semiconductor lasers for high bit rate optical...

Semiconductor Lasers for High Bit Rate

Optical Communication Networks

R. Paoletti, R&D managerR. Paoletti, R&D manager

TTCTTC-- Avago Technologies Italy, Via Schiaparelli 12, 10148 Torino, ITAvago Technologies Italy, Via Schiaparelli 12, 10148 Torino, ITALYALY

PaviaPavia, 22/06/2009, 22/06/2009

Roberto Paoletti, Pavia

22 June, 2009

Semiconductor Lasers for High Bit Rate

Optical Communication Networks

Abstract

• Internet and intranet data traffic constantly increased during last years, and the consequent band demand is pushing research and development of high speed optical modules operating at 17 Gb/s (Fiber Channel applications, FC) and 40 / 100Gb/s (Gigabit Ethernet, GbE). Main requirements of such a transmitters are high performances, reliability, power consumption and manufacturability. Talk will reports semiconductor laser sources suitable for such demanding applications, underlying the key technologies, and benchmarking the results proposed by the world’s most important research centers.

Roberto Paoletti (M’97) Roberto Paoletti was born in Vado Ligure (Savona), Italy, in 1966. He received the

laurea degree in Electronic Engineer from the University of Genova in 1991, and the Ph. D. degree in

Electronic at the Electronic Department of Politecnico of Turin, Italy, in 1995, studying high frequency

characterization, modeling and equivalent circuits of high speed semiconductor lasers and packages.

In 1995 he joined the former CSELT, Torino, Italy, then TTC Agilent technologies and now Avago

technologies, Italy. In the last years he has been responsible of research projects on new laser devices, and

mainly involved in design, characterization and reliability of high speed optoelectronic devices. He is author

and co-author of more than 50 papers, 5 patents and a book on high frequency modeling and characterization of semiconductor laser sources, as well as reviewer for technical journals.

Mr. Paoletti is a member of the IEEE Lasers and Electro-Optics Society.


22 June, 2009

Outline

• Avago: company overview

• Laser sources for ““pluggable transceiver pluggable transceiver worldworld””


22 June, 2009

1939

Hewlett Packard

Foundation (T&M)

• Test & Measurement

• Life science

• Semiconductor components

• Computers/imaging…

November 1, 1999

Agilent spin-off

~80000 employees

Computers, printers, imaging, …

~40000 employees

T&M, Life science,

semiconductor components

December 1, 2005

Avago spin off

~ 6500 employees

Semiconductor

components

T&M, Life science

Acquisition of TTC

former optoelectronic technology division of CSELT

R&D Corporate Center of STET (now Telecom)

Avago History


22 June, 2009

Product leadership in target markets


22 June, 2009

� 40-years heritage of innovation and technology leadership

� Over 2,000 patent and patent applications

� Over 1,000 design engineers

Leading in Technology Innovation

Fiber Optic Products Division


22 June, 2009

Technical Edge: Innovation History and ExpertiseStrategic Partner of Choice World’s 1st production transceiver

1st to market with VCSEL transceivers

1st OC-48 part (2x9 SC Duplex)

1st 10G XENPAK

Agilent is spun off from HP

1st 12x2.5G parallel optic transceiver

1st to market with 10G SFP+

1st to market with 8G SFPs

Avago is spun off from Agilent

1st to market with 12x6G parallel optic

1st to demonstrate full 17G SFP1st to demonstrate full 100G parallel optics

1999

2001

2002

2005

2007

2008

1998

1996

1978

• Module design

• 30 years of first’s and standards leadership

• Leading green technologies – lowest power / extended temperature SFP+ designs

• IC development

• Enable next generation optics, time to market

• Lower cost, best power roadmaps, features

• Only POD supplier with DMI, pre-emphasis, etc.

