase ;-<.

Fig 3.11!: conc$/t o- /h&s$ shi-t$# -o# silicon /hotonics


/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. -F

A Seminar Report on Silicon Photonics

9ou start b! splitting the laser beam in two and then appl!ing an electric field to one beam. $f the speed changes enough to dela! the beam b! half of one wavelength, that beam will be out of phase with its mate. When the beams recombine, the! will interfere with each other and cancel out. $f, on the other hand, no voltage is applied, the beams remain in phase, and the! will add constructivel! when recombined. 0ncoding the beam with -s and 's, then, means ma(ing the beams interfere ;6< or (eeping them in phase ;-<. #hese resulting changes in amplitude ;the strength of the light< are the basis on which the photo-detector recogni*es 's and -s. +ecause the amplitude is being modulated ;to encode the data<, this techni ue is referred to as amplitude modulation ;A&<. We expect to achieve even greater bandwidth b! multiplexing these data streams. #his approach could bring silicon photonics into an age where bandwidths of :6 Gbps bandwidths or more are common.

Fig 3.12! O/tic&l Ph&s$ Shi-t$#

#he $ntel modulator is based on a &ach-Qehnder interferometer with a reverse-biased pn @unction in each of the arms. When a reverse voltage is applied to the @unction, free carriers A electrons and holes resulting from the n- and p-dopants are pulled out of the @unction, changing its refractive index via the free-carrier effect. #he intensit! of the light transmitted through the
/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. -C

A Seminar Report on Silicon Photonics

&ach-Qehnder interferometer is modulated b! modulating the phase difference between the interferometer,s two arms. #his modulation can be ver! fast, because free carriers can be swept out of the @unction. #he modulator speed is thus limited b!, the parasitic effects such as R% time constant limit. #he high-speed silicon modulator could find use in various future applications. 7or example, a highl! integrated silicon photonic circuit ma! provide a cost effective solution for the future optical interconnect within computers and other devices. With the demonstration of the :6 Gbps silicon modulator and the electricall! pumped h!brid silicon laser, it will become possible to integrate multiple devices on a single chip that can transmit terabits of aggregate data per second trul! enabling tera-scale computing.

3.4. Silicon (&s$+ Photo+$t$cto#:

Fig 3.13! Silicon (&s$+ Photo+$t$cto#

Silicon can also be used in photo detection-the process b! which incoming wavelengths are converted into electrical signals representing bits. &odulation turns the light on and off to encode the data. When the wavelength is off, a *ero-bit is encoded= when the wavelength is on, a one-bit result. #he photo detector has the responsibilit! of converting those incoming bits bac( into their electrical counterparts. &a(ing a photo detector in silicon, however, has a significant
/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere. -D

A Seminar Report on Silicon Photonics

challenge" At the infrared fre uencies used b! toda!,s fiber-optic lasers, silicon is transparent. $t cannot detect incoming light because the photons that ma(e up the wavelengths pass right through it. $nterestingl!, if the laser wavelengths were in the visible spectrum where silicon is not transparent, silicon would be ideall! suited for photo detection. $ntel has developed a means of adding the element germanium to silicon to improve its light sensitivit! in the infrared spectrum. #his approach leverages one of germanium,s important properties" $t can extend the spectrum of wavelengths at which silicon absorbs light. #his achievement enables $ntel to build silicon detectors for optical communication.

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

56

A Seminar Report on Silicon Photonics

CHAPTER 4 A T=PICAL ASSE"(L= OF SILICON PHOTONICS

7$G. below shows an assembling structure of silicon photonics in which modulators, electronic chip, optical fiber, photo-detectors, a laser source mirrors etc.

Fig 4.1! Ass$.0l2 St#,ct,#$

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

5-

A Seminar Report on Silicon Photonics

CHAPTER 6 "ICRO3PHOTONICS > NANO3PHOTONICS

&icro-photonics is a branch of technolog! that deals with directing light on a microscopic scale. $t is used in networ(ing. &icro photonics emplo!s at least two different materials with a large differential index of refraction to s uee*e the light down to a small si*e. Generall! spea(ing virtuall! all of micro photonics relies on 7resnel reflection to guide the light. $f the photons reside mainl! in the higher index material, the confinement is due to total internal reflection. $f the confinement is due man! distributed 7resnel reflections, the device is termed a photonic cr!stal. #here are man! different t!pes of geometries used in micro photonics including optical waveguides, optical micro cavities, and Arra!ed Waveguide Gratings. Pano-photonics is the stud! of the behavior of light on the nanometer scale. #he abilit! to fabricate devices in nanoscale that has been developed recentl! provided the catal!st for this area of stud!. #he stud! of Panophotonics involves two broad themes -. 5. Stud! the novel properties of light at the nanometer scale. 0nabling highl! power efficient devices for engineering applications. #he stud! has the potential to revolutioni*e the telecommunications industr! b! providing low power, high speed, and interference-free devices such as electro optic and all-optical switches on a chip.