• Packaging and sub-assembly

• Proven mastery of bulk optic technology

• Capability for highly dense optical footprints

Flex Optics Bulk Optics


22 June, 2009

Outline

• Avago: company overview

• Laser sources for ““pluggable transceiver pluggable transceiver worldworld””


22 June, 2009

Outline

• Introduction

- Datacom – telecom networks history

- Pluggable solutions

• 10 Gb devices and technologies for pluggable transceivers

- Laser basic concepts

- Key design elements for high performances laser sources

- Advanced laser sources for pluggable transceivers

- TTC III-V technology overview


22 June, 2009

FIBER OPTIC History:

1960: First proposal of fiber optic for telecom.Basic design [Kao STC]

1970: Production of first fiber optic [Corning]

1977: First fiber optic network (1.5 miles)in Chicago [ATT Bell Labs]

1988: First transoceanic fiber-optic cable (3148 miles,40000 simultaeous telephone calls)

Dr. Kao

1960

LASER History:

1960: First Laser (Ruby) [Hughes Labs]

1962: First Semiconductor Laser (GaAs @T = - 200 oC) [GEC, IBM, MIT]

1970: First Semiconductor Laser at room temp. [Bell Labs, Ioffe Phys.Tech Inst. USSR]

1972: Invention of DFB Laser [Bell Labs]

1984: First Strained MQW in semiconductor laser

1990- Lasers for optical telecom

2000- Uncooled telecom lasers

GEC 1962

(Optical) Telecom History


22 June, 2009

DATACOM TELECOM

DFBFP

VCSEL EML

10 Gb/s LASER Sources

Long Haul

Metro

Storage

2-20 km 20-100 km

Enterprise

Today Optical Network

10km 100km100m10m 1km


22 June, 2009

~250 x 250 mm

LASER

XFP(78 x 18mm)

X2XENPAK(100 x 50mm)

Evolution (since 2000)

PLUGGABILITY: PAY AS YOU GOPLUGGABILITY: PAY AS YOU GO

• Strong limits in space available and power budget

• Wide temperature operation (0÷85oC)

⇒⇒⇒⇒ Requirement on lasers:

• High temperature operation (preferably uncooled)

• Low cost, high manufacturing yield, high reliability

• no compromise on High performance

(high bit rate, high optical power, high spectral purity, …)

• Monolithic integration can provide:

• Small and smart devices (High functionality)

Pluggability: a keyword…From transceiver cards to hot-pluggable transceiver modules


22 June, 2009

Optical coupling:

Lens+Optical Isolator +

pigtailed fiber

Chip

laser

From a Butterfly Laser Module….The 1990 Technology

Temperature

control

Back detectorElectrical / RF

connections


22 June, 2009

Avago Fiber Optics Portfolio

Enterprise SONETStorage Base StationParallel

10G leadership• 1st to market SFP+• 1st to market LRM• Superior VCSELs• Only extended temp range part• Lower power

High reliability, low cost 1G SFP

Proven supply assurance

Broad portfolio• OC-3 to 192 in multiple form factors and distances• Industrial tempavailable (includingXFP-LR/SR1)• Superior EMIperformance

#2 in market• 8G leadership• Industry’s most reliable 4G VCSEL• Industry’s best defect PPM rate• 1st 17G SFP demo• Only extended temp range 8G part

Strategic partner to key industry players

Innovative low cost design

Enabling time-to-mkt• 1st 100G POD demo• 1st 12x6G POD• World’s strategic parallel optics partner

Superior feature set –DMI, pre-emphasis

Unsurpassed delivery record

Proven quality

Market leader• 1st and only OBSAI/CPRI specific portfolio• High BER and industrial temp rates

Critical partner to leading players• Nextgen partner• Proven reliability


22 June, 2009

Outline

• Introduction



• 10 Gb devices and technologies for pluggabletransceivers






22 June, 2009

MQW Active layer

• Active material: MQW Multi Quantum Well structure �

dimensional control at nanometric scale

• Wavelength selection: grating structure � dimensional control

at sub-µm scale

• Lateral confinement: ridge structure � dimensional control at

µm range

QUANTUM WELL

BARRIER

X 300.000 Grating

Confinement

Structure

Lasers for pluggable modules: key device elements for outstanding performances


22 June, 2009

Fabry-Perot Laser

1280 1285 1290 1295 1300 1305

-70

-60

-50

-40

-30

-20

Wavelength (nm)

Amplitude (dBm)

Device:br00T20u; I = 30.0 mA; Peak: 1291.0 nm; 23-Jan-2003

Optical cavity = active material

Ibias

Multimodal emission

Short haul – not necessarily true…

L Nn

====λλλλ

2

N= integer

n=refractive index

λλλλ = wavelength

λλλλ

Cavity modes

0 50 1000

5

10

15

20Optical Power (T=20, 40, 60, 80, 85 °C)

Current (mA); - 24-Nov-2008

Power (m

W)


22 June, 2009

Grating

•Phase shifted grating allows to get

100% single mode yield, for ideal

AR/AR coatings on both facets

θ 90+θb

b+Λ = m (λ / n)

b = Λ sinθΛ (1+sinθ)=m(λ / n)

Λ =m(λ / 2n)Λ =m(λ / 2n)Bragg wavelength

DFB Laser

1280 1300 1320 1340 1360-70

-60

-50

-40

-30

-20

-10

0

10

Wavelength (nm)

Rel. Amplitude (dB)

λ Peak: 1316.9 nm; SMSR: 53 dB


22 June, 2009

Ridge structure

•No lateral blocking layers

•Very simple technological

process (one-step epi-growth)

•Suitable for Al-based lasers

and low cost devices

Optimised facet cleavage processOptimised facet cleavage process

Heat pathHeat path

Narrow reverse mesa (small cavity, fast chips)Narrow reverse mesa (small cavity, fast chips)

TiTi--PtPt--Au metalAu metal

SiOSiO22

Au plated padAu plated pad

To get performances: design (ridge )


22 June, 2009

To get performances: control (MQW)

QWs Period 150

152

154

156

158

160

162

164

FA048

FA041

FA056

FA038

FA044

FA052

DA035

DA038

DA034

DA032*

DA040

DA044

DA101U

DA115*

DA071*

DA126

DA129

DA133

DA107L

DA117

FA095*

DA158L

DA171

DA199-203

DA215-219

DA229

FA128-132

FA117-121

FA122-126

DA259-261

DA269-273

DA279-284

DA297-302

DA315-320

DA333-338

Process name

PERIOD [Å]

MANIFOLD

MAN

IFOLD

MAN

IFOLD

R & D

ON-LINE DATA POINTS ON-LINE - 2 MRbar/1.128 line USL

ON-LINE LCL ON-LINE - 1 MRbar/1.128 line TARGET

ON-LINE CL ON-LINE + 1 MRbar/1.128 line LSL

ON-LINE UCL ON-LINE +2 MRbar/1.128 line ON-LINE OUTLIERS

All production wafers are inspected before wafer fabprocessing MQW period control within 0.3 nm (3σσσσ)

MQW period control chart

Out of control:Out of control:

NO Performances; NO reliabilityNO Performances; NO reliability


22 June, 2009

0 10 20 30 40 50 60 70 80 90 1000

5

10

15

20

25

30

35

40

45

Current (mA); - 09-Aug-2002

Power (m

W) T=20-100 ºC

InGaAsP-based

MQW

InGaAsP-based

MQW

MQW active material optimization

0 10 20 30 40 50 60 70 80 90 1000

5

10

15

20

25

30

35

40

0 10 20 30 40 50 60 70 80 90 1000

Current (mA); - 20-Aug 2004

5

10

15

20

25

30

35

40

Power (m

W)

T=20-100 ºC

InGaAsAl-based

MQW

InGaAsAl-based

MQW

T0=95 KT0=48 K


22 June, 2009

Optical coupling:

Lens+Optical Isolator +

pigtailed fiber

Chip

laser

Direct Modulation of Laser ModuleButterfly Laser Module

Temperature

control

Back detectorElectrical / RF

connections

RF


Intensity modulation of laser sources

• Laser chip equivalent circuit

0 5 10 15 20-25

-20

-15

-10

-5

0

5

10

Frequency (GHz)

Rel. Amplitude (dB)

Ith+ 10 mA: (I

th= 6 mA); f

-3dB= 7.93 GHz

Ith+ 30 mA: (I

th= 6 mA); f

-3dB= 13.29 GHz

Ith+ 50 mA: (I

th= 6 mA); f

-3dB= 16.47 GHz

Ith+ 70 mA: (I

th= 6 mA); f

-3dB= 18.56 GHz

Ith+ 90 mA: (I

th= 6 mA); f

-3dB= 19.95 GHz

Increasing I-Ith

thr IIf −∝

thr IIf −∝Parasitics

Active material


22 June, 2009

Intensity modulation of laser sources:

chip parasitics

Geometrical analysis

Equivalent circuit

Simulation and experimental results


22 June, 2009

Outline

• Introduction



• 10 Gb devices and technologies for pluggabletransceivers






22 June, 2009

Agilent uncooled InGaAlAs ridge FP laser

for 10GBASE-LRM product (300 m MMF)

Device design:•Active layer: Al based materialDesigned to enhance T0

• 9 InGaAlAs 55Å wells; strain +0.8%

• 8 InGaAlAs100Å barriers; strain –0.4%

• 2xSCH1: InGaAlAs 350Å• 2xSCH2: InAlAs 500Å

• Device technology:High yield/low cost/Al compatible

• Reversed mesa ridge• Auto-aligned mesa

•Optical cavity:Very fast chip at high T operation

• Narrow cavity volume • Hr coating optimised versus

both Temperature and speed

200 µµµµm long x 250 µµµµm wide device

Avago/TTC Post

Deadline Paper at

OFC 2005


22 June, 2009

Device results - static

20 °C base chip temperature:• Threshold 7.6 mA•High power 23 mW

85 °C base chip temperature:• threshold 15.6 mA • power 16.8 mW

95 °C base chip temperature:• threshold as low as 18 mA• Still more than 15 mW

0 10 20 30 40 50 60 70 80 90 1000

5

10

15

20

25

fa021ak113: Optical Power (T=20, 40, 60, 80, 85, 90, 95 °C)

Current (mA); - 23-Dec-2004

Power (m

W)

T=20ºC; Ith= 7.6mA; PIth+30

= 8.9mW; Pmax

= 23.8mW

T=40ºC; Ith= 9.3mA; PIth+30

= 8.5mW, Pratio,Ith+30

=0.95; Pmax

= 22.4mW

T=60ºC; Ith=11.5mA; PIth+30


=0.88; Pmax

= 20.0mW



=0.80; Pmax

= 17.4mW



=0.78; Pmax

= 16.8mW



=0.78; Pmax

= 15.9mW



=0.76; Pmax

= 15.3mW

Key points:• High optical power enable

high coupling loss• Small threshold increasing up

to 95 C since high T0 • Small bias variation over T• Small efficiency degradation

over T

⇒ Constant eye quality with

constant modulation current!

Centre of eye power

Bias variation (0-85 °C)

Measured on chip on carrierMeasured on chip on carrier

Slope eff. ratio= 80%

110 °C


22 June, 2009

Technologies for LRM application: EDC (Electronic Technologies for LRM application: EDC (Electronic

Dispersion Compensation)Dispersion Compensation)

•EDC uses adaptive electrical

filtering techniques to compensate

for limitations incurred during fiber

propagation:

- modal dispersion

- chromatic dispersion

- polarization modal

dispersion

•EDC can be realized in an

integrated circuit.

•EDC can enhance the performance

of existing transceivers and

enabling new applications.