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

55

A Seminar Report on Silicon Photonics

CHAPTER 7 AD?ANTA)ES AND DISAD?ANTA)ES

A+4&nt&g$s: -. $t is inexpensive to use. 5. $mplementation is eas!. >. Speed is so high ;- #bit.sec<. :. $mplementation of the transistors on computer chips will increase 8. $t can send 1 receive -66 billion bits per second. E. %omputer components will become ver! small in si*e. F. #he speed of light can also be controlled using photonic cr!stal waveguides. ;up to -.>66th of usual speed of light< C. 'ptics is destined to be utili*ed in data centers since optical communications can meet the large bandwidth demands of high-performance computing s!stems b! bringing the immense advantages of high modulation rates and parallelism of wavelength division multiplexing. D. Silicon photonics offers high densit! integration of individual optical components on a single chip. -6. Strong light confinement enables dramatic scaling of the device area and allows unprecedented control over optical signals. --. Silicon Panophotonics devices have immense capacit! for low-loss, high-bandwidth data processing. 7abrication of silicon photonics s!stem in the complementar! metaloxide-semiconductor ;%&'S<-compatible silicon-on-insulator platform also results in further integration of optical and electrical circuitr!. Dis&+4&nt&g$s: -. %ircuit %omplexit! is more.

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

5>

A Seminar Report on Silicon Photonics

CHAPTER 8 APPLICATIONS AND PRACTICAL I"PLE"ENTATIUON

A//lic&tions -. %omputers, Servers, Storage s!stems 5. %ontinue high speed data processing from source to %P2 - potential processing in the #b.s range >. Smaller computers, less heat, elimination of motherboards ;b! toda!,s standards< :. High performance for data.computation intense applications. P#&ctic&l I./l$.$nt&tion
1.

7irst 0lectricall! Pumped H!brid Silicon 3aser (Sept -Cth 566E b! $P#03< 'ptical microelectromechanical s!stem ;&0&S< devices

5.

Fig 8.1 Silicon l&s$#

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

5:

A Seminar Report on Silicon Photonics

Fig 8.2 o/tic&ll2 int$#conn$ct$+ -,t,#$ .,ltico#$ /#oc$sso#s 'ith th#$$ /l&n$s: CPU5 .$.o#2 &n+ o/tic&l /l&n$s int$#conn$ct$+ 02 4i&s &n+ o/tic&ll2 conn$ct$+ +,&l inlin$ .$.o#2 .o+,l$ OC3DI""! 'ith o/tic&l .$.o#2 cont#oll$# O"C!

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

58

A Seminar Report on Silicon Photonics

CONCLUSION

Silicon will not win individual devices, but with integrated modules that brings total functionalit! 1 intelligence at a lower cost. What is certain is that this technolog! will once again revolutioni*e the technological world, which is alread! growing at an exponential rate. #he high-speed silicon modulator could find use in various future applications. 7or example, a highl! integrated silicon photonic circuit ma! provide a cost effective solution for the future optical interconnect within computers and other devices. Silicon photonics offers high densit! integration of individual optical components on a single chip. 7uture monolithic $%s will get manufactured b! this technolog! which help to increase +W, Speed, and thus reduces %ost.

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

5E

A Seminar Report on Silicon Photonics

REFERENCES

R-S R5S R>S R:S R8S RES

&asin G. Sahni, and %apellini G., IGe photodetectors integrated in + "! photonic circuitsJ, Silicon Photonics, 566C, ECDC, p. CDC6C R. A. Soref, I!ilicon.%ased optoelectronics,,, Proc. $000, vol. C-, no. -5, pp. -EF6E, /ec.-DDD. +ahram Talali and S. /mitri, I'a(an.Based !ilicon $hotonicsJ, 7ellow, $000 T. Sel. #op Uuantum 0lectron 566E, ->, ;E<, pp. -EFCA-ECF White Paper on I+ontinuous !ilicon Laser %ased on 'a(an e**ectJ, b! $ntel in 566: /.T. Shin, V. H. 3ee, I ach.0ehnder silicon (odulator toward 1'2 $000, 'pt. 0xpress, vol. -:, no. 56, pp. D-DFA D565, 'ct. 566E White Paper on I3n#eils !ilicon $hotonics Breakthrough ,igh.!peed !ilicon odulationJ b! /r. &ario Paniccia /irector, Photonics #echnolog! 3ab $ntel %orporation optical inter*aceJ,

RFS RCS RDS R-6S R--S R-5S R->S

3.#sung-9ang and A. Vah-Wee, I3ndertaking research in the *ield o* silicon optical (odulatorsJ, Sel #op.-Uuantum 0lectronics, $000 T. of, vol. -E, pp. >--58. +. R. +ennett, I+ontrol o* unwanted light in !i wa#eguide J, /))) 4. 5uantu( )lectron., vol. U0-5>, no. -, pp. -5>A-5D, Tan. -DDE. 9urii A. Wlasov $+& #homas T. Watson Research %enter, 9or(town Heights, P9 -68DC, 2SA , I!ilicon photonics *or next generation co(puting s6ste(sJ p. 6F:>-5, 5668. T. W. Pan, I!ilicon Nanophotonics 7 application in sensingJ, 'pt. 0xpress -E, 8F5566C. H. #a(esue,IGeneration o* polari8ation entangled photon pairs using silicon wire wa#eguideJ, 'pt. 0xpress -E, 8F5- ;566C<. Sodihi. T. +, I!ilicon photonics a chip.scale one.wa6 #al#e *or lightJ, Pature Photonics, 566D, >, ;5<, pp. FFAFD Soref. R,I&he past9 present9 and *uture o* silicon photonics J, $000 T. Sel. #op. Uuantum 0lectron, 566E, -5, ;E<, pp. -EFCA-ECF

/epartment of 0lectronics 1 #elecommunication 0ngineering /r. +abasaheb Ambed(ar #echnological 2niversit!, 3onere.

5F