EDC technology

10 Gbps to 300m over legacy

multimode fiberTX CDR/

LD

FP/DFB

LinTIA

220m MMF RX CDR

Electronic Dispersion

Compensation

algorithm

D

t1

D

t2

D

tn

LPF CDR

Signal

Feedback

Adaptive controller

t1 tn

EQ

Why?


22 June, 2009

Modal dispersion in MM fibersModal dispersion in MM fibers

back


22 June, 2009

Eye diagram in the XFP module

70 C module 4.5 dB e.r. 43%

mm

0 C module 4.5 dB e.r. 43% mm

0 10 20 30 40 50 60 70 80 90 1000

5

10

15

20

25

30

35

40

45

Current (mA); - 09-Aug-2002

Power (m

W)

20 °C

100 °C

40 °C

60°C

80 °C

90 °C

1270128012901300131013201330134013501360

-70

-60

-50

-40

-30

-20

-10

Wavelength (nm)

Rel. Amplitude (dB)

l Peak: 1314.0 nm; SMSR: 46 dB

T=100 C

1260127012801290130013101320133013401350

-80

-70

-60

-50

-40

-30

-20

-10

Wavelength (nm)

Rel. Amplitude (dB)

l Peak: 1306.7 nm; SMSR: 49 dB

T=20C

Agilent uncooled InGaAsP BH 10

Gb DFB for 10GBASE-LR product:

2001 results


22 June, 2009

Spectral broadening of a direct modulated

DFB laser

timetime

“0”

“1”


22 June, 2009

Dispersion penalty

Back-

to

back

After

a fibre

link

TX output RX optical input Decision input

Pulse broadening causes intersymbol interference (ISI) and then eye closure :

less signal for the same received power

Noise depends on receiver, not on link : same noise

To get the same S/N more input power is needed : this is dispersion penalty


22 June, 2009

Dispersion penalty at 2.5 Gb, 80 km, for a

directly modulated DFB laser

-35 -34 -33 -32 -31 -30 -29 -28 -27 -26

1e-3

1e-4

1e-5

1e-6

1e-7

1e-8

1e-9

1e-10

1e-11

1e-12

Optical input power [dBm]

BER

file: b28t20bb.; Sens. 10

-10: -29.6

file: b28t2080km.; Sens. 10

-10: -26.5

0, and 80 km

3.1 dB DP

DP


22 June, 2009

Electro-absorption

Modulated DFB Laser

(EML)

Electro-absorption

Modulated DFB Laser

(EML)

• Device based on the monolithic integration between a DFB laser

and an Electro-Absorption Modulator (EAM).

• The DFB is biased at a constant current (CW).

• The EAM switch from transparency to opacity by applying

modulated voltage (HF signal).

DMDM--DFBDFB EMLEML

Key advantage of EML: The high frequency response is directly related to the fast dynamics of the modulator and not to the dynamic of the DFB laser: better eye quality, lower dispersion

penaltyCarrier relaxation

in DFB laser

Integrated EAM-DFB (EML)


22 June, 2009

10Gb/ s filtered eye

10Gb/ s eye after 50km SMF

10Gb/ s unfiltered eye

BER test at 0, 50km SMF

0.6dB0 km

50 km SMF

10 Gb/s eye after 100km SMF BER test at 0,100km SMF

0.6dB

100 km SMF

0 km

50°C operation of 1550 nm 10Gb/s EML


22 June, 2009

And what next?

• Economic crisis ⇒⇒⇒⇒ profitability of Optics low since 2001;

• BUT

- need for more speed and ability of transport more data remains unsatisfied

- 100Gb/s Ethernet standardized and planned

- 1Tb/s Ethernet is starting to be discussed

• So Low cost, low volumes (right now, especially in telecom) applications

- Many believe that photonic integration is the only solution of the problem, like “discrete transistor versus ICs”

- Particularly true for telecom (data formats…)

- R&D still in good shape? Surely hit by the downturn…

Sources: Optical Fiber Conference 2009 and related overviews;Sources: Optical Fiber Conference 2009 and related overviews;


22 June, 2009

40 and 100 Gb/s standard

• (IEEE P802.3ba)

- for 40Gb/s (40GBASE – SR4, 850nm, 4 x 10GbE; 40GBASE – LR4, 1300nm, 4 x 10GbE CWDM)

- for 100 Gb/s (100GBASE – SR10, 850 nm, 10 x 10GbE; 100GBASE – LR4, 1300 nm, 4 x 25GbE, DWDM)

• 16G FC standard for Fiber channel

• Standard and MSA have focused the technology development:

- 16G FC transceiver SFP+ (Avago 980, Optnext 1300)

- First 40Gb/s and 100Gb/s CFP MSA (Multi-Source Agreement) Optnext, Finisar

• Pushing the DFB technology up to 25Gb/s uncooled !


22 June, 2009

DFB technology: beyond the limits….

• OFC session: Uncooled and semicooled DFB laser sources at 25 and 40 Gb/s

- Key players: Avago, Finisar, Optnext/Hitachi, Fujitsu, NTT

• Avago: best eye quality up to 70C, 25Gb/s


22 June, 2009

… but manufacturability

is key!

DF019DF013DF011DF007

12

10

8

6

4

2

0

Wafer

Threshold_Front

Boxplot of Threshold_Front

BoxPlot of Threshold / Max. Power / SMSR for 4 wafers• Several hundredths chip/wafer automatically tested

• Tight statistics

• Good reproducibility wafer to wafer

�� Uniform / high yield technological processDF019DF013DF011DF007

35

30

25

20

15

10

5

0

Wafer

Power__Imax_Front

Boxplot of Power__Imax_Front

DF019DF013DF011DF007

50

45

40

35

30

Wafer

SMSR__100mA

Boxplot of SMSR__100mA

Accelerated Life test

• ALT conditions: 95 ºC, 80 mA ACC

• Burn In conditions: 100 ºC, 90 mA, 24 h

•• Devices stable over > 5000Devices stable over > 5000

�� reliable technological process

Delta Output Power (%)

-40

-30

-20

-10

0

10

20

30

40

0 1000 2000 3000 4000 5000 6000

aging time T (h)

Delta Output Power (%)


22 June, 2009

Photonic integration trends

• On telecom network:

- Low cost, 100Gb/s, modulation format…

• ⇒⇒⇒⇒ Photonics on InP

- Infinera, (OFC2009, OThN2) Transmitter PIC for 10-Channel x 40Gb/s per Channel Polarization-Multiplexed RZ-DQPSK Modulation

- High complexity; system on a chip….

-


22 June, 2009

Passive and tuning

waveguides

DBR and SOA active

waveguide

Wide tunable laser: DBR Array

InP Monolithically

integrated

DBR Array + PICBent

waveguide

sMMI

SOA

4 grating pitches (EBL)

⇒⇒⇒⇒ 10 nm spaced

Bragg Wavelengths

Bent

output

Output beam

400 um

1670 um

40 um

Medium

tuning

DBR

array

Paoletti et al. ECOC 2003Paoletti et al. ECOC 2003


22 June, 2009

Device results: emitted spectrum

1.52 1.53 1.54 1.55 1.56 1.57 1.58 1.59 1.6-70

-60

-50

-40

-30

-20

-10

0

chip4-M35-2 @ 20°C: Iact

=50mA

Wavelength - nm -

Rel. Optical Power -dB -

41 nm tuning range

Paoletti et al. ECOC 2003Paoletti et al. ECOC 2003


22 June, 2009

Revolution on active material:Is Quantum Dot ready to go?

OFC2009, OWJ1, OFC2009, OWJ1, ““HighHigh--Speed and TemperatureSpeed and Temperature--Insensitive Operation in 1.3Insensitive Operation in 1.3--µµm m InAs/GaAsInAs/GaAsHighHigh--Density Quantum Dot LasersDensity Quantum Dot Lasers”” , Fujitsu, Fujitsu

200 um long 1.3um ridge FP laser, with amazing performances… on

GaAs . Announced to be “ready for production”….


22 June, 2009

The Turin Technology

Centre (TTC)

Via G. Schiaparelli 12 10148 Torino ItalyVia G. Schiaparelli 12 10148 Torino Italy

Acquisition by Agilent Technologies 19 April 2000

Activity: R&D and Production (laser chip)

• Short term Development projects (transceivers @ 2.5 Gbit/s and @10 Gbit/s)

• Medium term Research projects for active and passive devices

• Development and Production of 10G FP/DFB/EML laser source

Facilities

• EPI (2 MOCVD), material characterization, processing (including EBL), die fab (singulation, coating, testing, assembly and reliability tests)

Expertise: optoelectronic and photonic technologies

• New devices and components conception and design

• Semiconductors

• Device design, prototyping and characterisation

Offices surface: 1005 m2

Laboratory surface: 3281 m2


22 June, 2009

Key StrengthDeveloping a reliable technology….

• Reliability has always been the key strength in a III-V world

• Customer reliability expectation is almost compared to the ‘telecom”field, but for low cost – consumer products

⇒⇒⇒⇒ Reliability is the key investment in the III-V area

Example: DFB (SFP+, XFP)

qualified for Cisco:

• 7 M device* hours

• Long endurance test

Enormous investments…


22 June, 2009

Key StrengthTesting a production volume….

• Testing is one of the most expensive part in the III-V production

• Typical application require 100% testing of the laser chip!

• Only key team have testing capability suitable for mass-production

- Investment not scalable for start-up

⇒⇒⇒⇒ (low cost) testing developing is a key strength.. Also from R&D design perspective

SMSR 30 mA (dB)

0.00

4.00

8.00

12.00

16.00

20.00

24.00

28.00

32.00

36.00

40.00

44.00

48.00

DA012__Y2 (346)

DA012__Y4 (264)

DA012__Y8 (336)

DA013__Y1 (103)

DA013__Y3 (246)

DA017__Y2 (357)

DA017__Y4 (316)

DA017__Y6 (247)

DA022__Y2 (249)

DA022__Y5 (98)

DA022__Y7 (377)

DA043__Y2 (490)

DA043__Y4 (519)

DA044__Y1 (417)

DA044__Y3 (761)

DA046__Y2 (387)

DA046__Y4 (427)

DA046__Y6 (628)

DA047__Y2 (110)

DA047__Y4 (570)

DA054Z_Y1 (204)

DA054Z_Y3 (321)

DA055U_Y2 (289)

med 25% 75%

Example: DFB qualificationExample: DFB qualification•• Statistic of 9 qual wafers:Statistic of 9 qual wafers:•• Based on 15K chip tested across 8 Based on 15K chip tested across 8 waferswafers


22 June, 2009

www avagotech.com

www.torinoscienza.it/lab-vr/agilent

(visita virtuale ai laboratori TTC)

[email protected]

[email protected]

San Jose

-9h

Torino

Singapore

+6h

Chip development requires team work

across the globe: not an easy task!


22 June, 2009

The end!• Books

- G. Guekos, Photonic Devices, Springer, 1999, ISBN 3-540-64318-4

- L. A. Coldren, S. W. Corzine, “Diode lasers and photonic integrated circuits”, John Wiley and sons, inc.,

- P. Vasil’ev, “Ultrafast diode laser”, Artec House Boston-London

- K. Petermann, “Laser Diode Modulation” and Noise, Dordrecht, The Netherlands: Kluwer Academic Publishers

• Related published paper

- F. Delpiano, R. Paoletti, P. Audagnotto and R. Puleo, "High Frequency Modelling and Characterisation of High Performance DFB Laser Modules", IEEE Transaction on Components, Hybrids, and Manufacturing Technology, Part B, Vol. 17, No 3, pp. 412-417, august 1994.

- R. Paoletti, D. Bertone, A. Bricconi, R. Fang, L. Greborio, G. Magnetti, M. Meliga, "Comparison of Optical and Electrical Modulation Bandwidths in three different 1.55 µµµµm InGaAsP Buried Laser Structures", SPIE'S International Symposia - Photonics West '96, pp. 296-305, 30 Jan. - 1 Febr. 1996, S. Josè, CA, USA.

- R. Paoletti, M. Meliga, I. Montrosset, “Optical Modulation Technique for Carrier Lifetime Measurement in Semiconductor Lasers”, IEEE Photonics Technology Letters, Vol. 8, No. 11, pp. 1447-1449, November 1996.

- R. Paoletti, M. Meliga, G. Oliveti, M. Puleo, G. Rossi, L. Senepa, “10 Gbit/S Ultra-Low Chirp 1.55µµµµM Directly Modulated Hybrid Fiber Grating - Semiconductor Laser Source”, 23rd European Conference on Optical Communication ECOC '97, Mo 3B. 22-25 September 1997, Edimburgh (UK).

- R. Y. Fang, D. Bertone, M. Meliga, Montrosset*, S. Murgia, G. Morello, G. Magnetti, G. Oliveti, R. Paoletti, “A simple structure 1.55 µµµµm InGaAsP/InP Spot Size Converted (SSC) laser”, IEEE Photonics Technology Letters, Vol. 10, No. 6, pp. 775-777, June 1998.

- G. Rossi, R. Paoletti, M. Meliga, “SPICE simulation for analysis and design of fast 1.55 µµµµm MQW laser diodes”, IEEE Journal of Lightwave Technology, Vol. 16, No. 7, July 1998.

- R. Paoletti, M. Agresti, G. Burns, G. Berry, D. Bertone. P. Charles, P. Crump, A. Davies, R.Y. Fang, R. Ghin, P. Gotta, M. Holm, C. Kompocholis, G. Magnetti, J. Massa, G. Meneghini, G. Rossi, P. Ryder, A. Taylor, P. Valenti and M. Meliga, "100 °°°°C, 10 Gb/s directly modulated InGaAsP DFB lasers for uncooled Ethernet applications", post-deadline at European Conference on Optical Communication ECOC '2001, October 2001,Amstedam (NL).

- R. Paoletti, M. Meliga, "Uncooled, high speed DFB lasers for Gigabit Ethernet applications", invited paper at SPIE'S International Symposia - Photonics West Optoelectronics 20021, 19 - 25 Jan. 2002, S. Josè, CA, USA.

- R. Paoletti, C. Coriasso, M. Agresti, P. Gotta, G. Magnetti, A. Moro, D. Sarocchi, D. Soderstrom, C. Cacciatore, L. Fratta, M. Vallone, A. Stano, E. Liotti, P. Valenti, G. Roggero, G. Fornuto, G. Burns, R.Harrell, P. Charles, D. Clark, G. Berry and M. Meliga, "Small chip size, low power consumption, fully electronic controlled tunable laser source with 40 nm tuning range and 20 mW output power for WDM applications", ECOC 2003, 21 - 25 Sept. 2003, Rimini, Italy

- R. Paoletti, M. Agresti, D. Bertone, L. Bianco, C. Bruschi, A. Buccieri, R. Campi, C.Dorigoni, P. Gotta, M. Liotti, G. Magnetti, P. Montangero, G. Morello, C. Rigo, E. Riva, D. Soderstrom, S. Stano, P. Valenti, M. Vallone, M. Meliga" Highly reliable and high yield 1300 nm InGaAlAs directly modulated ridge Fabry-Perot lasers, operating at 10 Gb/s, up to 110 ºC, with constant current swing ", Post deadline at Optical Fiber Conference OFC 2005, Anaheim (CA)

semiconductor lasers for high bit rate optical...

Documents